This project, originally inspired by Benchmarking SQLite Performance in Go, bechmarks various SQLite implementations:
- github.com/eatonphil/gosqlite (aka "eatonphil")
- github.com/glebarez/go-sqlite (aka "glebarez")
- github.com/mattn/go-sqlite3 (aka "mattn")
- github.com/ncruces/go-sqlite3 (aka "ncruces")
- github.com/tailscale/sqlite (aka "tailscale")
- github.com/zombiezen/go-sqlite (aka "zombiezen")
- gitlab.com/cznic/sqlite (aka "modernc")
Here are some quick descriptions:
-
eatonphil
eatonphil is a CGO-based implementation. It offers a direct interface.
-
glebarez
glebarez is a non-CGO transpilation-based implementation, based on modernc. It offers a
database/sqlinterface. -
mattn
mattn is a CGO-based implementation. It offers a
database/sqlinterface. -
modernc
modernc is a non-CGO transpilation-based implementation. It offers a
database/sqlinterface. -
ncruces
ncruces is a non-CGO WASM-based implementation. It offers both direct and
database/sqlinterfaces. -
tailscale
tailscale is a CGO-based implementation. It offers both direct and
database/sqlinterfaces. -
zombiezen
zombiezen is a non-CGO transpilation-based implementation, based on modernc. It offers a direct interface.
The benchmark results are below.
The benchmark results in the quick directory were generated using:
make benchstat-by-categoryIt takes ~20m to run the "quick" benchmarks on my laptop.
The benchmark results in the slow directory were generated using:
make benchstat-by-category BENCH_BIG=1 BENCH_SLOW=1It takes ~1h10m to run the "slow" benchmarks on my laptop.
The tests in the tests directory were generated using:
make test-allAmong other things, the tests capture the compile-time options, pragmas and SQLite version used by each implementation.
(You can diff one test file against another to get a sense of how and where the implementations... well, differ.)
There are lots of ways to play the benchmark game; take a look at the examples below.
$ make help
Targets:
bench-all - Run all benchmarks
bench-by-category - Run all benchmark categories
bench-category-baseline - Run baseline benchmarks
bench-category-baseline-parallel - Run baseline benchmarks in parallel
bench-category-populate - Run populate benchmarks
bench-category-readwrite - Run readwrite benchmarks
bench-category-readwrite-parallel - Run readwrite benchmarks in parallel
bench-category-query-correlated - Run correlated query benchmarks
bench-category-query-correlated-parallel - Run correlated query benchmarks in parallel
bench-category-query-groupby - Run groupby query benchmarks
bench-category-query-groupby-parallel - Run groupby query benchmarks in parallel
bench-category-query-json - Run json query benchmarks
bench-category-query-json-parallel - Run json query benchmarks in parallel
bench-category-query-orderby - Run orderby query benchmarks
bench-category-query-orderby-parallel - Run orderby query benchmarks in parallel
bench-category-query-recursivecte - Run recursivecte query benchmarks
bench-category-query-recursivecte-parallel - Run recursivecte query benchmarks in parallel
bench-category-query-window - Run window query benchmarks
bench-category-query-window-parallel - Run window query benchmarks in parallel
benchstat-all - Compare all benchmarks
benchstat-by-category - Run and compare all benchmark categories
benchstat-category-baseline - Run and compare baseline benchmarks
benchstat-category-baseline-parallel - Run and compare baseline benchmarks in parallel
benchstat-category-populate - Run and compare populate benchmarks
benchstat-category-readwrite - Run and compare readwrite benchmarks
benchstat-category-readwrite-parallel - Run and compare readwrite benchmarks in parallel
benchstat-category-query-correlated - Run and compare correlated query benchmarks
benchstat-category-query-correlated-parallel - Run and compare correlated query benchmarks in parallel
benchstat-category-query-groupby - Run and compare groupby query benchmarks
benchstat-category-query-groupby-parallel - Run and compare groupby query benchmarks in parallel
benchstat-category-query-json - Run and compare json query benchmarks
benchstat-category-query-json-parallel - Run and compare json query benchmarks in parallel
benchstat-category-query-orderby - Run and compare orderby query benchmarks
benchstat-category-query-orderby-parallel - Run and compare orderby query benchmarks in parallel
benchstat-category-query-recursivecte - Run and compare recursivecte query benchmarks
benchstat-category-query-recursivecte-parallel - Run and compare recursivecte query benchmarks in parallel
benchstat-category-query-window - Run and compare window query benchmarks
benchstat-category-query-window-parallel - Run and compare window query benchmarks in parallel
clean - Remove all benchmark, benchstat and test files
test-all - Run all tests
update-all - Update the quick/, slow/, and tests/ directories and README
update-quick - Update the quick/ directory
update-slow - Update the slow/ directory
update-tests - Update the tests/ directory
update-readme - Update the README
Variables:
TAGS="ncruces_direct ncruces_driver eatonphil_direct glebarez_driver mattn_driver modernc_driver tailscale_driver zombiezen_direct"
The first TAG listed becomes the baseline for benchstat comparisons.
BENCH_COUNT=1
BENCH_CPU=''
BENCH_CPU_PARALLEL=1,2,4,8,16
BENCH_OPTS="-benchmem -short"
BENCH_PATTERN=.
BENCH_SKIP=''
BENCH_TIME=1s
BENCH_TIMEOUT=15m
BENCH_SLOW=
BENCH_SLOW changes some of the values shown above. To see the BENCH_SLOW values,
try "make help BENCH_SLOW=1"
BENCH_MAX_READ_CONNECTIONS=0
BENCH_MAX_WRITE_CONNECTIONS=32
BENCH_POSTS=200
BENCH_POST_PARAGRAPHS=10
BENCH_COMMENTS=10
BENCH_COMMENT_PARAGRAPHS=1
BENCH_BIG=
BENCH_BIG changes some of the values shown above. To see the BENCH_BIG values,
try "make help BENCH_BIG=1"
TEST_COUNT=1
TEST_CPU=''
TEST_OPTS="-short -v"
TEST_PATTERN=.
TEST_SKIP=''
Examples:
make bench-all
make benchstat-all
make bench-all BENCH_COUNT=3
make benchstat-all
make bench-all BENCH_SLOW=1
make benchstat-all
make bench-all BENCH_BIG=1
make benchstat-all
make bench-all BENCH_SLOW=1 BENCH_BIG=1
make benchstat-all
make bench-all TAGS="ncruces_direct ncruces_driver"
make benchstat-all TAGS="ncruces_direct ncruces_driver"
make benchstat-by-category
make benchstat-by-category BENCH_COUNT=3
make benchstat-by-category BENCH_SLOW=1
make benchstat-by-category BENCH_BIG=1
make benchstat-by-category BENCH_SLOW=1 BENCH_BIG=1
make benchstat-by-category TAGS="ncruces_direct ncruces_driver"
make test-all
make test-all TAGS="ncruces_direct ncruces_driver"
make clean
-
Configuration
All implementations use the following configuration. Some implementations use additional configuration.
PRAGMA busy_timeout(5000); PRAGMA foreign_keys(true); PRAGMA journal_mode(wal); PRAGMA synchronous(normal); -
Sequential vs. Parallel
Benchmarks that don't end with "Parallel" are sequential benchmarks; they run on a single goroutine, in a loop:
for b.Loop() { ... }Benchmarks that end with "Parallel" are parallel benchmarks; they run on
GOMAXPROCSgoroutines, in a loop:b.RunParallel(func(pb *testing.PB) { for pb.Next() { ... } })"Sequential & Parallel Performance" runs have lines like these:
Baseline/Conn-12 Baseline/ConnParallel-12They show the same benchmark, "Conn", being run both sequentially and in parallel (with
GOMAXPROCSset to 12 in this case, the default value on my laptop)."Parallel Performance" runs have lines like these:
Baseline/ConnParallel Baseline/ConnParallel-2 Baseline/ConnParallel-4 Baseline/ConnParallel-8 Baseline/ConnParallel-16They show the same benchmark, "ConnParallel", being run in parallel (with
GOMAXPROCSset to 1, 2, 4, 8, and 16 in this case). -
Queries
The query benchmarks were really meant to push the non-CGO-based implementations. They do that — but they also revealed unexpected differences between the CGO-based implementations.
-
Sequential Performance
Sequential benchmarks answer the question: if I'm doing this one thing, how fast (or slow) will it be?
-
Parallel Performance
Parallel benchmarks answer the question: if I'm doing more than one of these things at the same time, how fast (or slow) will they be?
Depending on the nature of the task, the answer might be "faster than sequential;" "the same as sequential;" or "slower than sequential."
-
Memory Usage
How much memory does one implementation use compared to another? Is the difference big enough to matter in your environment?
(Just keep in mind that the benchmarks only measure memory usage that's visible to Go.)
-
Conn
Get one connection from the pool and then return it to the pool. (Every benchmark does that, but this benchmark does only that.)
-
Select1
Execute a "SELECT 1" query, preparing it each time.
-
Select1PrePrepared
Execute a "SELECT 1" query, preparing it an advance.
benchstat bench_ncruces_direct.txt bench_ncruces_driver.txt bench_eatonphil_direct.txt bench_glebarez_driver.txt bench_mattn_driver.txt bench_modernc_driver.txt bench_tailscale_driver.txt bench_zombiezen_direct.txt
goos: darwin
goarch: arm64
pkg: github.com/michaellenaghan/go-sqlite-bench
cpu: Apple M2 Pro
│ bench_ncruces_direct.txt │ bench_ncruces_driver.txt │ bench_eatonphil_direct.txt │ bench_glebarez_driver.txt │ bench_mattn_driver.txt │ bench_modernc_driver.txt │ bench_tailscale_driver.txt │ bench_zombiezen_direct.txt │
│ sec/op │ sec/op vs base │ sec/op vs base │ sec/op vs base │ sec/op vs base │ sec/op vs base │ sec/op vs base │ sec/op vs base │
Baseline/Conn-12 189.2n ± ∞ ¹ 224.1n ± ∞ ¹ ~ (p=0.100 n=3) ² 187.9n ± ∞ ¹ ~ (p=0.100 n=3) ² 225.4n ± ∞ ¹ ~ (p=0.100 n=3) ² 222.5n ± ∞ ¹ ~ (p=0.100 n=3) ² 226.0n ± ∞ ¹ ~ (p=0.100 n=3) ² 227.7n ± ∞ ¹ ~ (p=0.100 n=3) ² 1119.0n ± ∞ ¹ ~ (p=0.100 n=3) ²
Baseline/ConnParallel-12 233.1n ± ∞ ¹ 534.1n ± ∞ ¹ ~ (p=0.100 n=3) ² 223.0n ± ∞ ¹ ~ (p=0.100 n=3) ² 542.5n ± ∞ ¹ ~ (p=0.100 n=3) ² 538.0n ± ∞ ¹ ~ (p=0.100 n=3) ² 537.2n ± ∞ ¹ ~ (p=0.100 n=3) ² 541.0n ± ∞ ¹ ~ (p=0.100 n=3) ² 1314.0n ± ∞ ¹ ~ (p=0.100 n=3) ²
Baseline/Select1-12 1902.0n ± ∞ ¹ 1947.0n ± ∞ ¹ ~ (p=0.200 n=3) ² 924.1n ± ∞ ¹ ~ (p=0.100 n=3) ² 2001.0n ± ∞ ¹ ~ (p=0.100 n=3) ² 2635.0n ± ∞ ¹ ~ (p=0.100 n=3) ² 2024.0n ± ∞ ¹ ~ (p=0.100 n=3) ² 1118.0n ± ∞ ¹ ~ (p=0.100 n=3) ² 3110.0n ± ∞ ¹ ~ (p=0.100 n=3) ²
Baseline/Select1Parallel-12 1239.0n ± ∞ ¹ 1159.0n ± ∞ ¹ ~ (p=0.100 n=3) ² 751.2n ± ∞ ¹ ~ (p=0.100 n=3) ² 2292.0n ± ∞ ¹ ~ (p=0.100 n=3) ² 1513.0n ± ∞ ¹ ~ (p=0.100 n=3) ² 2305.0n ± ∞ ¹ ~ (p=0.100 n=3) ² 914.9n ± ∞ ¹ ~ (p=0.100 n=3) ² 2652.0n ± ∞ ¹ ~ (p=0.100 n=3) ²
Baseline/Select1PrePrepared-12 504.4n ± ∞ ¹ 467.3n ± ∞ ¹ ~ (p=0.100 n=3) ² 487.9n ± ∞ ¹ ~ (p=0.100 n=3) ² 1976.0n ± ∞ ¹ ~ (p=0.100 n=3) ² 1304.0n ± ∞ ¹ ~ (p=0.100 n=3) ² 2002.0n ± ∞ ¹ ~ (p=0.100 n=3) ² 409.9n ± ∞ ¹ ~ (p=0.100 n=3) ² 1507.0n ± ∞ ¹ ~ (p=0.100 n=3) ²
Baseline/Select1PrePreparedParallel-12 654.0n ± ∞ ¹ 897.4n ± ∞ ¹ ~ (p=0.100 n=3) ² 477.8n ± ∞ ¹ ~ (p=0.100 n=3) ² 1787.0n ± ∞ ¹ ~ (p=0.100 n=3) ² 1108.0n ± ∞ ¹ ~ (p=0.100 n=3) ² 1784.0n ± ∞ ¹ ~ (p=0.100 n=3) ² 774.5n ± ∞ ¹ ~ (p=0.100 n=3) ² 1533.0n ± ∞ ¹ ~ (p=0.100 n=3) ²
geomean 570.0n 695.6n +22.04% 435.1n -23.67% 1.121µ +96.61% 939.9n +64.91% 1.125µ +97.31% 584.8n +2.60% 1.743µ +205.76%
¹ need >= 6 samples for confidence interval at level 0.95
² need >= 4 samples to detect a difference at alpha level 0.05
│ bench_ncruces_direct.txt │ bench_ncruces_driver.txt │ bench_eatonphil_direct.txt │ bench_glebarez_driver.txt │ bench_mattn_driver.txt │ bench_modernc_driver.txt │ bench_tailscale_driver.txt │ bench_zombiezen_direct.txt │
│ B/op │ B/op vs base │ B/op vs base │ B/op vs base │ B/op vs base │ B/op vs base │ B/op vs base │ B/op vs base │
Baseline/Conn-12 0.00 ± ∞ ¹ 64.00 ± ∞ ¹ ~ (p=0.100 n=3) ² 0.00 ± ∞ ¹ ~ (p=1.000 n=3) ³ 64.00 ± ∞ ¹ ~ (p=0.100 n=3) ² 64.00 ± ∞ ¹ ~ (p=0.100 n=3) ² 64.00 ± ∞ ¹ ~ (p=0.100 n=3) ² 64.00 ± ∞ ¹ ~ (p=0.100 n=3) ² 353.00 ± ∞ ¹ ~ (p=0.100 n=3) ²
Baseline/ConnParallel-12 0.00 ± ∞ ¹ 64.00 ± ∞ ¹ ~ (p=0.100 n=3) ² 0.00 ± ∞ ¹ ~ (p=1.000 n=3) ³ 64.00 ± ∞ ¹ ~ (p=0.100 n=3) ² 64.00 ± ∞ ¹ ~ (p=0.100 n=3) ² 64.00 ± ∞ ¹ ~ (p=0.100 n=3) ² 64.00 ± ∞ ¹ ~ (p=0.100 n=3) ² 352.00 ± ∞ ¹ ~ (p=0.100 n=3) ²
Baseline/Select1-12 48.00 ± ∞ ¹ 32.00 ± ∞ ¹ ~ (p=0.100 n=3) ² 96.00 ± ∞ ¹ ~ (p=0.100 n=3) ² 252.00 ± ∞ ¹ ~ (p=0.100 n=3) ² 288.00 ± ∞ ¹ ~ (p=0.100 n=3) ² 252.00 ± ∞ ¹ ~ (p=0.100 n=3) ² 240.00 ± ∞ ¹ ~ (p=0.100 n=3) ² 705.00 ± ∞ ¹ ~ (p=0.100 n=3) ²
Baseline/Select1Parallel-12 48.00 ± ∞ ¹ 32.00 ± ∞ ¹ ~ (p=0.100 n=3) ² 96.00 ± ∞ ¹ ~ (p=0.100 n=3) ² 252.00 ± ∞ ¹ ~ (p=0.100 n=3) ² 288.00 ± ∞ ¹ ~ (p=0.100 n=3) ² 252.00 ± ∞ ¹ ~ (p=0.100 n=3) ² 240.00 ± ∞ ¹ ~ (p=0.100 n=3) ² 704.00 ± ∞ ¹ ~ (p=0.100 n=3) ²
Baseline/Select1PrePrepared-12 0.00 ± ∞ ¹ 48.00 ± ∞ ¹ ~ (p=0.100 n=3) ² 0.00 ± ∞ ¹ ~ (p=1.000 n=3) ³ 252.00 ± ∞ ¹ ~ (p=0.100 n=3) ² 240.00 ± ∞ ¹ ~ (p=0.100 n=3) ² 252.00 ± ∞ ¹ ~ (p=0.100 n=3) ² 48.00 ± ∞ ¹ ~ (p=0.100 n=3) ² 352.00 ± ∞ ¹ ~ (p=0.100 n=3) ²
Baseline/Select1PrePreparedParallel-12 0.00 ± ∞ ¹ 48.00 ± ∞ ¹ ~ (p=0.100 n=3) ² 0.00 ± ∞ ¹ ~ (p=1.000 n=3) ³ 252.00 ± ∞ ¹ ~ (p=0.100 n=3) ² 240.00 ± ∞ ¹ ~ (p=0.100 n=3) ² 252.00 ± ∞ ¹ ~ (p=0.100 n=3) ² 48.00 ± ∞ ¹ ~ (p=0.100 n=3) ² 353.00 ± ∞ ¹ ~ (p=0.100 n=3) ²
geomean ⁴ 46.15 ? +25.99% ⁴ 159.6 ? 164.2 ? 159.6 ? 90.34 ? 444.0 ?
¹ need >= 6 samples for confidence interval at level 0.95
² need >= 4 samples to detect a difference at alpha level 0.05
³ all samples are equal
⁴ summaries must be >0 to compute geomean
│ bench_ncruces_direct.txt │ bench_ncruces_driver.txt │ bench_eatonphil_direct.txt │ bench_glebarez_driver.txt │ bench_mattn_driver.txt │ bench_modernc_driver.txt │ bench_tailscale_driver.txt │ bench_zombiezen_direct.txt │
│ allocs/op │ allocs/op vs base │ allocs/op vs base │ allocs/op vs base │ allocs/op vs base │ allocs/op vs base │ allocs/op vs base │ allocs/op vs base │
Baseline/Conn-12 0.000 ± ∞ ¹ 1.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 0.000 ± ∞ ¹ ~ (p=1.000 n=3) ³ 1.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 1.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 1.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 1.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 5.000 ± ∞ ¹ ~ (p=0.100 n=3) ²
Baseline/ConnParallel-12 0.000 ± ∞ ¹ 1.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 0.000 ± ∞ ¹ ~ (p=1.000 n=3) ³ 1.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 1.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 1.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 1.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 5.000 ± ∞ ¹ ~ (p=0.100 n=3) ²
Baseline/Select1-12 1.000 ± ∞ ¹ 1.000 ± ∞ ¹ ~ (p=1.000 n=3) ³ 4.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 7.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 9.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 7.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 5.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 8.000 ± ∞ ¹ ~ (p=0.100 n=3) ²
Baseline/Select1Parallel-12 1.000 ± ∞ ¹ 1.000 ± ∞ ¹ ~ (p=1.000 n=3) ³ 4.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 7.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 9.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 7.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 5.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 8.000 ± ∞ ¹ ~ (p=0.100 n=3) ²
Baseline/Select1PrePrepared-12 0.000 ± ∞ ¹ 2.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 0.000 ± ∞ ¹ ~ (p=1.000 n=3) ³ 7.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 7.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 7.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 2.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 5.000 ± ∞ ¹ ~ (p=0.100 n=3) ²
Baseline/Select1PrePreparedParallel-12 0.000 ± ∞ ¹ 2.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 0.000 ± ∞ ¹ ~ (p=1.000 n=3) ³ 7.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 7.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 7.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 2.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 5.000 ± ∞ ¹ ~ (p=0.100 n=3) ²
geomean ⁴ 1.260 ? +58.74% ⁴ 3.659 ? 3.979 ? 3.659 ? 2.154 ? 5.848 ?
¹ need >= 6 samples for confidence interval at level 0.95
² need >= 4 samples to detect a difference at alpha level 0.05
³ all samples are equal
⁴ summaries must be >0 to compute geomean
benchstat bench_ncruces_direct.txt bench_ncruces_driver.txt bench_eatonphil_direct.txt bench_glebarez_driver.txt bench_mattn_driver.txt bench_modernc_driver.txt bench_tailscale_driver.txt bench_zombiezen_direct.txt
goos: darwin
goarch: arm64
pkg: github.com/michaellenaghan/go-sqlite-bench
cpu: Apple M2 Pro
│ bench_ncruces_direct.txt │ bench_ncruces_driver.txt │ bench_eatonphil_direct.txt │ bench_glebarez_driver.txt │ bench_mattn_driver.txt │ bench_modernc_driver.txt │ bench_tailscale_driver.txt │ bench_zombiezen_direct.txt │
│ sec/op │ sec/op vs base │ sec/op vs base │ sec/op vs base │ sec/op vs base │ sec/op vs base │ sec/op vs base │ sec/op vs base │
Baseline/ConnParallel 186.4n ± ∞ ¹ 250.7n ± ∞ ¹ ~ (p=0.100 n=3) ² 188.1n ± ∞ ¹ ~ (p=0.200 n=3) ² 245.0n ± ∞ ¹ ~ (p=0.100 n=3) ² 233.7n ± ∞ ¹ ~ (p=0.100 n=3) ² 243.9n ± ∞ ¹ ~ (p=0.100 n=3) ² 233.6n ± ∞ ¹ ~ (p=0.100 n=3) ² 1351.0n ± ∞ ¹ ~ (p=0.100 n=3) ²
Baseline/ConnParallel-2 160.4n ± ∞ ¹ 216.7n ± ∞ ¹ ~ (p=0.100 n=3) ² 155.8n ± ∞ ¹ ~ (p=0.100 n=3) ² 221.7n ± ∞ ¹ ~ (p=0.100 n=3) ² 218.0n ± ∞ ¹ ~ (p=0.100 n=3) ² 224.6n ± ∞ ¹ ~ (p=0.100 n=3) ² 218.9n ± ∞ ¹ ~ (p=0.100 n=3) ² 693.6n ± ∞ ¹ ~ (p=0.100 n=3) ²
Baseline/ConnParallel-4 217.1n ± ∞ ¹ 266.0n ± ∞ ¹ ~ (p=0.100 n=3) ² 205.0n ± ∞ ¹ ~ (p=0.100 n=3) ² 269.3n ± ∞ ¹ ~ (p=0.100 n=3) ² 271.7n ± ∞ ¹ ~ (p=0.100 n=3) ² 265.2n ± ∞ ¹ ~ (p=0.100 n=3) ² 268.1n ± ∞ ¹ ~ (p=0.100 n=3) ² 685.7n ± ∞ ¹ ~ (p=0.100 n=3) ²
Baseline/ConnParallel-8 229.7n ± ∞ ¹ 509.3n ± ∞ ¹ ~ (p=0.100 n=3) ² 216.6n ± ∞ ¹ ~ (p=0.100 n=3) ² 509.9n ± ∞ ¹ ~ (p=0.100 n=3) ² 514.7n ± ∞ ¹ ~ (p=0.100 n=3) ² 518.9n ± ∞ ¹ ~ (p=0.100 n=3) ² 516.0n ± ∞ ¹ ~ (p=0.100 n=3) ² 1247.0n ± ∞ ¹ ~ (p=0.100 n=3) ²
Baseline/ConnParallel-16 227.0n ± ∞ ¹ 542.7n ± ∞ ¹ ~ (p=0.100 n=3) ² 219.6n ± ∞ ¹ ~ (p=0.100 n=3) ² 538.8n ± ∞ ¹ ~ (p=0.100 n=3) ² 547.6n ± ∞ ¹ ~ (p=0.100 n=3) ² 551.0n ± ∞ ¹ ~ (p=0.100 n=3) ² 547.5n ± ∞ ¹ ~ (p=0.100 n=3) ² 1378.0n ± ∞ ¹ ~ (p=0.100 n=3) ²
Baseline/Select1Parallel 1933.0n ± ∞ ¹ 1927.0n ± ∞ ¹ ~ (p=1.000 n=3) ² 935.6n ± ∞ ¹ ~ (p=0.100 n=3) ² 1898.0n ± ∞ ¹ ~ (p=0.100 n=3) ² 1647.0n ± ∞ ¹ ~ (p=0.100 n=3) ² 1902.0n ± ∞ ¹ ~ (p=0.100 n=3) ² 1156.0n ± ∞ ¹ ~ (p=0.100 n=3) ² 2858.0n ± ∞ ¹ ~ (p=0.100 n=3) ²
Baseline/Select1Parallel-2 1057.0n ± ∞ ¹ 1019.0n ± ∞ ¹ ~ (p=0.200 n=3) ² 550.9n ± ∞ ¹ ~ (p=0.100 n=3) ² 1098.0n ± ∞ ¹ ~ (p=0.100 n=3) ² 1005.0n ± ∞ ¹ ~ (p=0.400 n=3) ² 1085.0n ± ∞ ¹ ~ (p=0.100 n=3) ² 649.8n ± ∞ ¹ ~ (p=0.100 n=3) ² 1529.0n ± ∞ ¹ ~ (p=0.100 n=3) ²
Baseline/Select1Parallel-4 622.6n ± ∞ ¹ 563.6n ± ∞ ¹ ~ (p=0.200 n=3) ² 349.0n ± ∞ ¹ ~ (p=0.100 n=3) ² 1004.0n ± ∞ ¹ ~ (p=0.100 n=3) ² 631.5n ± ∞ ¹ ~ (p=0.700 n=3) ² 983.3n ± ∞ ¹ ~ (p=0.100 n=3) ² 399.0n ± ∞ ¹ ~ (p=0.100 n=3) ² 1110.0n ± ∞ ¹ ~ (p=0.100 n=3) ²
Baseline/Select1Parallel-8 1340.0n ± ∞ ¹ 970.1n ± ∞ ¹ ~ (p=0.100 n=3) ² 760.4n ± ∞ ¹ ~ (p=0.100 n=3) ² 2214.0n ± ∞ ¹ ~ (p=0.100 n=3) ² 1473.0n ± ∞ ¹ ~ (p=0.100 n=3) ² 2234.0n ± ∞ ¹ ~ (p=0.100 n=3) ² 879.1n ± ∞ ¹ ~ (p=0.100 n=3) ² 2481.0n ± ∞ ¹ ~ (p=0.100 n=3) ²
Baseline/Select1Parallel-16 1369.0n ± ∞ ¹ 1024.0n ± ∞ ¹ ~ (p=0.100 n=3) ² 766.5n ± ∞ ¹ ~ (p=0.100 n=3) ² 2269.0n ± ∞ ¹ ~ (p=0.100 n=3) ² 1493.0n ± ∞ ¹ ~ (p=0.100 n=3) ² 2295.0n ± ∞ ¹ ~ (p=0.100 n=3) ² 926.8n ± ∞ ¹ ~ (p=0.100 n=3) ² 2703.0n ± ∞ ¹ ~ (p=0.100 n=3) ²
Baseline/Select1PrePreparedParallel 498.8n ± ∞ ¹ 468.5n ± ∞ ¹ ~ (p=0.100 n=3) ² 488.1n ± ∞ ¹ ~ (p=0.100 n=3) ² 1868.0n ± ∞ ¹ ~ (p=0.100 n=3) ² 858.3n ± ∞ ¹ ~ (p=0.100 n=3) ² 1885.0n ± ∞ ¹ ~ (p=0.100 n=3) ² 429.8n ± ∞ ¹ ~ (p=0.100 n=3) ² 1616.0n ± ∞ ¹ ~ (p=0.100 n=3) ²
Baseline/Select1PrePreparedParallel-2 326.0n ± ∞ ¹ 425.9n ± ∞ ¹ ~ (p=0.100 n=3) ² 339.5n ± ∞ ¹ ~ (p=0.100 n=3) ² 1053.0n ± ∞ ¹ ~ (p=0.100 n=3) ² 634.9n ± ∞ ¹ ~ (p=0.100 n=3) ² 1043.0n ± ∞ ¹ ~ (p=0.100 n=3) ² 354.4n ± ∞ ¹ ~ (p=0.100 n=3) ² 853.2n ± ∞ ¹ ~ (p=0.100 n=3) ²
Baseline/Select1PrePreparedParallel-4 440.8n ± ∞ ¹ 304.7n ± ∞ ¹ ~ (p=0.100 n=3) ² 313.0n ± ∞ ¹ ~ (p=0.100 n=3) ² 759.3n ± ∞ ¹ ~ (p=0.100 n=3) ² 524.7n ± ∞ ¹ ~ (p=0.400 n=3) ² 774.9n ± ∞ ¹ ~ (p=0.100 n=3) ² 288.0n ± ∞ ¹ ~ (p=0.100 n=3) ² 719.7n ± ∞ ¹ ~ (p=0.100 n=3) ²
Baseline/Select1PrePreparedParallel-8 660.7n ± ∞ ¹ 817.7n ± ∞ ¹ ~ (p=0.100 n=3) ² 488.8n ± ∞ ¹ ~ (p=0.100 n=3) ² 1741.0n ± ∞ ¹ ~ (p=0.100 n=3) ² 1049.0n ± ∞ ¹ ~ (p=0.100 n=3) ² 1742.0n ± ∞ ¹ ~ (p=0.100 n=3) ² 738.6n ± ∞ ¹ ~ (p=0.100 n=3) ² 1467.0n ± ∞ ¹ ~ (p=0.100 n=3) ²
Baseline/Select1PrePreparedParallel-16 665.4n ± ∞ ¹ 853.1n ± ∞ ¹ ~ (p=0.100 n=3) ² 491.8n ± ∞ ¹ ~ (p=0.100 n=3) ² 1779.0n ± ∞ ¹ ~ (p=0.100 n=3) ² 1097.0n ± ∞ ¹ ~ (p=0.100 n=3) ² 1784.0n ± ∞ ¹ ~ (p=0.100 n=3) ² 778.3n ± ∞ ¹ ~ (p=0.100 n=3) ² 1595.0n ± ∞ ¹ ~ (p=0.100 n=3) ²
geomean 493.2n 564.1n +14.36% 372.8n -24.42% 896.9n +81.83% 678.2n +37.49% 899.6n +82.39% 491.9n -0.28% 1.343µ +172.28%
¹ need >= 6 samples for confidence interval at level 0.95
² need >= 4 samples to detect a difference at alpha level 0.05
│ bench_ncruces_direct.txt │ bench_ncruces_driver.txt │ bench_eatonphil_direct.txt │ bench_glebarez_driver.txt │ bench_mattn_driver.txt │ bench_modernc_driver.txt │ bench_tailscale_driver.txt │ bench_zombiezen_direct.txt │
│ B/op │ B/op vs base │ B/op vs base │ B/op vs base │ B/op vs base │ B/op vs base │ B/op vs base │ B/op vs base │
Baseline/ConnParallel 0.00 ± ∞ ¹ 64.00 ± ∞ ¹ ~ (p=0.100 n=3) ² 0.00 ± ∞ ¹ ~ (p=1.000 n=3) ³ 64.00 ± ∞ ¹ ~ (p=0.100 n=3) ² 64.00 ± ∞ ¹ ~ (p=0.100 n=3) ² 64.00 ± ∞ ¹ ~ (p=0.100 n=3) ² 64.00 ± ∞ ¹ ~ (p=0.100 n=3) ² 381.00 ± ∞ ¹ ~ (p=0.100 n=3) ²
Baseline/ConnParallel-2 0.00 ± ∞ ¹ 64.00 ± ∞ ¹ ~ (p=0.100 n=3) ² 0.00 ± ∞ ¹ ~ (p=1.000 n=3) ³ 64.00 ± ∞ ¹ ~ (p=0.100 n=3) ² 64.00 ± ∞ ¹ ~ (p=0.100 n=3) ² 64.00 ± ∞ ¹ ~ (p=0.100 n=3) ² 64.00 ± ∞ ¹ ~ (p=0.100 n=3) ² 352.00 ± ∞ ¹ ~ (p=0.100 n=3) ²
Baseline/ConnParallel-4 0.00 ± ∞ ¹ 64.00 ± ∞ ¹ ~ (p=0.100 n=3) ² 0.00 ± ∞ ¹ ~ (p=1.000 n=3) ³ 64.00 ± ∞ ¹ ~ (p=0.100 n=3) ² 64.00 ± ∞ ¹ ~ (p=0.100 n=3) ² 64.00 ± ∞ ¹ ~ (p=0.100 n=3) ² 64.00 ± ∞ ¹ ~ (p=0.100 n=3) ² 352.00 ± ∞ ¹ ~ (p=0.100 n=3) ²
Baseline/ConnParallel-8 0.00 ± ∞ ¹ 64.00 ± ∞ ¹ ~ (p=0.100 n=3) ² 0.00 ± ∞ ¹ ~ (p=1.000 n=3) ³ 64.00 ± ∞ ¹ ~ (p=0.100 n=3) ² 64.00 ± ∞ ¹ ~ (p=0.100 n=3) ² 64.00 ± ∞ ¹ ~ (p=0.100 n=3) ² 64.00 ± ∞ ¹ ~ (p=0.100 n=3) ² 352.00 ± ∞ ¹ ~ (p=0.100 n=3) ²
Baseline/ConnParallel-16 0.00 ± ∞ ¹ 64.00 ± ∞ ¹ ~ (p=0.100 n=3) ² 0.00 ± ∞ ¹ ~ (p=1.000 n=3) ³ 64.00 ± ∞ ¹ ~ (p=0.100 n=3) ² 64.00 ± ∞ ¹ ~ (p=0.100 n=3) ² 64.00 ± ∞ ¹ ~ (p=0.100 n=3) ² 64.00 ± ∞ ¹ ~ (p=0.100 n=3) ² 352.00 ± ∞ ¹ ~ (p=0.100 n=3) ²
Baseline/Select1Parallel 48.00 ± ∞ ¹ 32.00 ± ∞ ¹ ~ (p=0.100 n=3) ² 96.00 ± ∞ ¹ ~ (p=0.100 n=3) ² 252.00 ± ∞ ¹ ~ (p=0.100 n=3) ² 288.00 ± ∞ ¹ ~ (p=0.100 n=3) ² 252.00 ± ∞ ¹ ~ (p=0.100 n=3) ² 240.00 ± ∞ ¹ ~ (p=0.100 n=3) ² 727.00 ± ∞ ¹ ~ (p=0.100 n=3) ²
Baseline/Select1Parallel-2 48.00 ± ∞ ¹ 32.00 ± ∞ ¹ ~ (p=0.100 n=3) ² 96.00 ± ∞ ¹ ~ (p=0.100 n=3) ² 252.00 ± ∞ ¹ ~ (p=0.100 n=3) ² 288.00 ± ∞ ¹ ~ (p=0.100 n=3) ² 252.00 ± ∞ ¹ ~ (p=0.100 n=3) ² 240.00 ± ∞ ¹ ~ (p=0.100 n=3) ² 704.00 ± ∞ ¹ ~ (p=0.100 n=3) ²
Baseline/Select1Parallel-4 48.00 ± ∞ ¹ 32.00 ± ∞ ¹ ~ (p=0.100 n=3) ² 96.00 ± ∞ ¹ ~ (p=0.100 n=3) ² 252.00 ± ∞ ¹ ~ (p=0.100 n=3) ² 288.00 ± ∞ ¹ ~ (p=0.100 n=3) ² 252.00 ± ∞ ¹ ~ (p=0.100 n=3) ² 240.00 ± ∞ ¹ ~ (p=0.100 n=3) ² 704.00 ± ∞ ¹ ~ (p=0.100 n=3) ²
Baseline/Select1Parallel-8 48.00 ± ∞ ¹ 32.00 ± ∞ ¹ ~ (p=0.100 n=3) ² 96.00 ± ∞ ¹ ~ (p=0.100 n=3) ² 252.00 ± ∞ ¹ ~ (p=0.100 n=3) ² 288.00 ± ∞ ¹ ~ (p=0.100 n=3) ² 252.00 ± ∞ ¹ ~ (p=0.100 n=3) ² 240.00 ± ∞ ¹ ~ (p=0.100 n=3) ² 704.00 ± ∞ ¹ ~ (p=0.100 n=3) ²
Baseline/Select1Parallel-16 48.00 ± ∞ ¹ 32.00 ± ∞ ¹ ~ (p=0.100 n=3) ² 96.00 ± ∞ ¹ ~ (p=0.100 n=3) ² 252.00 ± ∞ ¹ ~ (p=0.100 n=3) ² 288.00 ± ∞ ¹ ~ (p=0.100 n=3) ² 252.00 ± ∞ ¹ ~ (p=0.100 n=3) ² 240.00 ± ∞ ¹ ~ (p=0.100 n=3) ² 704.00 ± ∞ ¹ ~ (p=0.100 n=3) ²
Baseline/Select1PrePreparedParallel 0.00 ± ∞ ¹ 48.00 ± ∞ ¹ ~ (p=0.100 n=3) ² 0.00 ± ∞ ¹ ~ (p=1.000 n=3) ³ 252.00 ± ∞ ¹ ~ (p=0.100 n=3) ² 240.00 ± ∞ ¹ ~ (p=0.100 n=3) ² 252.00 ± ∞ ¹ ~ (p=0.100 n=3) ² 48.00 ± ∞ ¹ ~ (p=0.100 n=3) ² 373.00 ± ∞ ¹ ~ (p=0.100 n=3) ²
Baseline/Select1PrePreparedParallel-2 0.00 ± ∞ ¹ 48.00 ± ∞ ¹ ~ (p=0.100 n=3) ² 0.00 ± ∞ ¹ ~ (p=1.000 n=3) ³ 252.00 ± ∞ ¹ ~ (p=0.100 n=3) ² 240.00 ± ∞ ¹ ~ (p=0.100 n=3) ² 252.00 ± ∞ ¹ ~ (p=0.100 n=3) ² 48.00 ± ∞ ¹ ~ (p=0.100 n=3) ² 352.00 ± ∞ ¹ ~ (p=0.100 n=3) ²
Baseline/Select1PrePreparedParallel-4 0.00 ± ∞ ¹ 48.00 ± ∞ ¹ ~ (p=0.100 n=3) ² 0.00 ± ∞ ¹ ~ (p=1.000 n=3) ³ 252.00 ± ∞ ¹ ~ (p=0.100 n=3) ² 240.00 ± ∞ ¹ ~ (p=0.100 n=3) ² 252.00 ± ∞ ¹ ~ (p=0.100 n=3) ² 48.00 ± ∞ ¹ ~ (p=0.100 n=3) ² 352.00 ± ∞ ¹ ~ (p=0.100 n=3) ²
Baseline/Select1PrePreparedParallel-8 0.00 ± ∞ ¹ 48.00 ± ∞ ¹ ~ (p=0.100 n=3) ² 0.00 ± ∞ ¹ ~ (p=1.000 n=3) ³ 252.00 ± ∞ ¹ ~ (p=0.100 n=3) ² 240.00 ± ∞ ¹ ~ (p=0.100 n=3) ² 252.00 ± ∞ ¹ ~ (p=0.100 n=3) ² 48.00 ± ∞ ¹ ~ (p=0.100 n=3) ² 352.00 ± ∞ ¹ ~ (p=0.100 n=3) ²
Baseline/Select1PrePreparedParallel-16 0.00 ± ∞ ¹ 48.00 ± ∞ ¹ ~ (p=0.100 n=3) ² 0.00 ± ∞ ¹ ~ (p=1.000 n=3) ³ 252.00 ± ∞ ¹ ~ (p=0.100 n=3) ² 240.00 ± ∞ ¹ ~ (p=0.100 n=3) ² 252.00 ± ∞ ¹ ~ (p=0.100 n=3) ² 48.00 ± ∞ ¹ ~ (p=0.100 n=3) ² 352.00 ± ∞ ¹ ~ (p=0.100 n=3) ²
geomean ⁴ 46.15 ? +25.99% ⁴ 159.6 ? 164.2 ? 159.6 ? 90.34 ? 448.5 ?
¹ need >= 6 samples for confidence interval at level 0.95
² need >= 4 samples to detect a difference at alpha level 0.05
³ all samples are equal
⁴ summaries must be >0 to compute geomean
│ bench_ncruces_direct.txt │ bench_ncruces_driver.txt │ bench_eatonphil_direct.txt │ bench_glebarez_driver.txt │ bench_mattn_driver.txt │ bench_modernc_driver.txt │ bench_tailscale_driver.txt │ bench_zombiezen_direct.txt │
│ allocs/op │ allocs/op vs base │ allocs/op vs base │ allocs/op vs base │ allocs/op vs base │ allocs/op vs base │ allocs/op vs base │ allocs/op vs base │
Baseline/ConnParallel 0.000 ± ∞ ¹ 1.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 0.000 ± ∞ ¹ ~ (p=1.000 n=3) ³ 1.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 1.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 1.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 1.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 5.000 ± ∞ ¹ ~ (p=0.100 n=3) ²
Baseline/ConnParallel-2 0.000 ± ∞ ¹ 1.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 0.000 ± ∞ ¹ ~ (p=1.000 n=3) ³ 1.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 1.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 1.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 1.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 5.000 ± ∞ ¹ ~ (p=0.100 n=3) ²
Baseline/ConnParallel-4 0.000 ± ∞ ¹ 1.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 0.000 ± ∞ ¹ ~ (p=1.000 n=3) ³ 1.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 1.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 1.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 1.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 5.000 ± ∞ ¹ ~ (p=0.100 n=3) ²
Baseline/ConnParallel-8 0.000 ± ∞ ¹ 1.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 0.000 ± ∞ ¹ ~ (p=1.000 n=3) ³ 1.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 1.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 1.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 1.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 5.000 ± ∞ ¹ ~ (p=0.100 n=3) ²
Baseline/ConnParallel-16 0.000 ± ∞ ¹ 1.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 0.000 ± ∞ ¹ ~ (p=1.000 n=3) ³ 1.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 1.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 1.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 1.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 5.000 ± ∞ ¹ ~ (p=0.100 n=3) ²
Baseline/Select1Parallel 1.000 ± ∞ ¹ 1.000 ± ∞ ¹ ~ (p=1.000 n=3) ³ 4.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 7.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 9.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 7.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 5.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 8.000 ± ∞ ¹ ~ (p=0.100 n=3) ²
Baseline/Select1Parallel-2 1.000 ± ∞ ¹ 1.000 ± ∞ ¹ ~ (p=1.000 n=3) ³ 4.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 7.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 9.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 7.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 5.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 8.000 ± ∞ ¹ ~ (p=0.100 n=3) ²
Baseline/Select1Parallel-4 1.000 ± ∞ ¹ 1.000 ± ∞ ¹ ~ (p=1.000 n=3) ³ 4.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 7.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 9.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 7.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 5.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 8.000 ± ∞ ¹ ~ (p=0.100 n=3) ²
Baseline/Select1Parallel-8 1.000 ± ∞ ¹ 1.000 ± ∞ ¹ ~ (p=1.000 n=3) ³ 4.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 7.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 9.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 7.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 5.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 8.000 ± ∞ ¹ ~ (p=0.100 n=3) ²
Baseline/Select1Parallel-16 1.000 ± ∞ ¹ 1.000 ± ∞ ¹ ~ (p=1.000 n=3) ³ 4.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 7.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 9.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 7.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 5.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 8.000 ± ∞ ¹ ~ (p=0.100 n=3) ²
Baseline/Select1PrePreparedParallel 0.000 ± ∞ ¹ 2.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 0.000 ± ∞ ¹ ~ (p=1.000 n=3) ³ 7.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 7.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 7.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 2.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 5.000 ± ∞ ¹ ~ (p=0.100 n=3) ²
Baseline/Select1PrePreparedParallel-2 0.000 ± ∞ ¹ 2.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 0.000 ± ∞ ¹ ~ (p=1.000 n=3) ³ 7.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 7.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 7.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 2.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 5.000 ± ∞ ¹ ~ (p=0.100 n=3) ²
Baseline/Select1PrePreparedParallel-4 0.000 ± ∞ ¹ 2.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 0.000 ± ∞ ¹ ~ (p=1.000 n=3) ³ 7.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 7.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 7.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 2.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 5.000 ± ∞ ¹ ~ (p=0.100 n=3) ²
Baseline/Select1PrePreparedParallel-8 0.000 ± ∞ ¹ 2.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 0.000 ± ∞ ¹ ~ (p=1.000 n=3) ³ 7.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 7.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 7.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 2.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 5.000 ± ∞ ¹ ~ (p=0.100 n=3) ²
Baseline/Select1PrePreparedParallel-16 0.000 ± ∞ ¹ 2.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 0.000 ± ∞ ¹ ~ (p=1.000 n=3) ³ 7.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 7.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 7.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 2.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 5.000 ± ∞ ¹ ~ (p=0.100 n=3) ²
geomean ⁴ 1.260 ? +58.74% ⁴ 3.659 ? 3.979 ? 3.659 ? 2.154 ? 5.848 ?
¹ need >= 6 samples for confidence interval at level 0.95
² need >= 4 samples to detect a difference at alpha level 0.05
³ all samples are equal
⁴ summaries must be >0 to compute geomean
-
PopulateDB
Populate the database without using any explicit transactions.
-
PopulateDBWithTx
Populate the database using one (big) explicit write transaction.
-
PopulateDBWithTxs
Populate the database using one explicit write transaction per post + comments.
benchstat bench_ncruces_direct.txt bench_ncruces_driver.txt bench_eatonphil_direct.txt bench_glebarez_driver.txt bench_mattn_driver.txt bench_modernc_driver.txt bench_tailscale_driver.txt bench_zombiezen_direct.txt
goos: darwin
goarch: arm64
pkg: github.com/michaellenaghan/go-sqlite-bench
cpu: Apple M2 Pro
│ bench_ncruces_direct.txt │ bench_ncruces_driver.txt │ bench_eatonphil_direct.txt │ bench_glebarez_driver.txt │ bench_mattn_driver.txt │ bench_modernc_driver.txt │ bench_tailscale_driver.txt │ bench_zombiezen_direct.txt │
│ sec/op │ sec/op vs base │ sec/op vs base │ sec/op vs base │ sec/op vs base │ sec/op vs base │ sec/op vs base │ sec/op vs base │
Populate/PopulateDB-12 2.490 ± ∞ ¹ 2.524 ± ∞ ¹ ~ (p=0.700 n=3) ² 2.050 ± ∞ ¹ ~ (p=0.100 n=3) ² 2.736 ± ∞ ¹ ~ (p=0.100 n=3) ² 2.219 ± ∞ ¹ ~ (p=0.100 n=3) ² 2.751 ± ∞ ¹ ~ (p=0.100 n=3) ² 2.038 ± ∞ ¹ ~ (p=0.100 n=3) ² 2.388 ± ∞ ¹ ~ (p=0.100 n=3) ²
Populate/PopulateDBWithTx-12 1166.8m ± ∞ ¹ 1170.1m ± ∞ ¹ ~ (p=1.000 n=3) ² 913.3m ± ∞ ¹ ~ (p=0.100 n=3) ² 1478.1m ± ∞ ¹ ~ (p=0.100 n=3) ² 1129.6m ± ∞ ¹ ~ (p=1.000 n=3) ² 1409.9m ± ∞ ¹ ~ (p=0.100 n=3) ² 1024.0m ± ∞ ¹ ~ (p=0.100 n=3) ² 1130.9m ± ∞ ¹ ~ (p=0.400 n=3) ²
Populate/PopulateDBWithTxs-12 1.115 ± ∞ ¹ 1.381 ± ∞ ¹ ~ (p=0.100 n=3) ² 1.088 ± ∞ ¹ ~ (p=0.200 n=3) ² 1.398 ± ∞ ¹ ~ (p=0.100 n=3) ² 1.188 ± ∞ ¹ ~ (p=0.100 n=3) ² 1.391 ± ∞ ¹ ~ (p=0.100 n=3) ² 1.137 ± ∞ ¹ ~ (p=1.000 n=3) ² 1.194 ± ∞ ¹ ~ (p=0.100 n=3) ²
geomean 1.480 1.598 +7.99% 1.268 -14.32% 1.781 +20.41% 1.439 -2.75% 1.754 +18.56% 1.334 -9.86% 1.477 -0.15%
¹ need >= 6 samples for confidence interval at level 0.95
² need >= 4 samples to detect a difference at alpha level 0.05
│ bench_ncruces_direct.txt │ bench_ncruces_driver.txt │ bench_eatonphil_direct.txt │ bench_glebarez_driver.txt │ bench_mattn_driver.txt │ bench_modernc_driver.txt │ bench_tailscale_driver.txt │ bench_zombiezen_direct.txt │
│ B/op │ B/op vs base │ B/op vs base │ B/op vs base │ B/op vs base │ B/op vs base │ B/op vs base │ B/op vs base │
Populate/PopulateDB-12 2.362Mi ± ∞ ¹ 18.968Mi ± ∞ ¹ ~ (p=0.100 n=3) ² 5.771Mi ± ∞ ¹ ~ (p=0.100 n=3) ² 31.505Mi ± ∞ ¹ ~ (p=0.100 n=3) ² 273.296Mi ± ∞ ¹ ~ (p=0.100 n=3) ² 31.492Mi ± ∞ ¹ ~ (p=0.100 n=3) ² 18.963Mi ± ∞ ¹ ~ (p=0.100 n=3) ² 6.167Mi ± ∞ ¹ ~ (p=0.100 n=3) ²
Populate/PopulateDBWithTx-12 2.541Mi ± ∞ ¹ 32.394Mi ± ∞ ¹ ~ (p=0.100 n=3) ² 5.774Mi ± ∞ ¹ ~ (p=0.100 n=3) ² 44.496Mi ± ∞ ¹ ~ (p=0.100 n=3) ² 286.530Mi ± ∞ ¹ ~ (p=0.100 n=3) ² 44.762Mi ± ∞ ¹ ~ (p=0.100 n=3) ² 32.553Mi ± ∞ ¹ ~ (p=0.100 n=3) ² 6.167Mi ± ∞ ¹ ~ (p=0.100 n=3) ²
Populate/PopulateDBWithTxs-12 2.373Mi ± ∞ ¹ 31.219Mi ± ∞ ¹ ~ (p=0.100 n=3) ² 5.772Mi ± ∞ ¹ ~ (p=0.100 n=3) ² 43.667Mi ± ∞ ¹ ~ (p=0.100 n=3) ² 286.004Mi ± ∞ ¹ ~ (p=0.100 n=3) ² 43.848Mi ± ∞ ¹ ~ (p=0.100 n=3) ² 31.256Mi ± ∞ ¹ ~ (p=0.100 n=3) ² 6.213Mi ± ∞ ¹ ~ (p=0.100 n=3) ²
geomean 2.424Mi 26.77Mi +1004.38% 5.772Mi +138.13% 39.41Mi +1525.94% 281.9Mi +11529.03% 39.54Mi +1531.19% 26.82Mi +1006.52% 6.182Mi +155.05%
¹ need >= 6 samples for confidence interval at level 0.95
² need >= 4 samples to detect a difference at alpha level 0.05
│ bench_ncruces_direct.txt │ bench_ncruces_driver.txt │ bench_eatonphil_direct.txt │ bench_glebarez_driver.txt │ bench_mattn_driver.txt │ bench_modernc_driver.txt │ bench_tailscale_driver.txt │ bench_zombiezen_direct.txt │
│ allocs/op │ allocs/op vs base │ allocs/op vs base │ allocs/op vs base │ allocs/op vs base │ allocs/op vs base │ allocs/op vs base │ allocs/op vs base │
Populate/PopulateDB-12 140.0k ± ∞ ¹ 598.1k ± ∞ ¹ ~ (p=0.100 n=3) ² 394.0k ± ∞ ¹ ~ (p=0.100 n=3) ² 1209.5k ± ∞ ¹ ~ (p=0.100 n=3) ² 1057.5k ± ∞ ¹ ~ (p=0.100 n=3) ² 1209.4k ± ∞ ¹ ~ (p=0.100 n=3) ² 598.1k ± ∞ ¹ ~ (p=0.100 n=3) ² 445.1k ± ∞ ¹ ~ (p=0.100 n=3) ²
Populate/PopulateDBWithTx-12 140.1k ± ∞ ¹ 751.1k ± ∞ ¹ ~ (p=0.100 n=3) ² 394.0k ± ∞ ¹ ~ (p=0.100 n=3) ² 1362.4k ± ∞ ¹ ~ (p=0.100 n=3) ² 1210.5k ± ∞ ¹ ~ (p=0.100 n=3) ² 1362.3k ± ∞ ¹ ~ (p=0.100 n=3) ² 751.1k ± ∞ ¹ ~ (p=0.100 n=3) ² 445.1k ± ∞ ¹ ~ (p=0.100 n=3) ²
Populate/PopulateDBWithTxs-12 140.0k ± ∞ ¹ 765.2k ± ∞ ¹ ~ (p=0.100 n=3) ² 394.0k ± ∞ ¹ ~ (p=0.100 n=3) ² 1376.3k ± ∞ ¹ ~ (p=0.100 n=3) ² 1239.0k ± ∞ ¹ ~ (p=0.100 n=3) ² 1380.3k ± ∞ ¹ ~ (p=0.100 n=3) ² 767.2k ± ∞ ¹ ~ (p=0.100 n=3) ² 448.1k ± ∞ ¹ ~ (p=0.100 n=3) ²
geomean 140.1k 700.5k +400.17% 394.0k +181.34% 1.314M +838.08% 1.166M +732.65% 1.315M +838.94% 701.1k +400.60% 446.1k +218.50%
¹ need >= 6 samples for confidence interval at level 0.95
² need >= 4 samples to detect a difference at alpha level 0.05
-
ReadPost
Read one post, without a transaction.
-
ReadPostWithTx
Read one post, in a read transaction.
-
ReadPostAndComments
Read one post + comments, without a transaction.
-
ReadPostAndCommentsWithTx
Read one post + comments, in a read transaction.
-
WritePost
Write one post, without a transaction.
-
WritePostWithTx
Write one post, in a write transaction.
-
WritePostAndComments
Write one post + comments, without a transaction.
-
WritePostAndCommentsWithTx
Write one post + comments, in a write transaction.
-
ReadOrWritePostAndComments
Read or write one post + comments, without a transaction.
("Read or write" is determined by a "write rate" — either 10% or 90% in the current benchmarks.)
-
ReadOrWritePostAndCommentsWithTx
Read or write one post + comments, in a read or write transaction.
("Read or write" is determined by a "write rate" — either 10% or 90% in the current benchmarks.)
benchstat bench_ncruces_direct.txt bench_ncruces_driver.txt bench_eatonphil_direct.txt bench_glebarez_driver.txt bench_mattn_driver.txt bench_modernc_driver.txt bench_tailscale_driver.txt bench_zombiezen_direct.txt
goos: darwin
goarch: arm64
pkg: github.com/michaellenaghan/go-sqlite-bench
cpu: Apple M2 Pro
│ bench_ncruces_direct.txt │ bench_ncruces_driver.txt │ bench_eatonphil_direct.txt │ bench_glebarez_driver.txt │ bench_mattn_driver.txt │ bench_modernc_driver.txt │ bench_tailscale_driver.txt │ bench_zombiezen_direct.txt │
│ sec/op │ sec/op vs base │ sec/op vs base │ sec/op vs base │ sec/op vs base │ sec/op vs base │ sec/op vs base │ sec/op vs base │
ReadWrite/ReadPost-12 13.405µ ± ∞ ¹ 21.673µ ± ∞ ¹ ~ (p=0.100 n=3) ² 9.511µ ± ∞ ¹ ~ (p=0.100 n=3) ² 36.905µ ± ∞ ¹ ~ (p=0.100 n=3) ² 17.720µ ± ∞ ¹ ~ (p=0.100 n=3) ² 36.533µ ± ∞ ¹ ~ (p=0.100 n=3) ² 16.470µ ± ∞ ¹ ~ (p=0.100 n=3) ² 25.550µ ± ∞ ¹ ~ (p=0.100 n=3) ²
ReadWrite/ReadPostWithTx-12 15.449µ ± ∞ ¹ 37.716µ ± ∞ ¹ ~ (p=0.100 n=3) ² 9.971µ ± ∞ ¹ ~ (p=0.100 n=3) ² 39.974µ ± ∞ ¹ ~ (p=0.100 n=3) ² 22.053µ ± ∞ ¹ ~ (p=0.100 n=3) ² 40.170µ ± ∞ ¹ ~ (p=0.100 n=3) ² 24.476µ ± ∞ ¹ ~ (p=0.100 n=3) ² 25.849µ ± ∞ ¹ ~ (p=0.100 n=3) ²
ReadWrite/ReadPostAndComments-12 92.93µ ± ∞ ¹ 113.07µ ± ∞ ¹ ~ (p=0.100 n=3) ² 72.39µ ± ∞ ¹ ~ (p=0.100 n=3) ² 177.97µ ± ∞ ¹ ~ (p=0.100 n=3) ² 173.02µ ± ∞ ¹ ~ (p=0.100 n=3) ² 183.85µ ± ∞ ¹ ~ (p=0.100 n=3) ² 98.54µ ± ∞ ¹ ~ (p=0.100 n=3) ² 102.35µ ± ∞ ¹ ~ (p=0.100 n=3) ²
ReadWrite/ReadPostAndCommentsWithTx-12 92.03µ ± ∞ ¹ 141.16µ ± ∞ ¹ ~ (p=0.100 n=3) ² 70.75µ ± ∞ ¹ ~ (p=0.100 n=3) ² 179.73µ ± ∞ ¹ ~ (p=0.100 n=3) ² 176.75µ ± ∞ ¹ ~ (p=0.100 n=3) ² 185.43µ ± ∞ ¹ ~ (p=0.100 n=3) ² 110.15µ ± ∞ ¹ ~ (p=0.100 n=3) ² 101.50µ ± ∞ ¹ ~ (p=0.100 n=3) ²
ReadWrite/WritePost-12 134.2µ ± ∞ ¹ 119.4µ ± ∞ ¹ ~ (p=0.100 n=3) ² 124.1µ ± ∞ ¹ ~ (p=0.100 n=3) ² 158.0µ ± ∞ ¹ ~ (p=0.100 n=3) ² 126.1µ ± ∞ ¹ ~ (p=0.100 n=3) ² 153.9µ ± ∞ ¹ ~ (p=0.100 n=3) ² 122.7µ ± ∞ ¹ ~ (p=0.100 n=3) ² 176.8µ ± ∞ ¹ ~ (p=0.100 n=3) ²
ReadWrite/WritePostWithTx-12 139.8µ ± ∞ ¹ 128.8µ ± ∞ ¹ ~ (p=0.100 n=3) ² 129.3µ ± ∞ ¹ ~ (p=0.100 n=3) ² 157.6µ ± ∞ ¹ ~ (p=0.100 n=3) ² 134.8µ ± ∞ ¹ ~ (p=0.100 n=3) ² 159.2µ ± ∞ ¹ ~ (p=0.100 n=3) ² 124.3µ ± ∞ ¹ ~ (p=0.100 n=3) ² 176.6µ ± ∞ ¹ ~ (p=0.100 n=3) ²
ReadWrite/WritePostAndComments-12 2.618m ± ∞ ¹ 2.542m ± ∞ ¹ ~ (p=0.100 n=3) ² 2.160m ± ∞ ¹ ~ (p=0.100 n=3) ² 2.899m ± ∞ ¹ ~ (p=0.100 n=3) ² 2.341m ± ∞ ¹ ~ (p=0.100 n=3) ² 2.820m ± ∞ ¹ ~ (p=0.100 n=3) ² 2.146m ± ∞ ¹ ~ (p=0.100 n=3) ² 2.607m ± ∞ ¹ ~ (p=0.100 n=3) ²
ReadWrite/WritePostAndCommentsWithTx-12 1005.0µ ± ∞ ¹ 968.8µ ± ∞ ¹ ~ (p=0.100 n=3) ² 822.8µ ± ∞ ¹ ~ (p=0.100 n=3) ² 1241.9µ ± ∞ ¹ ~ (p=0.100 n=3) ² 997.4µ ± ∞ ¹ ~ (p=0.100 n=3) ² 1233.3µ ± ∞ ¹ ~ (p=0.100 n=3) ² 841.8µ ± ∞ ¹ ~ (p=0.100 n=3) ² 1102.5µ ± ∞ ¹ ~ (p=0.100 n=3) ²
ReadWrite/ReadOrWritePostAndComments/write_rate=10-12 430.2µ ± ∞ ¹ 350.0µ ± ∞ ¹ ~ (p=0.100 n=3) ² 311.6µ ± ∞ ¹ ~ (p=0.100 n=3) ² 461.7µ ± ∞ ¹ ~ (p=0.100 n=3) ² 400.9µ ± ∞ ¹ ~ (p=0.100 n=3) ² 451.4µ ± ∞ ¹ ~ (p=0.200 n=3) ² 303.2µ ± ∞ ¹ ~ (p=0.100 n=3) ² 434.8µ ± ∞ ¹ ~ (p=1.000 n=3) ²
ReadWrite/ReadOrWritePostAndComments/write_rate=90-12 2.409m ± ∞ ¹ 2.315m ± ∞ ¹ ~ (p=0.100 n=3) ² 1.990m ± ∞ ¹ ~ (p=0.100 n=3) ² 2.639m ± ∞ ¹ ~ (p=0.100 n=3) ² 2.088m ± ∞ ¹ ~ (p=0.100 n=3) ² 2.586m ± ∞ ¹ ~ (p=0.100 n=3) ² 2.002m ± ∞ ¹ ~ (p=0.100 n=3) ² 2.380m ± ∞ ¹ ~ (p=0.700 n=3) ²
ReadWrite/ReadOrWritePostAndCommentsParallel/write_rate=10-12 368.8µ ± ∞ ¹ 375.6µ ± ∞ ¹ ~ (p=0.400 n=3) ² 299.8µ ± ∞ ¹ ~ (p=0.100 n=3) ² 414.2µ ± ∞ ¹ ~ (p=0.100 n=3) ² 336.2µ ± ∞ ¹ ~ (p=0.100 n=3) ² 429.0µ ± ∞ ¹ ~ (p=0.100 n=3) ² 300.0µ ± ∞ ¹ ~ (p=0.100 n=3) ² 415.5µ ± ∞ ¹ ~ (p=0.700 n=3) ²
ReadWrite/ReadOrWritePostAndCommentsParallel/write_rate=90-12 3.158m ± ∞ ¹ 3.022m ± ∞ ¹ ~ (p=0.700 n=3) ² 2.462m ± ∞ ¹ ~ (p=0.100 n=3) ² 2.758m ± ∞ ¹ ~ (p=0.100 n=3) ² 2.455m ± ∞ ¹ ~ (p=0.200 n=3) ² 2.880m ± ∞ ¹ ~ (p=0.200 n=3) ² 2.481m ± ∞ ¹ ~ (p=0.100 n=3) ² 3.386m ± ∞ ¹ ~ (p=0.700 n=3) ²
ReadWrite/ReadOrWritePostAndCommentsWithTx/write_rate=10-12 258.1µ ± ∞ ¹ 218.7µ ± ∞ ¹ ~ (p=0.100 n=3) ² 195.5µ ± ∞ ¹ ~ (p=0.100 n=3) ² 287.5µ ± ∞ ¹ ~ (p=0.100 n=3) ² 270.0µ ± ∞ ¹ ~ (p=0.700 n=3) ² 309.6µ ± ∞ ¹ ~ (p=0.100 n=3) ² 182.7µ ± ∞ ¹ ~ (p=0.100 n=3) ² 270.2µ ± ∞ ¹ ~ (p=0.700 n=3) ²
ReadWrite/ReadOrWritePostAndCommentsWithTx/write_rate=90-12 970.3µ ± ∞ ¹ 905.2µ ± ∞ ¹ ~ (p=0.100 n=3) ² 760.2µ ± ∞ ¹ ~ (p=0.100 n=3) ² 1162.7µ ± ∞ ¹ ~ (p=0.700 n=3) ² 937.3µ ± ∞ ¹ ~ (p=0.100 n=3) ² 1177.9µ ± ∞ ¹ ~ (p=0.700 n=3) ² 770.0µ ± ∞ ¹ ~ (p=0.100 n=3) ² 1049.1µ ± ∞ ¹ ~ (p=0.700 n=3) ²
ReadWrite/ReadOrWritePostAndCommentsWithTxParallel/write_rate=10-12 111.8µ ± ∞ ¹ 119.7µ ± ∞ ¹ ~ (p=0.200 n=3) ² 129.6µ ± ∞ ¹ ~ (p=0.100 n=3) ² 238.3µ ± ∞ ¹ ~ (p=0.100 n=3) ² 185.1µ ± ∞ ¹ ~ (p=0.100 n=3) ² 230.3µ ± ∞ ¹ ~ (p=0.100 n=3) ² 134.1µ ± ∞ ¹ ~ (p=0.100 n=3) ² 186.4µ ± ∞ ¹ ~ (p=0.100 n=3) ²
ReadWrite/ReadOrWritePostAndCommentsWithTxParallel/write_rate=90-12 869.3µ ± ∞ ¹ 887.8µ ± ∞ ¹ ~ (p=0.400 n=3) ² 940.8µ ± ∞ ¹ ~ (p=0.400 n=3) ² 3205.2µ ± ∞ ¹ ~ (p=0.100 n=3) ² 1259.0µ ± ∞ ¹ ~ (p=0.100 n=3) ² 5038.3µ ± ∞ ¹ ~ (p=0.100 n=3) ² 926.8µ ± ∞ ¹ ~ (p=0.100 n=3) ² 2418.6µ ± ∞ ¹ ~ (p=0.100 n=3) ²
geomean 290.5µ 315.3µ +8.57% 239.2µ -17.65% 432.6µ +48.93% 329.5µ +13.45% 447.7µ +54.13% 273.0µ -6.01% 368.3µ +26.80%
¹ need >= 6 samples for confidence interval at level 0.95
² need >= 4 samples to detect a difference at alpha level 0.05
│ bench_ncruces_direct.txt │ bench_ncruces_driver.txt │ bench_eatonphil_direct.txt │ bench_glebarez_driver.txt │ bench_mattn_driver.txt │ bench_modernc_driver.txt │ bench_tailscale_driver.txt │ bench_zombiezen_direct.txt │
│ B/op │ B/op vs base │ B/op vs base │ B/op vs base │ B/op vs base │ B/op vs base │ B/op vs base │ B/op vs base │
ReadWrite/ReadPost-12 40.17Ki ± ∞ ¹ 41.23Ki ± ∞ ¹ ~ (p=0.100 n=3) ² 40.18Ki ± ∞ ¹ ~ (p=0.400 n=3) ² 81.35Ki ± ∞ ¹ ~ (p=0.100 n=3) ² 41.39Ki ± ∞ ¹ ~ (p=0.100 n=3) ² 81.30Ki ± ∞ ¹ ~ (p=0.100 n=3) ² 41.19Ki ± ∞ ¹ ~ (p=0.100 n=3) ² 40.56Ki ± ∞ ¹ ~ (p=0.100 n=3) ²
ReadWrite/ReadPostWithTx-12 40.17Ki ± ∞ ¹ 42.29Ki ± ∞ ¹ ~ (p=0.100 n=3) ² 40.18Ki ± ∞ ¹ ~ (p=0.100 n=3) ² 82.22Ki ± ∞ ¹ ~ (p=0.100 n=3) ² 42.60Ki ± ∞ ¹ ~ (p=0.100 n=3) ² 82.36Ki ± ∞ ¹ ~ (p=0.100 n=3) ² 42.13Ki ± ∞ ¹ ~ (p=0.100 n=3) ² 40.61Ki ± ∞ ¹ ~ (p=0.100 n=3) ²
ReadWrite/ReadPostAndComments-12 250.2Ki ± ∞ ¹ 260.3Ki ± ∞ ¹ ~ (p=0.100 n=3) ² 250.2Ki ± ∞ ¹ ~ (p=0.600 n=3) ² 501.7Ki ± ∞ ¹ ~ (p=0.100 n=3) ² 263.9Ki ± ∞ ¹ ~ (p=0.100 n=3) ² 501.7Ki ± ∞ ¹ ~ (p=0.100 n=3) ² 257.2Ki ± ∞ ¹ ~ (p=0.100 n=3) ² 250.6Ki ± ∞ ¹ ~ (p=0.100 n=3) ²
ReadWrite/ReadPostAndCommentsWithTx-12 250.1Ki ± ∞ ¹ 261.9Ki ± ∞ ¹ ~ (p=0.100 n=3) ² 250.2Ki ± ∞ ¹ ~ (p=0.100 n=3) ² 503.1Ki ± ∞ ¹ ~ (p=0.100 n=3) ² 265.6Ki ± ∞ ¹ ~ (p=0.100 n=3) ² 503.3Ki ± ∞ ¹ ~ (p=0.100 n=3) ² 258.7Ki ± ∞ ¹ ~ (p=0.100 n=3) ² 250.6Ki ± ∞ ¹ ~ (p=0.100 n=3) ²
ReadWrite/WritePost-12 0.00 ± ∞ ¹ 240.00 ± ∞ ¹ ~ (p=0.100 n=3) ² 48.00 ± ∞ ¹ ~ (p=0.400 n=3) ² 496.00 ± ∞ ¹ ~ (p=0.100 n=3) ² 42288.00 ± ∞ ¹ ~ (p=0.100 n=3) ² 496.00 ± ∞ ¹ ~ (p=0.100 n=3) ² 240.00 ± ∞ ¹ ~ (p=0.100 n=3) ² 410.00 ± ∞ ¹ ~ (p=0.100 n=3) ²
ReadWrite/WritePostWithTx-12 0.00 ± ∞ ¹ 857.00 ± ∞ ¹ ~ (p=0.100 n=3) ² 48.00 ± ∞ ¹ ~ (p=0.100 n=3) ² 1104.00 ± ∞ ¹ ~ (p=0.100 n=3) ² 43255.00 ± ∞ ¹ ~ (p=0.100 n=3) ² 1290.00 ± ∞ ¹ ~ (p=0.100 n=3) ² 880.00 ± ∞ ¹ ~ (p=0.100 n=3) ² 458.00 ± ∞ ¹ ~ (p=0.100 n=3) ²
ReadWrite/WritePostAndComments-12 0.000Ki ± ∞ ¹ 13.907Ki ± ∞ ¹ ~ (p=0.100 n=3) ² 2.781Ki ± ∞ ¹ ~ (p=0.100 n=3) ² 26.663Ki ± ∞ ¹ ~ (p=0.100 n=3) ² 308.876Ki ± ∞ ¹ ~ (p=0.100 n=3) ² 26.669Ki ± ∞ ¹ ~ (p=0.100 n=3) ² 13.906Ki ± ∞ ¹ ~ (p=0.100 n=3) ² 3.535Ki ± ∞ ¹ ~ (p=0.100 n=3) ²
ReadWrite/WritePostAndCommentsWithTx-12 0.000Ki ± ∞ ¹ 26.448Ki ± ∞ ¹ ~ (p=0.100 n=3) ² 2.781Ki ± ∞ ¹ ~ (p=0.100 n=3) ² 39.193Ki ± ∞ ¹ ~ (p=0.100 n=3) ² 321.743Ki ± ∞ ¹ ~ (p=0.100 n=3) ² 39.380Ki ± ∞ ¹ ~ (p=0.100 n=3) ² 26.465Ki ± ∞ ¹ ~ (p=0.100 n=3) ² 3.579Ki ± ∞ ¹ ~ (p=0.100 n=3) ²
ReadWrite/ReadOrWritePostAndComments/write_rate=10-12 224.2Ki ± ∞ ¹ 236.6Ki ± ∞ ¹ ~ (p=0.100 n=3) ² 226.6Ki ± ∞ ¹ ~ (p=0.400 n=3) ² 453.2Ki ± ∞ ¹ ~ (p=0.100 n=3) ² 268.4Ki ± ∞ ¹ ~ (p=0.100 n=3) ² 455.5Ki ± ∞ ¹ ~ (p=0.100 n=3) ² 233.0Ki ± ∞ ¹ ~ (p=0.100 n=3) ² 224.7Ki ± ∞ ¹ ~ (p=0.400 n=3) ²
ReadWrite/ReadOrWritePostAndComments/write_rate=90-12 22.70Ki ± ∞ ¹ 38.12Ki ± ∞ ¹ ~ (p=0.100 n=3) ² 27.95Ki ± ∞ ¹ ~ (p=1.000 n=3) ² 66.26Ki ± ∞ ¹ ~ (p=0.100 n=3) ² 303.61Ki ± ∞ ¹ ~ (p=0.100 n=3) ² 76.57Ki ± ∞ ¹ ~ (p=0.100 n=3) ² 34.67Ki ± ∞ ¹ ~ (p=0.100 n=3) ² 30.29Ki ± ∞ ¹ ~ (p=0.700 n=3) ²
ReadWrite/ReadOrWritePostAndCommentsParallel/write_rate=10-12 225.9Ki ± ∞ ¹ 236.1Ki ± ∞ ¹ ~ (p=0.100 n=3) ² 225.2Ki ± ∞ ¹ ~ (p=0.200 n=3) ² 456.5Ki ± ∞ ¹ ~ (p=0.100 n=3) ² 268.4Ki ± ∞ ¹ ~ (p=0.100 n=3) ² 453.6Ki ± ∞ ¹ ~ (p=0.100 n=3) ² 232.4Ki ± ∞ ¹ ~ (p=0.100 n=3) ² 225.4Ki ± ∞ ¹ ~ (p=0.700 n=3) ²
ReadWrite/ReadOrWritePostAndCommentsParallel/write_rate=90-12 30.05Ki ± ∞ ¹ 40.05Ki ± ∞ ¹ ~ (p=0.100 n=3) ² 27.09Ki ± ∞ ¹ ~ (p=0.700 n=3) ² 69.96Ki ± ∞ ¹ ~ (p=0.100 n=3) ² 303.92Ki ± ∞ ¹ ~ (p=0.100 n=3) ² 74.44Ki ± ∞ ¹ ~ (p=0.100 n=3) ² 40.76Ki ± ∞ ¹ ~ (p=0.100 n=3) ² 26.83Ki ± ∞ ¹ ~ (p=1.000 n=3) ²
ReadWrite/ReadOrWritePostAndCommentsWithTx/write_rate=10-12 225.1Ki ± ∞ ¹ 238.7Ki ± ∞ ¹ ~ (p=0.100 n=3) ² 225.3Ki ± ∞ ¹ ~ (p=0.700 n=3) ² 457.7Ki ± ∞ ¹ ~ (p=0.100 n=3) ² 271.4Ki ± ∞ ¹ ~ (p=0.100 n=3) ² 452.7Ki ± ∞ ¹ ~ (p=0.100 n=3) ² 234.9Ki ± ∞ ¹ ~ (p=0.100 n=3) ² 225.4Ki ± ∞ ¹ ~ (p=0.400 n=3) ²
ReadWrite/ReadOrWritePostAndCommentsWithTx/write_rate=90-12 23.22Ki ± ∞ ¹ 50.64Ki ± ∞ ¹ ~ (p=0.100 n=3) ² 30.00Ki ± ∞ ¹ ~ (p=0.100 n=3) ² 81.15Ki ± ∞ ¹ ~ (p=0.100 n=3) ² 315.79Ki ± ∞ ¹ ~ (p=0.100 n=3) ² 90.18Ki ± ∞ ¹ ~ (p=0.100 n=3) ² 51.17Ki ± ∞ ¹ ~ (p=0.100 n=3) ² 28.61Ki ± ∞ ¹ ~ (p=0.100 n=3) ²
ReadWrite/ReadOrWritePostAndCommentsWithTxParallel/write_rate=10-12 225.2Ki ± ∞ ¹ 237.9Ki ± ∞ ¹ ~ (p=0.100 n=3) ² 226.0Ki ± ∞ ¹ ~ (p=0.100 n=3) ² 455.3Ki ± ∞ ¹ ~ (p=0.100 n=3) ² 271.4Ki ± ∞ ¹ ~ (p=0.100 n=3) ² 457.3Ki ± ∞ ¹ ~ (p=0.100 n=3) ² 235.1Ki ± ∞ ¹ ~ (p=0.100 n=3) ² 226.5Ki ± ∞ ¹ ~ (p=0.200 n=3) ²
ReadWrite/ReadOrWritePostAndCommentsWithTxParallel/write_rate=90-12 25.64Ki ± ∞ ¹ 50.02Ki ± ∞ ¹ ~ (p=0.100 n=3) ² 27.23Ki ± ∞ ¹ ~ (p=0.400 n=3) ² 82.79Ki ± ∞ ¹ ~ (p=0.100 n=3) ² 315.55Ki ± ∞ ¹ ~ (p=0.100 n=3) ² 83.97Ki ± ∞ ¹ ~ (p=0.100 n=3) ² 48.76Ki ± ∞ ¹ ~ (p=0.100 n=3) ² 32.62Ki ± ∞ ¹ ~ (p=0.100 n=3) ²
geomean ³ 42.31Ki ? 21.88Ki ? 75.99Ki ? 178.3Ki ? 78.31Ki ? 41.90Ki ? 30.12Ki ?
¹ need >= 6 samples for confidence interval at level 0.95
² need >= 4 samples to detect a difference at alpha level 0.05
³ summaries must be >0 to compute geomean
│ bench_ncruces_direct.txt │ bench_ncruces_driver.txt │ bench_eatonphil_direct.txt │ bench_glebarez_driver.txt │ bench_mattn_driver.txt │ bench_modernc_driver.txt │ bench_tailscale_driver.txt │ bench_zombiezen_direct.txt │
│ allocs/op │ allocs/op vs base │ allocs/op vs base │ allocs/op vs base │ allocs/op vs base │ allocs/op vs base │ allocs/op vs base │ allocs/op vs base │
ReadWrite/ReadPost-12 5.000 ± ∞ ¹ 39.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 6.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 43.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 42.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 42.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 31.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 12.000 ± ∞ ¹ ~ (p=0.100 n=3) ²
ReadWrite/ReadPostWithTx-12 5.000 ± ∞ ¹ 60.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 6.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 57.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 70.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 60.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 50.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 15.000 ± ∞ ¹ ~ (p=0.100 n=3) ²
ReadWrite/ReadPostAndComments-12 262.0 ± ∞ ¹ 773.0 ± ∞ ¹ ~ (p=0.100 n=3) ² 264.0 ± ∞ ¹ ~ (p=0.100 n=3) ² 976.0 ± ∞ ¹ ~ (p=0.100 n=3) ² 684.0 ± ∞ ¹ ~ (p=0.100 n=3) ² 975.0 ± ∞ ¹ ~ (p=0.100 n=3) ² 562.0 ± ∞ ¹ ~ (p=0.100 n=3) ² 272.0 ± ∞ ¹ ~ (p=0.100 n=3) ²
ReadWrite/ReadPostAndCommentsWithTx-12 262.0 ± ∞ ¹ 801.0 ± ∞ ¹ ~ (p=0.100 n=3) ² 264.0 ± ∞ ¹ ~ (p=0.100 n=3) ² 997.0 ± ∞ ¹ ~ (p=0.100 n=3) ² 719.0 ± ∞ ¹ ~ (p=0.100 n=3) ² 1000.0 ± ∞ ¹ ~ (p=0.100 n=3) ² 587.0 ± ∞ ¹ ~ (p=0.100 n=3) ² 275.0 ± ∞ ¹ ~ (p=0.100 n=3) ²
ReadWrite/WritePost-12 0.000 ± ∞ ¹ 7.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 3.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 17.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 15.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 17.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 7.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 9.000 ± ∞ ¹ ~ (p=0.100 n=3) ²
ReadWrite/WritePostWithTx-12 0.000 ± ∞ ¹ 18.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 3.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 28.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 40.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 32.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 20.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 12.000 ± ∞ ¹ ~ (p=0.100 n=3) ²
ReadWrite/WritePostAndComments-12 0.0 ± ∞ ¹ 407.0 ± ∞ ¹ ~ (p=0.100 n=3) ² 203.0 ± ∞ ¹ ~ (p=0.100 n=3) ² 967.0 ± ∞ ¹ ~ (p=0.100 n=3) ² 815.0 ± ∞ ¹ ~ (p=0.100 n=3) ² 967.0 ± ∞ ¹ ~ (p=0.100 n=3) ² 407.0 ± ∞ ¹ ~ (p=0.100 n=3) ² 259.0 ± ∞ ¹ ~ (p=0.100 n=3) ²
ReadWrite/WritePostAndCommentsWithTx-12 0.0 ± ∞ ¹ 574.0 ± ∞ ¹ ~ (p=0.100 n=3) ² 203.0 ± ∞ ¹ ~ (p=0.100 n=3) ² 1134.0 ± ∞ ¹ ~ (p=0.100 n=3) ² 996.0 ± ∞ ¹ ~ (p=0.100 n=3) ² 1138.0 ± ∞ ¹ ~ (p=0.100 n=3) ² 576.0 ± ∞ ¹ ~ (p=0.100 n=3) ² 262.0 ± ∞ ¹ ~ (p=0.100 n=3) ²
ReadWrite/ReadOrWritePostAndComments/write_rate=10-12 234.0 ± ∞ ¹ 737.0 ± ∞ ¹ ~ (p=0.100 n=3) ² 258.0 ± ∞ ¹ ~ (p=0.100 n=3) ² 975.0 ± ∞ ¹ ~ (p=0.100 n=3) ² 697.0 ± ∞ ¹ ~ (p=0.100 n=3) ² 975.0 ± ∞ ¹ ~ (p=0.100 n=3) ² 546.0 ± ∞ ¹ ~ (p=0.100 n=3) ² 271.0 ± ∞ ¹ ~ (p=0.100 n=3) ²
ReadWrite/ReadOrWritePostAndComments/write_rate=90-12 23.00 ± ∞ ¹ 443.00 ± ∞ ¹ ~ (p=0.100 n=3) ² 209.00 ± ∞ ¹ ~ (p=0.100 n=3) ² 967.00 ± ∞ ¹ ~ (p=0.100 n=3) ² 799.00 ± ∞ ¹ ~ (p=0.100 n=3) ² 968.00 ± ∞ ¹ ~ (p=0.100 n=3) ² 420.00 ± ∞ ¹ ~ (p=0.100 n=3) ² 260.00 ± ∞ ¹ ~ (p=0.100 n=3) ²
ReadWrite/ReadOrWritePostAndCommentsParallel/write_rate=10-12 236.0 ± ∞ ¹ 737.0 ± ∞ ¹ ~ (p=0.100 n=3) ² 257.0 ± ∞ ¹ ~ (p=0.100 n=3) ² 975.0 ± ∞ ¹ ~ (p=0.100 n=3) ² 697.0 ± ∞ ¹ ~ (p=0.100 n=3) ² 975.0 ± ∞ ¹ ~ (p=0.100 n=3) ² 546.0 ± ∞ ¹ ~ (p=0.100 n=3) ² 271.0 ± ∞ ¹ ~ (p=0.100 n=3) ²
ReadWrite/ReadOrWritePostAndCommentsParallel/write_rate=90-12 30.00 ± ∞ ¹ 445.00 ± ∞ ¹ ~ (p=0.100 n=3) ² 209.00 ± ∞ ¹ ~ (p=0.100 n=3) ² 968.00 ± ∞ ¹ ~ (p=0.100 n=3) ² 800.00 ± ∞ ¹ ~ (p=0.100 n=3) ² 968.00 ± ∞ ¹ ~ (p=0.100 n=3) ² 424.00 ± ∞ ¹ ~ (p=0.100 n=3) ² 260.00 ± ∞ ¹ ~ (p=0.100 n=3) ²
ReadWrite/ReadOrWritePostAndCommentsWithTx/write_rate=10-12 235.0 ± ∞ ¹ 778.0 ± ∞ ¹ ~ (p=0.100 n=3) ² 257.0 ± ∞ ¹ ~ (p=0.100 n=3) ² 1010.0 ± ∞ ¹ ~ (p=0.100 n=3) ² 747.0 ± ∞ ¹ ~ (p=0.100 n=3) ² 1015.0 ± ∞ ¹ ~ (p=0.100 n=3) ² 586.0 ± ∞ ¹ ~ (p=0.100 n=3) ² 274.0 ± ∞ ¹ ~ (p=0.100 n=3) ²
ReadWrite/ReadOrWritePostAndCommentsWithTx/write_rate=90-12 24.00 ± ∞ ¹ 597.00 ± ∞ ¹ ~ (p=0.100 n=3) ² 209.00 ± ∞ ¹ ~ (p=0.100 n=3) ² 1121.00 ± ∞ ¹ ~ (p=0.100 n=3) ² 966.00 ± ∞ ¹ ~ (p=0.100 n=3) ² 1123.00 ± ∞ ¹ ~ (p=0.100 n=3) ² 577.00 ± ∞ ¹ ~ (p=0.100 n=3) ² 263.00 ± ∞ ¹ ~ (p=0.100 n=3) ²
ReadWrite/ReadOrWritePostAndCommentsWithTxParallel/write_rate=10-12 235.0 ± ∞ ¹ 777.0 ± ∞ ¹ ~ (p=0.100 n=3) ² 258.0 ± ∞ ¹ ~ (p=0.100 n=3) ² 1011.0 ± ∞ ¹ ~ (p=0.100 n=3) ² 747.0 ± ∞ ¹ ~ (p=0.100 n=3) ² 1014.0 ± ∞ ¹ ~ (p=0.100 n=3) ² 586.0 ± ∞ ¹ ~ (p=0.100 n=3) ² 274.0 ± ∞ ¹ ~ (p=0.100 n=3) ²
ReadWrite/ReadOrWritePostAndCommentsWithTxParallel/write_rate=90-12 26.00 ± ∞ ¹ 596.00 ± ∞ ¹ ~ (p=0.100 n=3) ² 209.00 ± ∞ ¹ ~ (p=0.100 n=3) ² 1121.00 ± ∞ ¹ ~ (p=0.100 n=3) ² 965.00 ± ∞ ¹ ~ (p=0.100 n=3) ² 1125.00 ± ∞ ¹ ~ (p=0.100 n=3) ² 577.00 ± ∞ ¹ ~ (p=0.100 n=3) ² 263.00 ± ∞ ¹ ~ (p=0.100 n=3) ²
geomean ³ 273.5 ? 85.27 ? 431.1 ? 368.1 ? 436.0 ? 237.4 ? 122.4 ?
¹ need >= 6 samples for confidence interval at level 0.95
² need >= 4 samples to detect a difference at alpha level 0.05
³ summaries must be >0 to compute geomean
benchstat bench_ncruces_direct.txt bench_ncruces_driver.txt bench_eatonphil_direct.txt bench_glebarez_driver.txt bench_mattn_driver.txt bench_modernc_driver.txt bench_tailscale_driver.txt bench_zombiezen_direct.txt
goos: darwin
goarch: arm64
pkg: github.com/michaellenaghan/go-sqlite-bench
cpu: Apple M2 Pro
│ bench_ncruces_direct.txt │ bench_ncruces_driver.txt │ bench_eatonphil_direct.txt │ bench_glebarez_driver.txt │ bench_mattn_driver.txt │ bench_modernc_driver.txt │ bench_tailscale_driver.txt │ bench_zombiezen_direct.txt │
│ sec/op │ sec/op vs base │ sec/op vs base │ sec/op vs base │ sec/op vs base │ sec/op vs base │ sec/op vs base │ sec/op vs base │
ReadWrite/ReadOrWritePostAndCommentsParallel/write_rate=10 451.0µ ± ∞ ¹ 346.3µ ± ∞ ¹ ~ (p=0.100 n=3) ² 319.7µ ± ∞ ¹ ~ (p=0.100 n=3) ² 428.7µ ± ∞ ¹ ~ (p=0.100 n=3) ² 347.2µ ± ∞ ¹ ~ (p=0.100 n=3) ² 465.1µ ± ∞ ¹ ~ (p=0.100 n=3) ² 287.3µ ± ∞ ¹ ~ (p=0.100 n=3) ² 442.1µ ± ∞ ¹ ~ (p=0.700 n=3) ²
ReadWrite/ReadOrWritePostAndCommentsParallel/write_rate=10-2 374.1µ ± ∞ ¹ 336.3µ ± ∞ ¹ ~ (p=0.100 n=3) ² 280.6µ ± ∞ ¹ ~ (p=0.100 n=3) ² 389.2µ ± ∞ ¹ ~ (p=0.400 n=3) ² 310.2µ ± ∞ ¹ ~ (p=0.100 n=3) ² 448.2µ ± ∞ ¹ ~ (p=0.100 n=3) ² 256.6µ ± ∞ ¹ ~ (p=0.100 n=3) ² 378.2µ ± ∞ ¹ ~ (p=1.000 n=3) ²
ReadWrite/ReadOrWritePostAndCommentsParallel/write_rate=10-4 334.0µ ± ∞ ¹ 352.5µ ± ∞ ¹ ~ (p=0.700 n=3) ² 286.9µ ± ∞ ¹ ~ (p=0.100 n=3) ² 365.3µ ± ∞ ¹ ~ (p=0.100 n=3) ² 297.2µ ± ∞ ¹ ~ (p=0.100 n=3) ² 389.0µ ± ∞ ¹ ~ (p=0.100 n=3) ² 271.5µ ± ∞ ¹ ~ (p=0.100 n=3) ² 364.5µ ± ∞ ¹ ~ (p=0.100 n=3) ²
ReadWrite/ReadOrWritePostAndCommentsParallel/write_rate=10-8 373.3µ ± ∞ ¹ 350.2µ ± ∞ ¹ ~ (p=0.400 n=3) ² 288.7µ ± ∞ ¹ ~ (p=0.100 n=3) ² 408.8µ ± ∞ ¹ ~ (p=0.200 n=3) ² 326.1µ ± ∞ ¹ ~ (p=0.100 n=3) ² 418.4µ ± ∞ ¹ ~ (p=0.100 n=3) ² 268.1µ ± ∞ ¹ ~ (p=0.100 n=3) ² 356.8µ ± ∞ ¹ ~ (p=1.000 n=3) ²
ReadWrite/ReadOrWritePostAndCommentsParallel/write_rate=10-16 427.3µ ± ∞ ¹ 430.7µ ± ∞ ¹ ~ (p=0.700 n=3) ² 331.7µ ± ∞ ¹ ~ (p=0.100 n=3) ² 441.8µ ± ∞ ¹ ~ (p=0.400 n=3) ² 339.5µ ± ∞ ¹ ~ (p=0.100 n=3) ² 481.7µ ± ∞ ¹ ~ (p=0.400 n=3) ² 307.8µ ± ∞ ¹ ~ (p=0.100 n=3) ² 387.3µ ± ∞ ¹ ~ (p=0.100 n=3) ²
ReadWrite/ReadOrWritePostAndCommentsParallel/write_rate=90 2.413m ± ∞ ¹ 2.294m ± ∞ ¹ ~ (p=0.100 n=3) ² 1.946m ± ∞ ¹ ~ (p=0.100 n=3) ² 2.446m ± ∞ ¹ ~ (p=0.100 n=3) ² 2.042m ± ∞ ¹ ~ (p=0.100 n=3) ² 2.638m ± ∞ ¹ ~ (p=0.100 n=3) ² 2.022m ± ∞ ¹ ~ (p=0.100 n=3) ² 2.386m ± ∞ ¹ ~ (p=0.700 n=3) ²
ReadWrite/ReadOrWritePostAndCommentsParallel/write_rate=90-2 2.562m ± ∞ ¹ 2.446m ± ∞ ¹ ~ (p=0.100 n=3) ² 1.935m ± ∞ ¹ ~ (p=0.100 n=3) ² 2.607m ± ∞ ¹ ~ (p=1.000 n=3) ² 2.063m ± ∞ ¹ ~ (p=0.100 n=3) ² 3.033m ± ∞ ¹ ~ (p=0.100 n=3) ² 1.989m ± ∞ ¹ ~ (p=0.100 n=3) ² 2.418m ± ∞ ¹ ~ (p=0.100 n=3) ²
ReadWrite/ReadOrWritePostAndCommentsParallel/write_rate=90-4 2.705m ± ∞ ¹ 2.685m ± ∞ ¹ ~ (p=0.700 n=3) ² 2.086m ± ∞ ¹ ~ (p=0.100 n=3) ² 2.549m ± ∞ ¹ ~ (p=0.100 n=3) ² 2.105m ± ∞ ¹ ~ (p=0.100 n=3) ² 2.899m ± ∞ ¹ ~ (p=0.100 n=3) ² 2.091m ± ∞ ¹ ~ (p=0.100 n=3) ² 2.378m ± ∞ ¹ ~ (p=0.100 n=3) ²
ReadWrite/ReadOrWritePostAndCommentsParallel/write_rate=90-8 2.836m ± ∞ ¹ 2.942m ± ∞ ¹ ~ (p=0.700 n=3) ² 2.301m ± ∞ ¹ ~ (p=0.100 n=3) ² 2.912m ± ∞ ¹ ~ (p=0.700 n=3) ² 2.344m ± ∞ ¹ ~ (p=0.100 n=3) ² 3.115m ± ∞ ¹ ~ (p=0.700 n=3) ² 2.222m ± ∞ ¹ ~ (p=0.100 n=3) ² 2.522m ± ∞ ¹ ~ (p=0.100 n=3) ²
ReadWrite/ReadOrWritePostAndCommentsParallel/write_rate=90-16 3.367m ± ∞ ¹ 3.339m ± ∞ ¹ ~ (p=0.400 n=3) ² 2.855m ± ∞ ¹ ~ (p=0.100 n=3) ² 3.425m ± ∞ ¹ ~ (p=0.700 n=3) ² 2.818m ± ∞ ¹ ~ (p=0.100 n=3) ² 3.958m ± ∞ ¹ ~ (p=0.100 n=3) ² 2.734m ± ∞ ¹ ~ (p=0.100 n=3) ² 8.487m ± ∞ ¹ ~ (p=0.100 n=3) ²
ReadWrite/ReadOrWritePostAndCommentsWithTxParallel/write_rate=10 275.0µ ± ∞ ¹ 218.0µ ± ∞ ¹ ~ (p=0.100 n=3) ² 190.9µ ± ∞ ¹ ~ (p=0.100 n=3) ² 288.1µ ± ∞ ¹ ~ (p=0.100 n=3) ² 230.3µ ± ∞ ¹ ~ (p=0.100 n=3) ² 307.8µ ± ∞ ¹ ~ (p=0.100 n=3) ² 181.1µ ± ∞ ¹ ~ (p=0.100 n=3) ² 288.1µ ± ∞ ¹ ~ (p=0.100 n=3) ²
ReadWrite/ReadOrWritePostAndCommentsWithTxParallel/write_rate=10-2 163.4µ ± ∞ ¹ 138.0µ ± ∞ ¹ ~ (p=0.100 n=3) ² 144.4µ ± ∞ ¹ ~ (p=0.100 n=3) ² 222.8µ ± ∞ ¹ ~ (p=0.100 n=3) ² 171.8µ ± ∞ ¹ ~ (p=0.400 n=3) ² 220.0µ ± ∞ ¹ ~ (p=0.100 n=3) ² 130.2µ ± ∞ ¹ ~ (p=0.100 n=3) ² 199.6µ ± ∞ ¹ ~ (p=0.100 n=3) ²
ReadWrite/ReadOrWritePostAndCommentsWithTxParallel/write_rate=10-4 119.8µ ± ∞ ¹ 119.2µ ± ∞ ¹ ~ (p=1.000 n=3) ² 121.9µ ± ∞ ¹ ~ (p=0.700 n=3) ² 199.9µ ± ∞ ¹ ~ (p=0.100 n=3) ² 142.9µ ± ∞ ¹ ~ (p=0.100 n=3) ² 195.6µ ± ∞ ¹ ~ (p=0.100 n=3) ² 115.4µ ± ∞ ¹ ~ (p=0.700 n=3) ² 175.9µ ± ∞ ¹ ~ (p=0.100 n=3) ²
ReadWrite/ReadOrWritePostAndCommentsWithTxParallel/write_rate=10-8 110.3µ ± ∞ ¹ 110.3µ ± ∞ ¹ ~ (p=1.000 n=3) ² 121.5µ ± ∞ ¹ ~ (p=0.100 n=3) ² 209.7µ ± ∞ ¹ ~ (p=0.100 n=3) ² 180.3µ ± ∞ ¹ ~ (p=0.100 n=3) ² 223.8µ ± ∞ ¹ ~ (p=0.100 n=3) ² 126.7µ ± ∞ ¹ ~ (p=0.100 n=3) ² 188.7µ ± ∞ ¹ ~ (p=0.100 n=3) ²
ReadWrite/ReadOrWritePostAndCommentsWithTxParallel/write_rate=10-16 122.0µ ± ∞ ¹ 129.7µ ± ∞ ¹ ~ (p=0.100 n=3) ² 131.6µ ± ∞ ¹ ~ (p=0.100 n=3) ² 240.5µ ± ∞ ¹ ~ (p=0.100 n=3) ² 222.1µ ± ∞ ¹ ~ (p=0.100 n=3) ² 234.5µ ± ∞ ¹ ~ (p=0.100 n=3) ² 136.7µ ± ∞ ¹ ~ (p=0.100 n=3) ² 204.4µ ± ∞ ¹ ~ (p=0.100 n=3) ²
ReadWrite/ReadOrWritePostAndCommentsWithTxParallel/write_rate=90 953.2µ ± ∞ ¹ 922.2µ ± ∞ ¹ ~ (p=0.100 n=3) ² 781.3µ ± ∞ ¹ ~ (p=0.100 n=3) ² 1128.8µ ± ∞ ¹ ~ (p=0.100 n=3) ² 884.5µ ± ∞ ¹ ~ (p=0.100 n=3) ² 1158.8µ ± ∞ ¹ ~ (p=0.100 n=3) ² 774.3µ ± ∞ ¹ ~ (p=0.100 n=3) ² 1060.5µ ± ∞ ¹ ~ (p=0.100 n=3) ²
ReadWrite/ReadOrWritePostAndCommentsWithTxParallel/write_rate=90-2 856.6µ ± ∞ ¹ 776.1µ ± ∞ ¹ ~ (p=0.200 n=3) ² 764.1µ ± ∞ ¹ ~ (p=0.100 n=3) ² 1143.9µ ± ∞ ¹ ~ (p=0.100 n=3) ² 891.3µ ± ∞ ¹ ~ (p=0.400 n=3) ² 1222.7µ ± ∞ ¹ ~ (p=0.100 n=3) ² 752.4µ ± ∞ ¹ ~ (p=0.100 n=3) ² 1005.9µ ± ∞ ¹ ~ (p=0.100 n=3) ²
ReadWrite/ReadOrWritePostAndCommentsWithTxParallel/write_rate=90-4 774.0µ ± ∞ ¹ 788.0µ ± ∞ ¹ ~ (p=0.200 n=3) ² 760.0µ ± ∞ ¹ ~ (p=1.000 n=3) ² 1205.0µ ± ∞ ¹ ~ (p=0.100 n=3) ² 956.6µ ± ∞ ¹ ~ (p=0.100 n=3) ² 1246.6µ ± ∞ ¹ ~ (p=0.100 n=3) ² 791.6µ ± ∞ ¹ ~ (p=0.200 n=3) ² 1067.9µ ± ∞ ¹ ~ (p=0.100 n=3) ²
ReadWrite/ReadOrWritePostAndCommentsWithTxParallel/write_rate=90-8 816.5µ ± ∞ ¹ 829.4µ ± ∞ ¹ ~ (p=0.400 n=3) ² 810.5µ ± ∞ ¹ ~ (p=1.000 n=3) ² 1264.1µ ± ∞ ¹ ~ (p=0.100 n=3) ² 1040.9µ ± ∞ ¹ ~ (p=0.100 n=3) ² 1580.8µ ± ∞ ¹ ~ (p=0.100 n=3) ² 846.2µ ± ∞ ¹ ~ (p=0.200 n=3) ² 1120.5µ ± ∞ ¹ ~ (p=0.100 n=3) ²
ReadWrite/ReadOrWritePostAndCommentsWithTxParallel/write_rate=90-16 1012.0µ ± ∞ ¹ 1038.0µ ± ∞ ¹ ~ (p=0.100 n=3) ² 878.4µ ± ∞ ¹ ~ (p=0.100 n=3) ² 10245.0µ ± ∞ ¹ ~ (p=0.100 n=3) ² 1545.3µ ± ∞ ¹ ~ (p=0.100 n=3) ² 12614.1µ ± ∞ ¹ ~ (p=0.100 n=3) ² 1029.6µ ± ∞ ¹ ~ (p=0.700 n=3) ² 10587.5µ ± ∞ ¹ ~ (p=0.100 n=3) ²
geomean 612.0µ 586.0µ -4.24% 521.4µ -14.80% 828.6µ +35.40% 613.5µ +0.25% 897.3µ +46.62% 513.1µ -16.16% 804.9µ +31.53%
¹ need >= 6 samples for confidence interval at level 0.95
² need >= 4 samples to detect a difference at alpha level 0.05
│ bench_ncruces_direct.txt │ bench_ncruces_driver.txt │ bench_eatonphil_direct.txt │ bench_glebarez_driver.txt │ bench_mattn_driver.txt │ bench_modernc_driver.txt │ bench_tailscale_driver.txt │ bench_zombiezen_direct.txt │
│ B/op │ B/op vs base │ B/op vs base │ B/op vs base │ B/op vs base │ B/op vs base │ B/op vs base │ B/op vs base │
ReadWrite/ReadOrWritePostAndCommentsParallel/write_rate=10 224.2Ki ± ∞ ¹ 235.4Ki ± ∞ ¹ ~ (p=0.100 n=3) ² 225.1Ki ± ∞ ¹ ~ (p=0.400 n=3) ² 456.1Ki ± ∞ ¹ ~ (p=0.100 n=3) ² 268.4Ki ± ∞ ¹ ~ (p=0.100 n=3) ² 453.7Ki ± ∞ ¹ ~ (p=0.100 n=3) ² 232.9Ki ± ∞ ¹ ~ (p=0.100 n=3) ² 226.0Ki ± ∞ ¹ ~ (p=0.200 n=3) ²
ReadWrite/ReadOrWritePostAndCommentsParallel/write_rate=10-2 224.9Ki ± ∞ ¹ 235.2Ki ± ∞ ¹ ~ (p=0.100 n=3) ² 225.6Ki ± ∞ ¹ ~ (p=0.400 n=3) ² 456.9Ki ± ∞ ¹ ~ (p=0.100 n=3) ² 268.3Ki ± ∞ ¹ ~ (p=0.100 n=3) ² 452.8Ki ± ∞ ¹ ~ (p=0.100 n=3) ² 233.6Ki ± ∞ ¹ ~ (p=0.100 n=3) ² 224.1Ki ± ∞ ¹ ~ (p=1.000 n=3) ²
ReadWrite/ReadOrWritePostAndCommentsParallel/write_rate=10-4 225.6Ki ± ∞ ¹ 235.6Ki ± ∞ ¹ ~ (p=0.100 n=3) ² 225.1Ki ± ∞ ¹ ~ (p=0.400 n=3) ² 454.5Ki ± ∞ ¹ ~ (p=0.100 n=3) ² 268.3Ki ± ∞ ¹ ~ (p=0.100 n=3) ² 453.9Ki ± ∞ ¹ ~ (p=0.100 n=3) ² 232.1Ki ± ∞ ¹ ~ (p=0.100 n=3) ² 225.4Ki ± ∞ ¹ ~ (p=0.700 n=3) ²
ReadWrite/ReadOrWritePostAndCommentsParallel/write_rate=10-8 224.4Ki ± ∞ ¹ 236.6Ki ± ∞ ¹ ~ (p=0.100 n=3) ² 225.1Ki ± ∞ ¹ ~ (p=0.700 n=3) ² 455.2Ki ± ∞ ¹ ~ (p=0.100 n=3) ² 268.5Ki ± ∞ ¹ ~ (p=0.100 n=3) ² 452.7Ki ± ∞ ¹ ~ (p=0.100 n=3) ² 233.1Ki ± ∞ ¹ ~ (p=0.100 n=3) ² 227.4Ki ± ∞ ¹ ~ (p=0.200 n=3) ²
ReadWrite/ReadOrWritePostAndCommentsParallel/write_rate=10-16 226.0Ki ± ∞ ¹ 235.7Ki ± ∞ ¹ ~ (p=0.100 n=3) ² 224.0Ki ± ∞ ¹ ~ (p=0.700 n=3) ² 453.5Ki ± ∞ ¹ ~ (p=0.100 n=3) ² 268.3Ki ± ∞ ¹ ~ (p=0.100 n=3) ² 451.4Ki ± ∞ ¹ ~ (p=0.100 n=3) ² 232.1Ki ± ∞ ¹ ~ (p=0.100 n=3) ² 226.0Ki ± ∞ ¹ ~ (p=1.000 n=3) ²
ReadWrite/ReadOrWritePostAndCommentsParallel/write_rate=90 26.59Ki ± ∞ ¹ 38.10Ki ± ∞ ¹ ~ (p=0.100 n=3) ² 27.00Ki ± ∞ ¹ ~ (p=0.700 n=3) ² 75.05Ki ± ∞ ¹ ~ (p=0.100 n=3) ² 303.58Ki ± ∞ ¹ ~ (p=0.100 n=3) ² 73.14Ki ± ∞ ¹ ~ (p=0.100 n=3) ² 37.50Ki ± ∞ ¹ ~ (p=0.100 n=3) ² 25.94Ki ± ∞ ¹ ~ (p=1.000 n=3) ²
ReadWrite/ReadOrWritePostAndCommentsParallel/write_rate=90-2 21.52Ki ± ∞ ¹ 40.47Ki ± ∞ ¹ ~ (p=0.100 n=3) ² 28.25Ki ± ∞ ¹ ~ (p=0.200 n=3) ² 69.50Ki ± ∞ ¹ ~ (p=0.100 n=3) ² 304.60Ki ± ∞ ¹ ~ (p=0.100 n=3) ² 73.55Ki ± ∞ ¹ ~ (p=0.100 n=3) ² 37.43Ki ± ∞ ¹ ~ (p=0.100 n=3) ² 26.62Ki ± ∞ ¹ ~ (p=0.100 n=3) ²
ReadWrite/ReadOrWritePostAndCommentsParallel/write_rate=90-4 27.12Ki ± ∞ ¹ 34.83Ki ± ∞ ¹ ~ (p=0.100 n=3) ² 29.78Ki ± ∞ ¹ ~ (p=1.000 n=3) ² 76.10Ki ± ∞ ¹ ~ (p=0.100 n=3) ² 304.21Ki ± ∞ ¹ ~ (p=0.100 n=3) ² 65.67Ki ± ∞ ¹ ~ (p=0.100 n=3) ² 39.57Ki ± ∞ ¹ ~ (p=0.100 n=3) ² 30.30Ki ± ∞ ¹ ~ (p=1.000 n=3) ²
ReadWrite/ReadOrWritePostAndCommentsParallel/write_rate=90-8 28.49Ki ± ∞ ¹ 35.98Ki ± ∞ ¹ ~ (p=0.100 n=3) ² 27.95Ki ± ∞ ¹ ~ (p=1.000 n=3) ² 71.65Ki ± ∞ ¹ ~ (p=0.100 n=3) ² 304.48Ki ± ∞ ¹ ~ (p=0.100 n=3) ² 80.66Ki ± ∞ ¹ ~ (p=0.100 n=3) ² 40.79Ki ± ∞ ¹ ~ (p=0.100 n=3) ² 25.03Ki ± ∞ ¹ ~ (p=0.700 n=3) ²
ReadWrite/ReadOrWritePostAndCommentsParallel/write_rate=90-16 23.55Ki ± ∞ ¹ 40.41Ki ± ∞ ¹ ~ (p=0.100 n=3) ² 29.02Ki ± ∞ ¹ ~ (p=0.200 n=3) ² 75.05Ki ± ∞ ¹ ~ (p=0.100 n=3) ² 304.03Ki ± ∞ ¹ ~ (p=0.100 n=3) ² 71.53Ki ± ∞ ¹ ~ (p=0.100 n=3) ² 40.43Ki ± ∞ ¹ ~ (p=0.100 n=3) ² 23.36Ki ± ∞ ¹ ~ (p=0.700 n=3) ²
ReadWrite/ReadOrWritePostAndCommentsWithTxParallel/write_rate=10 224.4Ki ± ∞ ¹ 237.7Ki ± ∞ ¹ ~ (p=0.100 n=3) ² 225.5Ki ± ∞ ¹ ~ (p=1.000 n=3) ² 459.3Ki ± ∞ ¹ ~ (p=0.100 n=3) ² 271.0Ki ± ∞ ¹ ~ (p=0.100 n=3) ² 456.1Ki ± ∞ ¹ ~ (p=0.100 n=3) ² 235.8Ki ± ∞ ¹ ~ (p=0.100 n=3) ² 226.0Ki ± ∞ ¹ ~ (p=0.700 n=3) ²
ReadWrite/ReadOrWritePostAndCommentsWithTxParallel/write_rate=10-2 224.5Ki ± ∞ ¹ 238.3Ki ± ∞ ¹ ~ (p=0.100 n=3) ² 225.2Ki ± ∞ ¹ ~ (p=0.400 n=3) ² 454.1Ki ± ∞ ¹ ~ (p=0.100 n=3) ² 271.2Ki ± ∞ ¹ ~ (p=0.100 n=3) ² 456.2Ki ± ∞ ¹ ~ (p=0.100 n=3) ² 234.8Ki ± ∞ ¹ ~ (p=0.100 n=3) ² 226.8Ki ± ∞ ¹ ~ (p=0.100 n=3) ²
ReadWrite/ReadOrWritePostAndCommentsWithTxParallel/write_rate=10-4 225.5Ki ± ∞ ¹ 237.8Ki ± ∞ ¹ ~ (p=0.100 n=3) ² 225.9Ki ± ∞ ¹ ~ (p=0.100 n=3) ² 456.6Ki ± ∞ ¹ ~ (p=0.100 n=3) ² 271.1Ki ± ∞ ¹ ~ (p=0.100 n=3) ² 455.8Ki ± ∞ ¹ ~ (p=0.100 n=3) ² 235.6Ki ± ∞ ¹ ~ (p=0.100 n=3) ² 226.2Ki ± ∞ ¹ ~ (p=0.700 n=3) ²
ReadWrite/ReadOrWritePostAndCommentsWithTxParallel/write_rate=10-8 224.5Ki ± ∞ ¹ 239.3Ki ± ∞ ¹ ~ (p=0.100 n=3) ² 225.7Ki ± ∞ ¹ ~ (p=0.100 n=3) ² 457.3Ki ± ∞ ¹ ~ (p=0.100 n=3) ² 271.2Ki ± ∞ ¹ ~ (p=0.100 n=3) ² 455.3Ki ± ∞ ¹ ~ (p=0.100 n=3) ² 235.1Ki ± ∞ ¹ ~ (p=0.100 n=3) ² 224.9Ki ± ∞ ¹ ~ (p=0.700 n=3) ²
ReadWrite/ReadOrWritePostAndCommentsWithTxParallel/write_rate=10-16 226.0Ki ± ∞ ¹ 238.5Ki ± ∞ ¹ ~ (p=0.100 n=3) ² 225.7Ki ± ∞ ¹ ~ (p=0.700 n=3) ² 453.3Ki ± ∞ ¹ ~ (p=0.100 n=3) ² 271.3Ki ± ∞ ¹ ~ (p=0.100 n=3) ² 457.8Ki ± ∞ ¹ ~ (p=0.100 n=3) ² 234.9Ki ± ∞ ¹ ~ (p=0.100 n=3) ² 225.2Ki ± ∞ ¹ ~ (p=0.700 n=3) ²
ReadWrite/ReadOrWritePostAndCommentsWithTxParallel/write_rate=90 25.66Ki ± ∞ ¹ 50.97Ki ± ∞ ¹ ~ (p=0.100 n=3) ² 29.67Ki ± ∞ ¹ ~ (p=1.000 n=3) ² 89.82Ki ± ∞ ¹ ~ (p=0.100 n=3) ² 316.24Ki ± ∞ ¹ ~ (p=0.100 n=3) ² 81.22Ki ± ∞ ¹ ~ (p=0.100 n=3) ² 48.96Ki ± ∞ ¹ ~ (p=0.100 n=3) ² 27.44Ki ± ∞ ¹ ~ (p=1.000 n=3) ²
ReadWrite/ReadOrWritePostAndCommentsWithTxParallel/write_rate=90-2 23.68Ki ± ∞ ¹ 50.13Ki ± ∞ ¹ ~ (p=0.100 n=3) ² 27.57Ki ± ∞ ¹ ~ (p=0.100 n=3) ² 88.58Ki ± ∞ ¹ ~ (p=0.100 n=3) ² 315.94Ki ± ∞ ¹ ~ (p=0.100 n=3) ² 91.25Ki ± ∞ ¹ ~ (p=0.100 n=3) ² 50.02Ki ± ∞ ¹ ~ (p=0.100 n=3) ² 25.19Ki ± ∞ ¹ ~ (p=0.100 n=3) ²
ReadWrite/ReadOrWritePostAndCommentsWithTxParallel/write_rate=90-4 25.52Ki ± ∞ ¹ 49.48Ki ± ∞ ¹ ~ (p=0.100 n=3) ² 27.83Ki ± ∞ ¹ ~ (p=0.400 n=3) ² 84.91Ki ± ∞ ¹ ~ (p=0.100 n=3) ² 316.09Ki ± ∞ ¹ ~ (p=0.100 n=3) ² 84.41Ki ± ∞ ¹ ~ (p=0.100 n=3) ² 48.91Ki ± ∞ ¹ ~ (p=0.100 n=3) ² 28.50Ki ± ∞ ¹ ~ (p=0.200 n=3) ²
ReadWrite/ReadOrWritePostAndCommentsWithTxParallel/write_rate=90-8 24.92Ki ± ∞ ¹ 50.41Ki ± ∞ ¹ ~ (p=0.100 n=3) ² 25.55Ki ± ∞ ¹ ~ (p=0.700 n=3) ² 88.20Ki ± ∞ ¹ ~ (p=0.100 n=3) ² 315.39Ki ± ∞ ¹ ~ (p=0.100 n=3) ² 80.43Ki ± ∞ ¹ ~ (p=0.100 n=3) ² 51.00Ki ± ∞ ¹ ~ (p=0.100 n=3) ² 27.36Ki ± ∞ ¹ ~ (p=0.700 n=3) ²
ReadWrite/ReadOrWritePostAndCommentsWithTxParallel/write_rate=90-16 24.14Ki ± ∞ ¹ 49.54Ki ± ∞ ¹ ~ (p=0.100 n=3) ² 26.50Ki ± ∞ ¹ ~ (p=0.200 n=3) ² 86.19Ki ± ∞ ¹ ~ (p=0.100 n=3) ² 316.02Ki ± ∞ ¹ ~ (p=0.100 n=3) ² 81.17Ki ± ∞ ¹ ~ (p=0.100 n=3) ² 51.52Ki ± ∞ ¹ ~ (p=0.100 n=3) ² 23.50Ki ± ∞ ¹ ~ (p=1.000 n=3) ²
geomean 75.07Ki 101.6Ki +35.38% 79.26Ki +5.58% 191.1Ki +154.61% 289.2Ki +285.24% 188.3Ki +150.83% 101.8Ki +35.57% 76.98Ki +2.55%
¹ need >= 6 samples for confidence interval at level 0.95
² need >= 4 samples to detect a difference at alpha level 0.05
│ bench_ncruces_direct.txt │ bench_ncruces_driver.txt │ bench_eatonphil_direct.txt │ bench_glebarez_driver.txt │ bench_mattn_driver.txt │ bench_modernc_driver.txt │ bench_tailscale_driver.txt │ bench_zombiezen_direct.txt │
│ allocs/op │ allocs/op vs base │ allocs/op vs base │ allocs/op vs base │ allocs/op vs base │ allocs/op vs base │ allocs/op vs base │ allocs/op vs base │
ReadWrite/ReadOrWritePostAndCommentsParallel/write_rate=10 234.0 ± ∞ ¹ 734.0 ± ∞ ¹ ~ (p=0.100 n=3) ² 257.0 ± ∞ ¹ ~ (p=0.100 n=3) ² 973.0 ± ∞ ¹ ~ (p=0.100 n=3) ² 695.0 ± ∞ ¹ ~ (p=0.100 n=3) ² 973.0 ± ∞ ¹ ~ (p=0.100 n=3) ² 544.0 ± ∞ ¹ ~ (p=0.100 n=3) ² 270.0 ± ∞ ¹ ~ (p=0.100 n=3) ²
ReadWrite/ReadOrWritePostAndCommentsParallel/write_rate=10-2 235.0 ± ∞ ¹ 734.0 ± ∞ ¹ ~ (p=0.100 n=3) ² 257.0 ± ∞ ¹ ~ (p=0.100 n=3) ² 974.0 ± ∞ ¹ ~ (p=0.100 n=3) ² 696.0 ± ∞ ¹ ~ (p=0.100 n=3) ² 974.0 ± ∞ ¹ ~ (p=0.100 n=3) ² 546.0 ± ∞ ¹ ~ (p=0.100 n=3) ² 270.0 ± ∞ ¹ ~ (p=0.100 n=3) ²
ReadWrite/ReadOrWritePostAndCommentsParallel/write_rate=10-4 236.0 ± ∞ ¹ 736.0 ± ∞ ¹ ~ (p=0.100 n=3) ² 257.0 ± ∞ ¹ ~ (p=0.100 n=3) ² 975.0 ± ∞ ¹ ~ (p=0.100 n=3) ² 696.0 ± ∞ ¹ ~ (p=0.100 n=3) ² 974.0 ± ∞ ¹ ~ (p=0.100 n=3) ² 546.0 ± ∞ ¹ ~ (p=0.100 n=3) ² 271.0 ± ∞ ¹ ~ (p=0.100 n=3) ²
ReadWrite/ReadOrWritePostAndCommentsParallel/write_rate=10-8 235.0 ± ∞ ¹ 737.0 ± ∞ ¹ ~ (p=0.100 n=3) ² 257.0 ± ∞ ¹ ~ (p=0.100 n=3) ² 975.0 ± ∞ ¹ ~ (p=0.100 n=3) ² 697.0 ± ∞ ¹ ~ (p=0.100 n=3) ² 975.0 ± ∞ ¹ ~ (p=0.100 n=3) ² 546.0 ± ∞ ¹ ~ (p=0.100 n=3) ² 271.0 ± ∞ ¹ ~ (p=0.100 n=3) ²
ReadWrite/ReadOrWritePostAndCommentsParallel/write_rate=10-16 236.0 ± ∞ ¹ 736.0 ± ∞ ¹ ~ (p=0.100 n=3) ² 257.0 ± ∞ ¹ ~ (p=0.100 n=3) ² 975.0 ± ∞ ¹ ~ (p=0.100 n=3) ² 696.0 ± ∞ ¹ ~ (p=0.100 n=3) ² 975.0 ± ∞ ¹ ~ (p=0.100 n=3) ² 546.0 ± ∞ ¹ ~ (p=0.100 n=3) ² 271.0 ± ∞ ¹ ~ (p=0.100 n=3) ²
ReadWrite/ReadOrWritePostAndCommentsParallel/write_rate=90 27.00 ± ∞ ¹ 442.00 ± ∞ ¹ ~ (p=0.100 n=3) ² 208.00 ± ∞ ¹ ~ (p=0.100 n=3) ² 967.00 ± ∞ ¹ ~ (p=0.100 n=3) ² 799.00 ± ∞ ¹ ~ (p=0.100 n=3) ² 967.00 ± ∞ ¹ ~ (p=0.100 n=3) ² 421.00 ± ∞ ¹ ~ (p=0.100 n=3) ² 260.00 ± ∞ ¹ ~ (p=0.100 n=3) ²
ReadWrite/ReadOrWritePostAndCommentsParallel/write_rate=90-2 22.00 ± ∞ ¹ 446.00 ± ∞ ¹ ~ (p=0.100 n=3) ² 209.00 ± ∞ ¹ ~ (p=0.100 n=3) ² 967.00 ± ∞ ¹ ~ (p=0.100 n=3) ² 802.00 ± ∞ ¹ ~ (p=0.100 n=3) ² 967.00 ± ∞ ¹ ~ (p=0.100 n=3) ² 422.00 ± ∞ ¹ ~ (p=0.100 n=3) ² 260.00 ± ∞ ¹ ~ (p=0.100 n=3) ²
ReadWrite/ReadOrWritePostAndCommentsParallel/write_rate=90-4 28.00 ± ∞ ¹ 438.00 ± ∞ ¹ ~ (p=0.100 n=3) ² 209.00 ± ∞ ¹ ~ (p=0.100 n=3) ² 968.00 ± ∞ ¹ ~ (p=0.100 n=3) ² 801.00 ± ∞ ¹ ~ (p=0.100 n=3) ² 967.00 ± ∞ ¹ ~ (p=0.100 n=3) ² 423.00 ± ∞ ¹ ~ (p=0.100 n=3) ² 260.00 ± ∞ ¹ ~ (p=0.100 n=3) ²
ReadWrite/ReadOrWritePostAndCommentsParallel/write_rate=90-8 29.00 ± ∞ ¹ 439.00 ± ∞ ¹ ~ (p=0.100 n=3) ² 209.00 ± ∞ ¹ ~ (p=0.100 n=3) ² 968.00 ± ∞ ¹ ~ (p=0.100 n=3) ² 802.00 ± ∞ ¹ ~ (p=0.100 n=3) ² 968.00 ± ∞ ¹ ~ (p=0.100 n=3) ² 424.00 ± ∞ ¹ ~ (p=0.100 n=3) ² 260.00 ± ∞ ¹ ~ (p=0.100 n=3) ²
ReadWrite/ReadOrWritePostAndCommentsParallel/write_rate=90-16 24.00 ± ∞ ¹ 446.00 ± ∞ ¹ ~ (p=0.100 n=3) ² 209.00 ± ∞ ¹ ~ (p=0.100 n=3) ² 968.00 ± ∞ ¹ ~ (p=0.100 n=3) ² 801.00 ± ∞ ¹ ~ (p=0.100 n=3) ² 968.00 ± ∞ ¹ ~ (p=0.100 n=3) ² 424.00 ± ∞ ¹ ~ (p=0.100 n=3) ² 260.00 ± ∞ ¹ ~ (p=0.100 n=3) ²
ReadWrite/ReadOrWritePostAndCommentsWithTxParallel/write_rate=10 235.0 ± ∞ ¹ 777.0 ± ∞ ¹ ~ (p=0.100 n=3) ² 257.0 ± ∞ ¹ ~ (p=0.100 n=3) ² 1009.0 ± ∞ ¹ ~ (p=0.100 n=3) ² 745.0 ± ∞ ¹ ~ (p=0.100 n=3) ² 1014.0 ± ∞ ¹ ~ (p=0.100 n=3) ² 585.0 ± ∞ ¹ ~ (p=0.100 n=3) ² 273.0 ± ∞ ¹ ~ (p=0.100 n=3) ²
ReadWrite/ReadOrWritePostAndCommentsWithTxParallel/write_rate=10-2 235.0 ± ∞ ¹ 777.0 ± ∞ ¹ ~ (p=0.100 n=3) ² 257.0 ± ∞ ¹ ~ (p=0.100 n=3) ² 1011.0 ± ∞ ¹ ~ (p=0.100 n=3) ² 746.0 ± ∞ ¹ ~ (p=0.100 n=3) ² 1014.0 ± ∞ ¹ ~ (p=0.100 n=3) ² 585.0 ± ∞ ¹ ~ (p=0.100 n=3) ² 273.0 ± ∞ ¹ ~ (p=0.100 n=3) ²
ReadWrite/ReadOrWritePostAndCommentsWithTxParallel/write_rate=10-4 235.0 ± ∞ ¹ 777.0 ± ∞ ¹ ~ (p=0.100 n=3) ² 258.0 ± ∞ ¹ ~ (p=0.100 n=3) ² 1010.0 ± ∞ ¹ ~ (p=0.100 n=3) ² 745.0 ± ∞ ¹ ~ (p=0.100 n=3) ² 1014.0 ± ∞ ¹ ~ (p=0.100 n=3) ² 586.0 ± ∞ ¹ ~ (p=0.100 n=3) ² 273.0 ± ∞ ¹ ~ (p=0.100 n=3) ²
ReadWrite/ReadOrWritePostAndCommentsWithTxParallel/write_rate=10-8 235.0 ± ∞ ¹ 779.0 ± ∞ ¹ ~ (p=0.100 n=3) ² 258.0 ± ∞ ¹ ~ (p=0.100 n=3) ² 1011.0 ± ∞ ¹ ~ (p=0.100 n=3) ² 746.0 ± ∞ ¹ ~ (p=0.100 n=3) ² 1015.0 ± ∞ ¹ ~ (p=0.100 n=3) ² 586.0 ± ∞ ¹ ~ (p=0.100 n=3) ² 273.0 ± ∞ ¹ ~ (p=0.100 n=3) ²
ReadWrite/ReadOrWritePostAndCommentsWithTxParallel/write_rate=10-16 236.0 ± ∞ ¹ 778.0 ± ∞ ¹ ~ (p=0.100 n=3) ² 258.0 ± ∞ ¹ ~ (p=0.100 n=3) ² 1012.0 ± ∞ ¹ ~ (p=0.100 n=3) ² 747.0 ± ∞ ¹ ~ (p=0.100 n=3) ² 1014.0 ± ∞ ¹ ~ (p=0.100 n=3) ² 586.0 ± ∞ ¹ ~ (p=0.100 n=3) ² 273.0 ± ∞ ¹ ~ (p=0.100 n=3) ²
ReadWrite/ReadOrWritePostAndCommentsWithTxParallel/write_rate=90 26.00 ± ∞ ¹ 597.00 ± ∞ ¹ ~ (p=0.100 n=3) ² 209.00 ± ∞ ¹ ~ (p=0.100 n=3) ² 1119.00 ± ∞ ¹ ~ (p=0.100 n=3) ² 968.00 ± ∞ ¹ ~ (p=0.100 n=3) ² 1125.00 ± ∞ ¹ ~ (p=0.100 n=3) ² 577.00 ± ∞ ¹ ~ (p=0.100 n=3) ² 263.00 ± ∞ ¹ ~ (p=0.100 n=3) ²
ReadWrite/ReadOrWritePostAndCommentsWithTxParallel/write_rate=90-2 24.00 ± ∞ ¹ 596.00 ± ∞ ¹ ~ (p=0.100 n=3) ² 209.00 ± ∞ ¹ ~ (p=0.100 n=3) ² 1119.00 ± ∞ ¹ ~ (p=0.100 n=3) ² 967.00 ± ∞ ¹ ~ (p=0.100 n=3) ² 1122.00 ± ∞ ¹ ~ (p=0.100 n=3) ² 577.00 ± ∞ ¹ ~ (p=0.100 n=3) ² 263.00 ± ∞ ¹ ~ (p=0.100 n=3) ²
ReadWrite/ReadOrWritePostAndCommentsWithTxParallel/write_rate=90-4 26.00 ± ∞ ¹ 596.00 ± ∞ ¹ ~ (p=0.100 n=3) ² 209.00 ± ∞ ¹ ~ (p=0.100 n=3) ² 1120.00 ± ∞ ¹ ~ (p=0.100 n=3) ² 968.00 ± ∞ ¹ ~ (p=0.100 n=3) ² 1124.00 ± ∞ ¹ ~ (p=0.100 n=3) ² 577.00 ± ∞ ¹ ~ (p=0.100 n=3) ² 263.00 ± ∞ ¹ ~ (p=0.100 n=3) ²
ReadWrite/ReadOrWritePostAndCommentsWithTxParallel/write_rate=90-8 26.00 ± ∞ ¹ 597.00 ± ∞ ¹ ~ (p=0.100 n=3) ² 208.00 ± ∞ ¹ ~ (p=0.100 n=3) ² 1119.00 ± ∞ ¹ ~ (p=0.100 n=3) ² 964.00 ± ∞ ¹ ~ (p=0.100 n=3) ² 1126.00 ± ∞ ¹ ~ (p=0.100 n=3) ² 577.00 ± ∞ ¹ ~ (p=0.100 n=3) ² 263.00 ± ∞ ¹ ~ (p=0.100 n=3) ²
ReadWrite/ReadOrWritePostAndCommentsWithTxParallel/write_rate=90-16 25.00 ± ∞ ¹ 596.00 ± ∞ ¹ ~ (p=0.100 n=3) ² 208.00 ± ∞ ¹ ~ (p=0.100 n=3) ² 1121.00 ± ∞ ¹ ~ (p=0.100 n=3) ² 967.00 ± ∞ ¹ ~ (p=0.100 n=3) ² 1126.00 ± ∞ ¹ ~ (p=0.100 n=3) ² 577.00 ± ∞ ¹ ~ (p=0.100 n=3) ² 264.00 ± ∞ ¹ ~ (p=0.100 n=3) ²
geomean 77.63 623.2 +702.71% 231.7 +198.49% 1.016k +1209.09% 796.3 +925.65% 1.018k +1211.58% 528.4 +580.61% 266.6 +243.47%
¹ need >= 6 samples for confidence interval at level 0.95
² need >= 4 samples to detect a difference at alpha level 0.05
SELECT
id,
title,
(SELECT COUNT(*) FROM comments WHERE post_id = posts.id) as comment_count,
(SELECT AVG(LENGTH(content)) FROM comments WHERE post_id = posts.id) AS avg_comment_length,
(SELECT MAX(LENGTH(content)) FROM comments WHERE post_id = posts.id) AS max_comment_length
FROM postsbenchstat bench_ncruces_direct.txt bench_ncruces_driver.txt bench_eatonphil_direct.txt bench_glebarez_driver.txt bench_mattn_driver.txt bench_modernc_driver.txt bench_tailscale_driver.txt bench_zombiezen_direct.txt
goos: darwin
goarch: arm64
pkg: github.com/michaellenaghan/go-sqlite-bench
cpu: Apple M2 Pro
│ bench_ncruces_direct.txt │ bench_ncruces_driver.txt │ bench_eatonphil_direct.txt │ bench_glebarez_driver.txt │ bench_mattn_driver.txt │ bench_modernc_driver.txt │ bench_tailscale_driver.txt │ bench_zombiezen_direct.txt │
│ sec/op │ sec/op vs base │ sec/op vs base │ sec/op vs base │ sec/op vs base │ sec/op vs base │ sec/op vs base │ sec/op vs base │
Query/Correlated-12 635.3m ± ∞ ¹ 646.0m ± ∞ ¹ ~ (p=0.100 n=3) ² 288.5m ± ∞ ¹ ~ (p=0.100 n=3) ² 304.6m ± ∞ ¹ ~ (p=0.100 n=3) ² 176.5m ± ∞ ¹ ~ (p=0.100 n=3) ² 303.8m ± ∞ ¹ ~ (p=0.100 n=3) ² 173.6m ± ∞ ¹ ~ (p=0.100 n=3) ² 323.4m ± ∞ ¹ ~ (p=0.100 n=3) ²
Query/CorrelatedParallel-12 120.94m ± ∞ ¹ 120.86m ± ∞ ¹ ~ (p=1.000 n=3) ² 50.85m ± ∞ ¹ ~ (p=0.100 n=3) ² 123.25m ± ∞ ¹ ~ (p=0.400 n=3) ² 46.98m ± ∞ ¹ ~ (p=0.100 n=3) ² 120.70m ± ∞ ¹ ~ (p=1.000 n=3) ² 44.85m ± ∞ ¹ ~ (p=0.100 n=3) ² 186.00m ± ∞ ¹ ~ (p=0.100 n=3) ²
geomean 277.2m 279.4m +0.81% 121.1m -56.30% 193.8m -30.10% 91.05m -67.15% 191.5m -30.92% 88.24m -68.17% 245.3m -11.51%
¹ need >= 6 samples for confidence interval at level 0.95
² need >= 4 samples to detect a difference at alpha level 0.05
│ bench_ncruces_direct.txt │ bench_ncruces_driver.txt │ bench_eatonphil_direct.txt │ bench_glebarez_driver.txt │ bench_mattn_driver.txt │ bench_modernc_driver.txt │ bench_tailscale_driver.txt │ bench_zombiezen_direct.txt │
│ B/op │ B/op vs base │ B/op vs base │ B/op vs base │ B/op vs base │ B/op vs base │ B/op vs base │ B/op vs base │
Query/Correlated-12 57168.0 ± ∞ ¹ 71264.0 ± ∞ ¹ ~ (p=0.100 n=3) ² 732.0 ± ∞ ¹ ~ (p=0.100 n=3) ² 63072.0 ± ∞ ¹ ~ (p=0.100 n=3) ² 207197.0 ± ∞ ¹ ~ (p=0.100 n=3) ² 63160.0 ± ∞ ¹ ~ (p=0.100 n=3) ² 47021.0 ± ∞ ¹ ~ (p=0.100 n=3) ² 9280.0 ± ∞ ¹ ~ (p=0.100 n=3) ²
Query/CorrelatedParallel-12 335.0 ± ∞ ¹ 71391.0 ± ∞ ¹ ~ (p=0.100 n=3) ² 157.0 ± ∞ ¹ ~ (p=0.700 n=3) ² 64257.0 ± ∞ ¹ ~ (p=0.100 n=3) ² 207526.0 ± ∞ ¹ ~ (p=0.100 n=3) ² 63265.0 ± ∞ ¹ ~ (p=0.100 n=3) ² 47773.0 ± ∞ ¹ ~ (p=0.100 n=3) ² 1788.0 ± ∞ ¹ ~ (p=0.700 n=3) ²
geomean 4.274Ki 69.66Ki +1529.89% 339.0 -92.25% 62.17Ki +1354.72% 202.5Ki +4638.37% 61.73Ki +1344.45% 46.28Ki +983.02% 3.978Ki -6.92%
¹ need >= 6 samples for confidence interval at level 0.95
² need >= 4 samples to detect a difference at alpha level 0.05
│ bench_ncruces_direct.txt │ bench_ncruces_driver.txt │ bench_eatonphil_direct.txt │ bench_glebarez_driver.txt │ bench_mattn_driver.txt │ bench_modernc_driver.txt │ bench_tailscale_driver.txt │ bench_zombiezen_direct.txt │
│ allocs/op │ allocs/op vs base │ allocs/op vs base │ allocs/op vs base │ allocs/op vs base │ allocs/op vs base │ allocs/op vs base │ allocs/op vs base │
Query/Correlated-12 12.000 ± ∞ ¹ 5770.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 6.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 6769.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 6771.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 6770.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 3766.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 30.000 ± ∞ ¹ ~ (p=0.100 n=3) ²
Query/CorrelatedParallel-12 2.000 ± ∞ ¹ 5770.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 1.000 ± ∞ ¹ ~ (p=0.300 n=3) ² 6788.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 6773.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 6772.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 3768.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 15.000 ± ∞ ¹ ~ (p=0.100 n=3) ²
geomean 4.899 5.770k +117679.63% 2.449 -50.00% 6.778k +138265.42% 6.772k +138132.87% 6.771k +138112.46% 3.767k +76793.56% 21.21 +333.01%
¹ need >= 6 samples for confidence interval at level 0.95
² need >= 4 samples to detect a difference at alpha level 0.05
benchstat bench_ncruces_direct.txt bench_ncruces_driver.txt bench_eatonphil_direct.txt bench_glebarez_driver.txt bench_mattn_driver.txt bench_modernc_driver.txt bench_tailscale_driver.txt bench_zombiezen_direct.txt
goos: darwin
goarch: arm64
pkg: github.com/michaellenaghan/go-sqlite-bench
cpu: Apple M2 Pro
│ bench_ncruces_direct.txt │ bench_ncruces_driver.txt │ bench_eatonphil_direct.txt │ bench_glebarez_driver.txt │ bench_mattn_driver.txt │ bench_modernc_driver.txt │ bench_tailscale_driver.txt │ bench_zombiezen_direct.txt │
│ sec/op │ sec/op vs base │ sec/op vs base │ sec/op vs base │ sec/op vs base │ sec/op vs base │ sec/op vs base │ sec/op vs base │
Query/CorrelatedParallel 645.9m ± ∞ ¹ 647.5m ± ∞ ¹ ~ (p=1.000 n=3) ² 289.0m ± ∞ ¹ ~ (p=0.100 n=3) ² 302.7m ± ∞ ¹ ~ (p=0.100 n=3) ² 174.8m ± ∞ ¹ ~ (p=0.100 n=3) ² 302.6m ± ∞ ¹ ~ (p=0.100 n=3) ² 172.6m ± ∞ ¹ ~ (p=0.100 n=3) ² 327.1m ± ∞ ¹ ~ (p=0.100 n=3) ²
Query/CorrelatedParallel-2 443.68m ± ∞ ¹ 436.45m ± ∞ ¹ ~ (p=0.100 n=3) ² 170.37m ± ∞ ¹ ~ (p=0.100 n=3) ² 184.01m ± ∞ ¹ ~ (p=0.100 n=3) ² 96.78m ± ∞ ¹ ~ (p=0.100 n=3) ² 184.50m ± ∞ ¹ ~ (p=0.100 n=3) ² 95.84m ± ∞ ¹ ~ (p=0.100 n=3) ² 189.30m ± ∞ ¹ ~ (p=0.100 n=3) ²
Query/CorrelatedParallel-4 227.99m ± ∞ ¹ 225.66m ± ∞ ¹ ~ (p=0.200 n=3) ² 84.66m ± ∞ ¹ ~ (p=0.100 n=3) ² 117.50m ± ∞ ¹ ~ (p=0.100 n=3) ² 58.08m ± ∞ ¹ ~ (p=0.100 n=3) ² 120.16m ± ∞ ¹ ~ (p=0.100 n=3) ² 56.86m ± ∞ ¹ ~ (p=0.100 n=3) ² 123.51m ± ∞ ¹ ~ (p=0.100 n=3) ²
Query/CorrelatedParallel-8 153.55m ± ∞ ¹ 154.95m ± ∞ ¹ ~ (p=0.400 n=3) ² 53.84m ± ∞ ¹ ~ (p=0.100 n=3) ² 138.45m ± ∞ ¹ ~ (p=0.100 n=3) ² 43.37m ± ∞ ¹ ~ (p=0.100 n=3) ² 138.67m ± ∞ ¹ ~ (p=0.100 n=3) ² 42.05m ± ∞ ¹ ~ (p=0.100 n=3) ² 187.21m ± ∞ ¹ ~ (p=0.700 n=3) ²
Query/CorrelatedParallel-16 78.47m ± ∞ ¹ 78.53m ± ∞ ¹ ~ (p=1.000 n=3) ² 53.68m ± ∞ ¹ ~ (p=0.100 n=3) ² 123.65m ± ∞ ¹ ~ (p=0.100 n=3) ² 49.66m ± ∞ ¹ ~ (p=0.100 n=3) ² 117.56m ± ∞ ¹ ~ (p=0.100 n=3) ² 44.19m ± ∞ ¹ ~ (p=0.100 n=3) ² 184.86m ± ∞ ¹ ~ (p=0.100 n=3) ²
geomean 239.4m 238.8m -0.29% 103.8m -56.65% 162.1m -32.29% 73.31m -69.38% 161.4m -32.61% 70.55m -70.53% 192.5m -19.59%
¹ need >= 6 samples for confidence interval at level 0.95
² need >= 4 samples to detect a difference at alpha level 0.05
│ bench_ncruces_direct.txt │ bench_ncruces_driver.txt │ bench_eatonphil_direct.txt │ bench_glebarez_driver.txt │ bench_mattn_driver.txt │ bench_modernc_driver.txt │ bench_tailscale_driver.txt │ bench_zombiezen_direct.txt │
│ B/op │ B/op vs base │ B/op vs base │ B/op vs base │ B/op vs base │ B/op vs base │ B/op vs base │ B/op vs base │
Query/CorrelatedParallel 57140.0 ± ∞ ¹ 71132.0 ± ∞ ¹ ~ (p=0.100 n=3) ² 718.0 ± ∞ ¹ ~ (p=0.100 n=3) ² 63050.0 ± ∞ ¹ ~ (p=0.100 n=3) ² 207185.0 ± ∞ ¹ ~ (p=0.100 n=3) ² 63066.0 ± ∞ ¹ ~ (p=0.100 n=3) ² 47009.0 ± ∞ ¹ ~ (p=0.100 n=3) ² 2178.0 ± ∞ ¹ ~ (p=0.100 n=3) ²
Query/CorrelatedParallel-2 57157.00 ± ∞ ¹ 71082.00 ± ∞ ¹ ~ (p=0.100 n=3) ² 49.00 ± ∞ ¹ ~ (p=0.100 n=3) ² 63013.00 ± ∞ ¹ ~ (p=0.100 n=3) ² 207154.00 ± ∞ ¹ ~ (p=0.100 n=3) ² 63029.00 ± ∞ ¹ ~ (p=0.100 n=3) ² 47002.00 ± ∞ ¹ ~ (p=0.100 n=3) ² 492.00 ± ∞ ¹ ~ (p=0.100 n=3) ²
Query/CorrelatedParallel-4 11745.00 ± ∞ ¹ 71028.00 ± ∞ ¹ ~ (p=0.100 n=3) ² 48.00 ± ∞ ¹ ~ (p=0.100 n=3) ² 63078.00 ± ∞ ¹ ~ (p=0.100 n=3) ² 207175.00 ± ∞ ¹ ~ (p=0.100 n=3) ² 63080.00 ± ∞ ¹ ~ (p=0.100 n=3) ² 46991.00 ± ∞ ¹ ~ (p=0.100 n=3) ² 464.00 ± ∞ ¹ ~ (p=0.700 n=3) ²
Query/CorrelatedParallel-8 388.0 ± ∞ ¹ 71304.0 ± ∞ ¹ ~ (p=0.100 n=3) ² 208.0 ± ∞ ¹ ~ (p=0.400 n=3) ² 63190.0 ± ∞ ¹ ~ (p=0.100 n=3) ² 207386.0 ± ∞ ¹ ~ (p=0.100 n=3) ² 63250.0 ± ∞ ¹ ~ (p=0.100 n=3) ² 47204.0 ± ∞ ¹ ~ (p=0.100 n=3) ² 858.0 ± ∞ ¹ ~ (p=0.100 n=3) ²
Query/CorrelatedParallel-16 118.0 ± ∞ ¹ 72441.0 ± ∞ ¹ ~ (p=0.100 n=3) ² 625.0 ± ∞ ¹ ~ (p=0.300 n=3) ² 63888.0 ± ∞ ¹ ~ (p=0.100 n=3) ² 207687.0 ± ∞ ¹ ~ (p=0.100 n=3) ² 63420.0 ± ∞ ¹ ~ (p=0.100 n=3) ² 47213.0 ± ∞ ¹ ~ (p=0.100 n=3) ² 1782.0 ± ∞ ¹ ~ (p=0.200 n=3) ²
geomean 4.351Ki 69.72Ki +1502.34% 185.5 -95.84% 61.76Ki +1319.37% 202.5Ki +4552.86% 61.69Ki +1317.71% 45.98Ki +956.71% 946.6 -78.75%
¹ need >= 6 samples for confidence interval at level 0.95
² need >= 4 samples to detect a difference at alpha level 0.05
│ bench_ncruces_direct.txt │ bench_ncruces_driver.txt │ bench_eatonphil_direct.txt │ bench_glebarez_driver.txt │ bench_mattn_driver.txt │ bench_modernc_driver.txt │ bench_tailscale_driver.txt │ bench_zombiezen_direct.txt │
│ allocs/op │ allocs/op vs base │ allocs/op vs base │ allocs/op vs base │ allocs/op vs base │ allocs/op vs base │ allocs/op vs base │ allocs/op vs base │
Query/CorrelatedParallel 14.000 ± ∞ ¹ 5771.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 7.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 6770.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 6772.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 6770.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 3767.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 30.000 ± ∞ ¹ ~ (p=0.100 n=3) ²
Query/CorrelatedParallel-2 14.000 ± ∞ ¹ 5770.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 1.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 6769.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 6771.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 6769.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 3767.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 8.000 ± ∞ ¹ ~ (p=0.700 n=3) ²
Query/CorrelatedParallel-4 4.000 ± ∞ ¹ 5769.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 1.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 6770.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 6772.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 6770.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 3767.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 7.000 ± ∞ ¹ ~ (p=0.100 n=3) ²
Query/CorrelatedParallel-8 3.000 ± ∞ ¹ 5770.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 1.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 6771.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 6772.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 6772.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 3767.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 10.000 ± ∞ ¹ ~ (p=0.100 n=3) ²
Query/CorrelatedParallel-16 2.000 ± ∞ ¹ 5773.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 2.000 ± ∞ ¹ ~ (p=1.000 n=3) ² 6782.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 6774.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 6774.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 3769.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 15.000 ± ∞ ¹ ~ (p=0.100 n=3) ²
geomean 5.426 5.771k +106247.55% 1.695 -68.76% 6.772k +124709.90% 6.772k +124706.24% 6.771k +124684.12% 3.767k +69330.18% 12.03 +121.71%
¹ need >= 6 samples for confidence interval at level 0.95
² need >= 4 samples to detect a difference at alpha level 0.05
SELECT
strftime('%Y-%m', created) AS month,
COUNT(*) as month_total
FROM posts
GROUP BY month
ORDER BY monthbenchstat bench_ncruces_direct.txt bench_ncruces_driver.txt bench_eatonphil_direct.txt bench_glebarez_driver.txt bench_mattn_driver.txt bench_modernc_driver.txt bench_tailscale_driver.txt bench_zombiezen_direct.txt
goos: darwin
goarch: arm64
pkg: github.com/michaellenaghan/go-sqlite-bench
cpu: Apple M2 Pro
│ bench_ncruces_direct.txt │ bench_ncruces_driver.txt │ bench_eatonphil_direct.txt │ bench_glebarez_driver.txt │ bench_mattn_driver.txt │ bench_modernc_driver.txt │ bench_tailscale_driver.txt │ bench_zombiezen_direct.txt │
│ sec/op │ sec/op vs base │ sec/op vs base │ sec/op vs base │ sec/op vs base │ sec/op vs base │ sec/op vs base │ sec/op vs base │
Query/GroupBy-12 569.5µ ± ∞ ¹ 582.7µ ± ∞ ¹ ~ (p=0.100 n=3) ² 277.0µ ± ∞ ¹ ~ (p=0.100 n=3) ² 489.2µ ± ∞ ¹ ~ (p=0.100 n=3) ² 333.2µ ± ∞ ¹ ~ (p=0.100 n=3) ² 498.1µ ± ∞ ¹ ~ (p=0.100 n=3) ² 261.3µ ± ∞ ¹ ~ (p=0.100 n=3) ² 479.1µ ± ∞ ¹ ~ (p=0.100 n=3) ²
Query/GroupByParallel-12 62.25µ ± ∞ ¹ 66.33µ ± ∞ ¹ ~ (p=0.100 n=3) ² 38.59µ ± ∞ ¹ ~ (p=0.100 n=3) ² 687.89µ ± ∞ ¹ ~ (p=0.100 n=3) ² 404.86µ ± ∞ ¹ ~ (p=0.100 n=3) ² 691.06µ ± ∞ ¹ ~ (p=0.100 n=3) ² 28.07µ ± ∞ ¹ ~ (p=0.100 n=3) ² 689.54µ ± ∞ ¹ ~ (p=0.100 n=3) ²
geomean 188.3µ 196.6µ +4.42% 103.4µ -45.09% 580.1µ +208.10% 367.3µ +95.07% 586.7µ +211.61% 85.64µ -54.51% 574.8µ +205.28%
¹ need >= 6 samples for confidence interval at level 0.95
² need >= 4 samples to detect a difference at alpha level 0.05
│ bench_ncruces_direct.txt │ bench_ncruces_driver.txt │ bench_eatonphil_direct.txt │ bench_glebarez_driver.txt │ bench_mattn_driver.txt │ bench_modernc_driver.txt │ bench_tailscale_driver.txt │ bench_zombiezen_direct.txt │
│ B/op │ B/op vs base │ B/op vs base │ B/op vs base │ B/op vs base │ B/op vs base │ B/op vs base │ B/op vs base │
Query/GroupBy-12 0.0 ± ∞ ¹ 2340.0 ± ∞ ¹ ~ (p=0.100 n=3) ² 0.0 ± ∞ ¹ ~ (p=1.000 n=3) ² 1932.0 ± ∞ ¹ ~ (p=0.100 n=3) ² 6989.0 ± ∞ ¹ ~ (p=0.100 n=3) ² 1944.0 ± ∞ ¹ ~ (p=0.100 n=3) ² 1558.0 ± ∞ ¹ ~ (p=0.100 n=3) ² 361.0 ± ∞ ¹ ~ (p=0.600 n=3) ²
Query/GroupByParallel-12 0.0 ± ∞ ¹ 2328.0 ± ∞ ¹ ~ (p=0.100 n=3) ² 0.0 ± ∞ ¹ ~ (p=1.000 n=3) ³ 1935.0 ± ∞ ¹ ~ (p=0.100 n=3) ² 6995.0 ± ∞ ¹ ~ (p=0.100 n=3) ² 1948.0 ± ∞ ¹ ~ (p=0.100 n=3) ² 1503.0 ± ∞ ¹ ~ (p=0.100 n=3) ² 370.0 ± ∞ ¹ ~ (p=0.100 n=3) ²
geomean ⁴ 2.279Ki ? +0.00% ⁴ 1.888Ki ? 6.828Ki ? 1.900Ki ? 1.494Ki ? 365.5 ?
¹ need >= 6 samples for confidence interval at level 0.95
² need >= 4 samples to detect a difference at alpha level 0.05
³ all samples are equal
⁴ summaries must be >0 to compute geomean
│ bench_ncruces_direct.txt │ bench_ncruces_driver.txt │ bench_eatonphil_direct.txt │ bench_glebarez_driver.txt │ bench_mattn_driver.txt │ bench_modernc_driver.txt │ bench_tailscale_driver.txt │ bench_zombiezen_direct.txt │
│ allocs/op │ allocs/op vs base │ allocs/op vs base │ allocs/op vs base │ allocs/op vs base │ allocs/op vs base │ allocs/op vs base │ allocs/op vs base │
Query/GroupBy-12 0.000 ± ∞ ¹ 116.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 0.000 ± ∞ ¹ ~ (p=1.000 n=3) ³ 119.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 151.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 119.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 51.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 6.000 ± ∞ ¹ ~ (p=0.100 n=3) ²
Query/GroupByParallel-12 0.000 ± ∞ ¹ 115.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 0.000 ± ∞ ¹ ~ (p=1.000 n=3) ³ 119.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 151.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 119.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 50.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 6.000 ± ∞ ¹ ~ (p=0.100 n=3) ²
geomean ⁴ 115.5 ? +0.00% ⁴ 119.0 ? 151.0 ? 119.0 ? 50.50 ? 6.000 ?
¹ need >= 6 samples for confidence interval at level 0.95
² need >= 4 samples to detect a difference at alpha level 0.05
³ all samples are equal
⁴ summaries must be >0 to compute geomean
benchstat bench_ncruces_direct.txt bench_ncruces_driver.txt bench_eatonphil_direct.txt bench_glebarez_driver.txt bench_mattn_driver.txt bench_modernc_driver.txt bench_tailscale_driver.txt bench_zombiezen_direct.txt
goos: darwin
goarch: arm64
pkg: github.com/michaellenaghan/go-sqlite-bench
cpu: Apple M2 Pro
│ bench_ncruces_direct.txt │ bench_ncruces_driver.txt │ bench_eatonphil_direct.txt │ bench_glebarez_driver.txt │ bench_mattn_driver.txt │ bench_modernc_driver.txt │ bench_tailscale_driver.txt │ bench_zombiezen_direct.txt │
│ sec/op │ sec/op vs base │ sec/op vs base │ sec/op vs base │ sec/op vs base │ sec/op vs base │ sec/op vs base │ sec/op vs base │
Query/GroupByParallel 566.0µ ± ∞ ¹ 562.1µ ± ∞ ¹ ~ (p=0.200 n=3) ² 287.3µ ± ∞ ¹ ~ (p=0.100 n=3) ² 484.5µ ± ∞ ¹ ~ (p=0.100 n=3) ² 301.7µ ± ∞ ¹ ~ (p=0.100 n=3) ² 483.4µ ± ∞ ¹ ~ (p=0.100 n=3) ² 264.5µ ± ∞ ¹ ~ (p=0.100 n=3) ² 467.3µ ± ∞ ¹ ~ (p=0.100 n=3) ²
Query/GroupByParallel-2 285.1µ ± ∞ ¹ 287.5µ ± ∞ ¹ ~ (p=0.100 n=3) ² 146.0µ ± ∞ ¹ ~ (p=0.100 n=3) ² 384.0µ ± ∞ ¹ ~ (p=0.100 n=3) ² 370.9µ ± ∞ ¹ ~ (p=0.100 n=3) ² 384.9µ ± ∞ ¹ ~ (p=0.100 n=3) ² 130.6µ ± ∞ ¹ ~ (p=0.100 n=3) ² 378.6µ ± ∞ ¹ ~ (p=0.100 n=3) ²
Query/GroupByParallel-4 145.70µ ± ∞ ¹ 147.83µ ± ∞ ¹ ~ (p=0.100 n=3) ² 79.85µ ± ∞ ¹ ~ (p=0.100 n=3) ² 398.64µ ± ∞ ¹ ~ (p=0.100 n=3) ² 448.10µ ± ∞ ¹ ~ (p=0.100 n=3) ² 390.96µ ± ∞ ¹ ~ (p=0.100 n=3) ² 65.92µ ± ∞ ¹ ~ (p=0.100 n=3) ² 393.71µ ± ∞ ¹ ~ (p=0.100 n=3) ²
Query/GroupByParallel-8 73.51µ ± ∞ ¹ 75.90µ ± ∞ ¹ ~ (p=0.100 n=3) ² 42.98µ ± ∞ ¹ ~ (p=0.100 n=3) ² 651.60µ ± ∞ ¹ ~ (p=0.100 n=3) ² 453.48µ ± ∞ ¹ ~ (p=0.100 n=3) ² 656.66µ ± ∞ ¹ ~ (p=0.100 n=3) ² 34.31µ ± ∞ ¹ ~ (p=0.100 n=3) ² 651.81µ ± ∞ ¹ ~ (p=0.100 n=3) ²
Query/GroupByParallel-16 61.75µ ± ∞ ¹ 64.71µ ± ∞ ¹ ~ (p=0.100 n=3) ² 37.65µ ± ∞ ¹ ~ (p=0.100 n=3) ² 687.56µ ± ∞ ¹ ~ (p=0.100 n=3) ² 414.56µ ± ∞ ¹ ~ (p=0.100 n=3) ² 688.05µ ± ∞ ¹ ~ (p=0.100 n=3) ² 28.35µ ± ∞ ¹ ~ (p=0.100 n=3) ² 692.11µ ± ∞ ¹ ~ (p=0.100 n=3) ²
geomean 160.6µ 163.6µ +1.91% 88.47µ -44.90% 506.2µ +215.25% 393.4µ +145.04% 505.1µ +214.57% 73.98µ -53.92% 500.6µ +211.76%
¹ need >= 6 samples for confidence interval at level 0.95
² need >= 4 samples to detect a difference at alpha level 0.05
│ bench_ncruces_direct.txt │ bench_ncruces_driver.txt │ bench_eatonphil_direct.txt │ bench_glebarez_driver.txt │ bench_mattn_driver.txt │ bench_modernc_driver.txt │ bench_tailscale_driver.txt │ bench_zombiezen_direct.txt │
│ B/op │ B/op vs base │ B/op vs base │ B/op vs base │ B/op vs base │ B/op vs base │ B/op vs base │ B/op vs base │
Query/GroupByParallel 0.0 ± ∞ ¹ 2324.0 ± ∞ ¹ ~ (p=0.100 n=3) ² 0.0 ± ∞ ¹ ~ (p=1.000 n=3) ³ 1817.0 ± ∞ ¹ ~ (p=0.100 n=3) ² 6984.0 ± ∞ ¹ ~ (p=0.100 n=3) ² 1832.0 ± ∞ ¹ ~ (p=0.100 n=3) ² 1557.0 ± ∞ ¹ ~ (p=0.100 n=3) ² 362.0 ± ∞ ¹ ~ (p=0.100 n=3) ²
Query/GroupByParallel-2 0.0 ± ∞ ¹ 2324.0 ± ∞ ¹ ~ (p=0.100 n=3) ² 0.0 ± ∞ ¹ ~ (p=1.000 n=3) ³ 1928.0 ± ∞ ¹ ~ (p=0.100 n=3) ² 6979.0 ± ∞ ¹ ~ (p=0.100 n=3) ² 1874.0 ± ∞ ¹ ~ (p=0.100 n=3) ² 1557.0 ± ∞ ¹ ~ (p=0.100 n=3) ² 360.0 ± ∞ ¹ ~ (p=0.100 n=3) ²
Query/GroupByParallel-4 0.0 ± ∞ ¹ 2324.0 ± ∞ ¹ ~ (p=0.100 n=3) ² 0.0 ± ∞ ¹ ~ (p=1.000 n=3) ³ 1930.0 ± ∞ ¹ ~ (p=0.100 n=3) ² 6951.0 ± ∞ ¹ ~ (p=0.100 n=3) ² 1923.0 ± ∞ ¹ ~ (p=0.100 n=3) ² 1516.0 ± ∞ ¹ ~ (p=0.100 n=3) ² 361.0 ± ∞ ¹ ~ (p=0.100 n=3) ²
Query/GroupByParallel-8 0.0 ± ∞ ¹ 2326.0 ± ∞ ¹ ~ (p=0.100 n=3) ² 0.0 ± ∞ ¹ ~ (p=1.000 n=3) ³ 1932.0 ± ∞ ¹ ~ (p=0.100 n=3) ² 6982.0 ± ∞ ¹ ~ (p=0.100 n=3) ² 1948.0 ± ∞ ¹ ~ (p=0.100 n=3) ² 1547.0 ± ∞ ¹ ~ (p=0.100 n=3) ² 362.0 ± ∞ ¹ ~ (p=0.100 n=3) ²
Query/GroupByParallel-16 0.0 ± ∞ ¹ 2326.0 ± ∞ ¹ ~ (p=0.100 n=3) ² 0.0 ± ∞ ¹ ~ (p=1.000 n=3) ² 1938.0 ± ∞ ¹ ~ (p=0.100 n=3) ² 6988.0 ± ∞ ¹ ~ (p=0.100 n=3) ² 1953.0 ± ∞ ¹ ~ (p=0.100 n=3) ² 1490.0 ± ∞ ¹ ~ (p=0.100 n=3) ² 368.0 ± ∞ ¹ ~ (p=0.100 n=3) ²
geomean ⁴ 2.270Ki ? +0.00% ⁴ 1.864Ki ? 6.813Ki ? 1.861Ki ? 1.497Ki ? 362.6 ?
¹ need >= 6 samples for confidence interval at level 0.95
² need >= 4 samples to detect a difference at alpha level 0.05
³ all samples are equal
⁴ summaries must be >0 to compute geomean
│ bench_ncruces_direct.txt │ bench_ncruces_driver.txt │ bench_eatonphil_direct.txt │ bench_glebarez_driver.txt │ bench_mattn_driver.txt │ bench_modernc_driver.txt │ bench_tailscale_driver.txt │ bench_zombiezen_direct.txt │
│ allocs/op │ allocs/op vs base │ allocs/op vs base │ allocs/op vs base │ allocs/op vs base │ allocs/op vs base │ allocs/op vs base │ allocs/op vs base │
Query/GroupByParallel 0.000 ± ∞ ¹ 115.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 0.000 ± ∞ ¹ ~ (p=1.000 n=3) ³ 117.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 151.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 117.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 51.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 6.000 ± ∞ ¹ ~ (p=0.100 n=3) ²
Query/GroupByParallel-2 0.000 ± ∞ ¹ 115.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 0.000 ± ∞ ¹ ~ (p=1.000 n=3) ³ 119.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 150.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 118.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 51.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 6.000 ± ∞ ¹ ~ (p=0.100 n=3) ²
Query/GroupByParallel-4 0.000 ± ∞ ¹ 115.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 0.000 ± ∞ ¹ ~ (p=1.000 n=3) ³ 119.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 150.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 118.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 50.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 6.000 ± ∞ ¹ ~ (p=0.100 n=3) ²
Query/GroupByParallel-8 0.000 ± ∞ ¹ 115.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 0.000 ± ∞ ¹ ~ (p=1.000 n=3) ³ 119.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 150.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 119.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 50.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 6.000 ± ∞ ¹ ~ (p=0.100 n=3) ²
Query/GroupByParallel-16 0.000 ± ∞ ¹ 115.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 0.000 ± ∞ ¹ ~ (p=1.000 n=3) ³ 119.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 151.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 119.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 50.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 6.000 ± ∞ ¹ ~ (p=0.100 n=3) ²
geomean ⁴ 115.0 ? +0.00% ⁴ 118.6 ? 150.4 ? 118.2 ? 50.40 ? 6.000 ?
¹ need >= 6 samples for confidence interval at level 0.95
² need >= 4 samples to detect a difference at alpha level 0.05
³ all samples are equal
⁴ summaries must be >0 to compute geomean
SELECT
date(created) as day,
SUM(json_extract(stats, '$.lorem')) as sum_lorem,
AVG(json_extract(stats, '$.ipsum.dolor')) as avg_dolor,
MAX(json_extract(stats, '$.lorem.sit')) as max_sit
FROM posts
GROUP BY day
ORDER BY daybenchstat bench_ncruces_direct.txt bench_ncruces_driver.txt bench_eatonphil_direct.txt bench_glebarez_driver.txt bench_mattn_driver.txt bench_modernc_driver.txt bench_tailscale_driver.txt bench_zombiezen_direct.txt
goos: darwin
goarch: arm64
pkg: github.com/michaellenaghan/go-sqlite-bench
cpu: Apple M2 Pro
│ bench_ncruces_direct.txt │ bench_ncruces_driver.txt │ bench_eatonphil_direct.txt │ bench_glebarez_driver.txt │ bench_mattn_driver.txt │ bench_modernc_driver.txt │ bench_tailscale_driver.txt │ bench_zombiezen_direct.txt │
│ sec/op │ sec/op vs base │ sec/op vs base │ sec/op vs base │ sec/op vs base │ sec/op vs base │ sec/op vs base │ sec/op vs base │
Query/JSON-12 9.797m ± ∞ ¹ 9.787m ± ∞ ¹ ~ (p=1.000 n=3) ² 7.264m ± ∞ ¹ ~ (p=0.100 n=3) ² 9.312m ± ∞ ¹ ~ (p=0.100 n=3) ² 8.260m ± ∞ ¹ ~ (p=0.100 n=3) ² 9.375m ± ∞ ¹ ~ (p=0.100 n=3) ² 7.055m ± ∞ ¹ ~ (p=0.100 n=3) ² 10.277m ± ∞ ¹ ~ (p=0.100 n=3) ²
Query/JSONParallel-12 8.500m ± ∞ ¹ 9.483m ± ∞ ¹ ~ (p=0.100 n=3) ² 8.571m ± ∞ ¹ ~ (p=1.000 n=3) ² 16.088m ± ∞ ¹ ~ (p=0.100 n=3) ² 8.649m ± ∞ ¹ ~ (p=0.700 n=3) ² 15.201m ± ∞ ¹ ~ (p=0.100 n=3) ² 8.718m ± ∞ ¹ ~ (p=0.400 n=3) ² 13.913m ± ∞ ¹ ~ (p=0.100 n=3) ²
geomean 9.126m 9.634m +5.57% 7.891m -13.53% 12.24m +34.13% 8.452m -7.38% 11.94m +30.82% 7.842m -14.06% 11.96m +31.04%
¹ need >= 6 samples for confidence interval at level 0.95
² need >= 4 samples to detect a difference at alpha level 0.05
│ bench_ncruces_direct.txt │ bench_ncruces_driver.txt │ bench_eatonphil_direct.txt │ bench_glebarez_driver.txt │ bench_mattn_driver.txt │ bench_modernc_driver.txt │ bench_tailscale_driver.txt │ bench_zombiezen_direct.txt │
│ B/op │ B/op vs base │ B/op vs base │ B/op vs base │ B/op vs base │ B/op vs base │ B/op vs base │ B/op vs base │
Query/JSON-12 2.000 ± ∞ ¹ 64953.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 1.000 ± ∞ ¹ ~ (p=0.300 n=3) ² 56949.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 201026.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 56966.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 32891.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 403.000 ± ∞ ¹ ~ (p=0.600 n=3) ²
Query/JSONParallel-12 40.00 ± ∞ ¹ 64887.00 ± ∞ ¹ ~ (p=0.100 n=3) ² 43.00 ± ∞ ¹ ~ (p=0.200 n=3) ² 57450.00 ± ∞ ¹ ~ (p=0.100 n=3) ² 201235.00 ± ∞ ¹ ~ (p=0.100 n=3) ² 57471.00 ± ∞ ¹ ~ (p=0.100 n=3) ² 32945.00 ± ∞ ¹ ~ (p=0.100 n=3) ² 469.00 ± ∞ ¹ ~ (p=0.100 n=3) ²
geomean 8.944 63.40Ki +725727.57% 6.557 -26.69% 55.86Ki +639403.72% 196.4Ki +2248607.05% 55.88Ki +639616.05% 32.15Ki +367934.30% 434.7 +4760.65%
¹ need >= 6 samples for confidence interval at level 0.95
² need >= 4 samples to detect a difference at alpha level 0.05
│ bench_ncruces_direct.txt │ bench_ncruces_driver.txt │ bench_eatonphil_direct.txt │ bench_glebarez_driver.txt │ bench_mattn_driver.txt │ bench_modernc_driver.txt │ bench_tailscale_driver.txt │ bench_zombiezen_direct.txt │
│ allocs/op │ allocs/op vs base │ allocs/op vs base │ allocs/op vs base │ allocs/op vs base │ allocs/op vs base │ allocs/op vs base │ allocs/op vs base │
Query/JSON-12 0.000 ± ∞ ¹ 4019.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 0.000 ± ∞ ¹ ~ (p=1.000 n=3) ² 4022.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 5021.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 4022.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 2020.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 6.000 ± ∞ ¹ ~ (p=0.100 n=3) ²
Query/JSONParallel-12 0.000 ± ∞ ¹ 4019.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 0.000 ± ∞ ¹ ~ (p=1.000 n=3) ³ 4022.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 5021.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 4022.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 2020.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 6.000 ± ∞ ¹ ~ (p=0.100 n=3) ²
geomean ⁴ 4.019k ? +0.00% ⁴ 4.022k ? 5.021k ? 4.022k ? 2.020k ? 6.000 ?
¹ need >= 6 samples for confidence interval at level 0.95
² need >= 4 samples to detect a difference at alpha level 0.05
³ all samples are equal
⁴ summaries must be >0 to compute geomean
benchstat bench_ncruces_direct.txt bench_ncruces_driver.txt bench_eatonphil_direct.txt bench_glebarez_driver.txt bench_mattn_driver.txt bench_modernc_driver.txt bench_tailscale_driver.txt bench_zombiezen_direct.txt
goos: darwin
goarch: arm64
pkg: github.com/michaellenaghan/go-sqlite-bench
cpu: Apple M2 Pro
│ bench_ncruces_direct.txt │ bench_ncruces_driver.txt │ bench_eatonphil_direct.txt │ bench_glebarez_driver.txt │ bench_mattn_driver.txt │ bench_modernc_driver.txt │ bench_tailscale_driver.txt │ bench_zombiezen_direct.txt │
│ sec/op │ sec/op vs base │ sec/op vs base │ sec/op vs base │ sec/op vs base │ sec/op vs base │ sec/op vs base │ sec/op vs base │
Query/JSONParallel 10.041m ± ∞ ¹ 9.792m ± ∞ ¹ ~ (p=0.700 n=3) ² 7.274m ± ∞ ¹ ~ (p=0.100 n=3) ² 9.453m ± ∞ ¹ ~ (p=0.100 n=3) ² 7.732m ± ∞ ¹ ~ (p=0.100 n=3) ² 9.413m ± ∞ ¹ ~ (p=0.100 n=3) ² 7.072m ± ∞ ¹ ~ (p=0.100 n=3) ² 10.125m ± ∞ ¹ ~ (p=0.700 n=3) ²
Query/JSONParallel-2 7.359m ± ∞ ¹ 6.987m ± ∞ ¹ ~ (p=0.100 n=3) ² 5.792m ± ∞ ¹ ~ (p=0.400 n=3) ² 7.459m ± ∞ ¹ ~ (p=0.700 n=3) ² 7.342m ± ∞ ¹ ~ (p=1.000 n=3) ² 7.637m ± ∞ ¹ ~ (p=0.700 n=3) ² 5.789m ± ∞ ¹ ~ (p=0.100 n=3) ² 8.633m ± ∞ ¹ ~ (p=0.100 n=3) ²
Query/JSONParallel-4 4.882m ± ∞ ¹ 4.816m ± ∞ ¹ ~ (p=1.000 n=3) ² 4.619m ± ∞ ¹ ~ (p=0.400 n=3) ² 5.927m ± ∞ ¹ ~ (p=0.100 n=3) ² 5.256m ± ∞ ¹ ~ (p=0.200 n=3) ² 5.688m ± ∞ ¹ ~ (p=0.100 n=3) ² 4.441m ± ∞ ¹ ~ (p=0.100 n=3) ² 8.151m ± ∞ ¹ ~ (p=0.100 n=3) ²
Query/JSONParallel-8 7.702m ± ∞ ¹ 7.086m ± ∞ ¹ ~ (p=0.100 n=3) ² 7.254m ± ∞ ¹ ~ (p=0.100 n=3) ² 13.831m ± ∞ ¹ ~ (p=0.100 n=3) ² 7.696m ± ∞ ¹ ~ (p=1.000 n=3) ² 13.719m ± ∞ ¹ ~ (p=0.100 n=3) ² 6.870m ± ∞ ¹ ~ (p=0.100 n=3) ² 15.200m ± ∞ ¹ ~ (p=0.100 n=3) ²
Query/JSONParallel-16 9.570m ± ∞ ¹ 8.875m ± ∞ ¹ ~ (p=0.100 n=3) ² 9.005m ± ∞ ¹ ~ (p=0.100 n=3) ² 16.323m ± ∞ ¹ ~ (p=0.100 n=3) ² 9.646m ± ∞ ¹ ~ (p=0.700 n=3) ² 16.343m ± ∞ ¹ ~ (p=0.100 n=3) ² 8.795m ± ∞ ¹ ~ (p=0.100 n=3) ² 13.976m ± ∞ ¹ ~ (p=0.100 n=3) ²
geomean 7.672m 7.299m -4.86% 6.620m -13.72% 9.884m +28.83% 7.397m -3.59% 9.828m +28.09% 6.429m -16.20% 10.86m +41.60%
¹ need >= 6 samples for confidence interval at level 0.95
² need >= 4 samples to detect a difference at alpha level 0.05
│ bench_ncruces_direct.txt │ bench_ncruces_driver.txt │ bench_eatonphil_direct.txt │ bench_glebarez_driver.txt │ bench_mattn_driver.txt │ bench_modernc_driver.txt │ bench_tailscale_driver.txt │ bench_zombiezen_direct.txt │
│ B/op │ B/op vs base │ B/op vs base │ B/op vs base │ B/op vs base │ B/op vs base │ B/op vs base │ B/op vs base │
Query/JSONParallel 2.000 ± ∞ ¹ 64845.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 1.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 56877.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 201019.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 56861.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 32891.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 367.000 ± ∞ ¹ ~ (p=0.600 n=3) ²
Query/JSONParallel-2 4.000 ± ∞ ¹ 64849.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 2.000 ± ∞ ¹ ~ (p=0.400 n=3) ² 56949.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 201022.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 56905.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 32885.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 365.000 ± ∞ ¹ ~ (p=0.100 n=3) ²
Query/JSONParallel-4 6.000 ± ∞ ¹ 64846.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 6.000 ± ∞ ¹ ~ (p=1.000 n=3) ² 56955.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 201019.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 56948.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 32892.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 368.000 ± ∞ ¹ ~ (p=0.100 n=3) ²
Query/JSONParallel-8 20.000 ± ∞ ¹ 64847.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 5.000 ± ∞ ¹ ~ (p=1.000 n=3) ² 56995.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 201070.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 57002.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 32907.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 426.000 ± ∞ ¹ ~ (p=0.100 n=3) ²
Query/JSONParallel-16 48.00 ± ∞ ¹ 64944.00 ± ∞ ¹ ~ (p=0.100 n=3) ² 70.00 ± ∞ ¹ ~ (p=0.700 n=3) ² 57048.00 ± ∞ ¹ ~ (p=0.100 n=3) ² 201112.00 ± ∞ ¹ ~ (p=0.100 n=3) ² 57075.00 ± ∞ ¹ ~ (p=0.100 n=3) ² 33031.00 ± ∞ ¹ ~ (p=0.100 n=3) ² 538.00 ± ∞ ¹ ~ (p=0.100 n=3) ²
geomean 8.565 63.35Ki +757283.58% 5.305 -38.06% 55.63Ki +665025.92% 196.3Ki +2347359.56% 55.62Ki +664948.61% 32.15Ki +384290.41% 407.9 +4663.18%
¹ need >= 6 samples for confidence interval at level 0.95
² need >= 4 samples to detect a difference at alpha level 0.05
│ bench_ncruces_direct.txt │ bench_ncruces_driver.txt │ bench_eatonphil_direct.txt │ bench_glebarez_driver.txt │ bench_mattn_driver.txt │ bench_modernc_driver.txt │ bench_tailscale_driver.txt │ bench_zombiezen_direct.txt │
│ allocs/op │ allocs/op vs base │ allocs/op vs base │ allocs/op vs base │ allocs/op vs base │ allocs/op vs base │ allocs/op vs base │ allocs/op vs base │
Query/JSONParallel 0.000 ± ∞ ¹ 4019.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 0.000 ± ∞ ¹ ~ (p=1.000 n=3) ² 4021.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 5021.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 4021.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 2020.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 6.000 ± ∞ ¹ ~ (p=0.100 n=3) ²
Query/JSONParallel-2 0.000 ± ∞ ¹ 4019.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 0.000 ± ∞ ¹ ~ (p=1.000 n=3) ³ 4022.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 5021.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 4021.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 2020.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 6.000 ± ∞ ¹ ~ (p=0.100 n=3) ²
Query/JSONParallel-4 0.000 ± ∞ ¹ 4019.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 0.000 ± ∞ ¹ ~ (p=1.000 n=3) ³ 4022.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 5021.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 4021.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 2020.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 6.000 ± ∞ ¹ ~ (p=0.100 n=3) ²
Query/JSONParallel-8 0.000 ± ∞ ¹ 4019.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 0.000 ± ∞ ¹ ~ (p=1.000 n=3) ³ 4022.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 5021.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 4022.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 2020.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 6.000 ± ∞ ¹ ~ (p=0.100 n=3) ²
Query/JSONParallel-16 0.000 ± ∞ ¹ 4019.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 0.000 ± ∞ ¹ ~ (p=1.000 n=3) ³ 4023.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 5021.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 4023.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 2020.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 6.000 ± ∞ ¹ ~ (p=0.100 n=3) ²
geomean ⁴ 4.019k ? +0.00% ⁴ 4.022k ? 5.021k ? 4.022k ? 2.020k ? 6.000 ?
¹ need >= 6 samples for confidence interval at level 0.95
² need >= 4 samples to detect a difference at alpha level 0.05
³ all samples are equal
⁴ summaries must be >0 to compute geomean
SELECT
name, created, id
FROM comments
ORDER BY name, created, idbenchstat bench_ncruces_direct.txt bench_ncruces_driver.txt bench_eatonphil_direct.txt bench_glebarez_driver.txt bench_mattn_driver.txt bench_modernc_driver.txt bench_tailscale_driver.txt bench_zombiezen_direct.txt
goos: darwin
goarch: arm64
pkg: github.com/michaellenaghan/go-sqlite-bench
cpu: Apple M2 Pro
│ bench_ncruces_direct.txt │ bench_ncruces_driver.txt │ bench_eatonphil_direct.txt │ bench_glebarez_driver.txt │ bench_mattn_driver.txt │ bench_modernc_driver.txt │ bench_tailscale_driver.txt │ bench_zombiezen_direct.txt │
│ sec/op │ sec/op vs base │ sec/op vs base │ sec/op vs base │ sec/op vs base │ sec/op vs base │ sec/op vs base │ sec/op vs base │
Query/OrderBy-12 83.46m ± ∞ ¹ 90.57m ± ∞ ¹ ~ (p=0.100 n=3) ² 47.05m ± ∞ ¹ ~ (p=0.100 n=3) ² 81.18m ± ∞ ¹ ~ (p=0.700 n=3) ² 109.21m ± ∞ ¹ ~ (p=0.100 n=3) ² 83.44m ± ∞ ¹ ~ (p=1.000 n=3) ² 61.86m ± ∞ ¹ ~ (p=0.100 n=3) ² 84.56m ± ∞ ¹ ~ (p=0.700 n=3) ²
Query/OrderByParallel-12 48.06m ± ∞ ¹ 51.52m ± ∞ ¹ ~ (p=0.100 n=3) ² 48.50m ± ∞ ¹ ~ (p=0.400 n=3) ² 75.27m ± ∞ ¹ ~ (p=0.100 n=3) ² 91.74m ± ∞ ¹ ~ (p=0.100 n=3) ² 71.91m ± ∞ ¹ ~ (p=0.100 n=3) ² 51.47m ± ∞ ¹ ~ (p=0.100 n=3) ² 79.94m ± ∞ ¹ ~ (p=0.100 n=3) ²
geomean 63.33m 68.31m +7.85% 47.77m -24.58% 78.17m +23.43% 100.1m +58.04% 77.46m +22.31% 56.43m -10.91% 82.22m +29.82%
¹ need >= 6 samples for confidence interval at level 0.95
² need >= 4 samples to detect a difference at alpha level 0.05
│ bench_ncruces_direct.txt │ bench_ncruces_driver.txt │ bench_eatonphil_direct.txt │ bench_glebarez_driver.txt │ bench_mattn_driver.txt │ bench_modernc_driver.txt │ bench_tailscale_driver.txt │ bench_zombiezen_direct.txt │
│ B/op │ B/op vs base │ B/op vs base │ B/op vs base │ B/op vs base │ B/op vs base │ B/op vs base │ B/op vs base │
Query/OrderBy-12 57461.0 ± ∞ ¹ 6399308.0 ± ∞ ¹ ~ (p=0.100 n=3) ² 118.0 ± ∞ ¹ ~ (p=0.100 n=3) ² 6397822.0 ± ∞ ¹ ~ (p=0.100 n=3) ² 11999108.0 ± ∞ ¹ ~ (p=0.100 n=3) ² 6397808.0 ± ∞ ¹ ~ (p=0.100 n=3) ² 2798877.0 ± ∞ ¹ ~ (p=0.100 n=3) ² 7222.0 ± ∞ ¹ ~ (p=0.100 n=3) ²
Query/OrderByParallel-12 695.0 ± ∞ ¹ 6399480.0 ± ∞ ¹ ~ (p=0.100 n=3) ² 532.0 ± ∞ ¹ ~ (p=0.700 n=3) ² 6398311.0 ± ∞ ¹ ~ (p=0.100 n=3) ² 12000026.0 ± ∞ ¹ ~ (p=0.100 n=3) ² 6398237.0 ± ∞ ¹ ~ (p=0.100 n=3) ² 2799192.0 ± ∞ ¹ ~ (p=0.100 n=3) ² 1448.0 ± ∞ ¹ ~ (p=0.700 n=3) ²
geomean 6.171Ki 6.103Mi +101165.11% 250.6 -96.04% 6.102Mi +101144.11% 11.44Mi +189783.22% 6.102Mi +101143.41% 2.669Mi +44192.40% 3.158Ki -48.83%
¹ need >= 6 samples for confidence interval at level 0.95
² need >= 4 samples to detect a difference at alpha level 0.05
│ bench_ncruces_direct.txt │ bench_ncruces_driver.txt │ bench_eatonphil_direct.txt │ bench_glebarez_driver.txt │ bench_mattn_driver.txt │ bench_modernc_driver.txt │ bench_tailscale_driver.txt │ bench_zombiezen_direct.txt │
│ allocs/op │ allocs/op vs base │ allocs/op vs base │ allocs/op vs base │ allocs/op vs base │ allocs/op vs base │ allocs/op vs base │ allocs/op vs base │
Query/OrderBy-12 20.000 ± ∞ ¹ 349779.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 1.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 449782.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 349771.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 449781.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 149766.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 37.000 ± ∞ ¹ ~ (p=0.300 n=3) ²
Query/OrderByParallel-12 9.000 ± ∞ ¹ 349778.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 1.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 449784.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 349780.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 449785.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 149767.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 19.000 ± ∞ ¹ ~ (p=0.100 n=3) ²
geomean 13.42 349.8k +2606995.01% 1.000 -92.55% 449.8k +3352384.54% 349.8k +2606972.65% 449.8k +3352384.54% 149.8k +1116193.58% 26.51 +97.62%
¹ need >= 6 samples for confidence interval at level 0.95
² need >= 4 samples to detect a difference at alpha level 0.05
benchstat bench_ncruces_direct.txt bench_ncruces_driver.txt bench_eatonphil_direct.txt bench_glebarez_driver.txt bench_mattn_driver.txt bench_modernc_driver.txt bench_tailscale_driver.txt bench_zombiezen_direct.txt
goos: darwin
goarch: arm64
pkg: github.com/michaellenaghan/go-sqlite-bench
cpu: Apple M2 Pro
│ bench_ncruces_direct.txt │ bench_ncruces_driver.txt │ bench_eatonphil_direct.txt │ bench_glebarez_driver.txt │ bench_mattn_driver.txt │ bench_modernc_driver.txt │ bench_tailscale_driver.txt │ bench_zombiezen_direct.txt │
│ sec/op │ sec/op vs base │ sec/op vs base │ sec/op vs base │ sec/op vs base │ sec/op vs base │ sec/op vs base │ sec/op vs base │
Query/OrderByParallel 79.20m ± ∞ ¹ 90.64m ± ∞ ¹ ~ (p=0.100 n=3) ² 46.44m ± ∞ ¹ ~ (p=0.100 n=3) ² 83.80m ± ∞ ¹ ~ (p=0.100 n=3) ² 87.88m ± ∞ ¹ ~ (p=0.100 n=3) ² 82.67m ± ∞ ¹ ~ (p=0.100 n=3) ² 61.85m ± ∞ ¹ ~ (p=0.100 n=3) ² 87.37m ± ∞ ¹ ~ (p=0.700 n=3) ²
Query/OrderByParallel-2 49.64m ± ∞ ¹ 59.12m ± ∞ ¹ ~ (p=0.100 n=3) ² 34.91m ± ∞ ¹ ~ (p=0.100 n=3) ² 56.34m ± ∞ ¹ ~ (p=0.100 n=3) ² 63.23m ± ∞ ¹ ~ (p=0.100 n=3) ² 56.53m ± ∞ ¹ ~ (p=0.100 n=3) ² 42.48m ± ∞ ¹ ~ (p=0.100 n=3) ² 51.81m ± ∞ ¹ ~ (p=0.100 n=3) ²
Query/OrderByParallel-4 33.24m ± ∞ ¹ 37.22m ± ∞ ¹ ~ (p=0.100 n=3) ² 24.03m ± ∞ ¹ ~ (p=0.200 n=3) ² 38.87m ± ∞ ¹ ~ (p=0.100 n=3) ² 51.52m ± ∞ ¹ ~ (p=0.100 n=3) ² 37.12m ± ∞ ¹ ~ (p=0.100 n=3) ² 29.48m ± ∞ ¹ ~ (p=0.100 n=3) ² 43.71m ± ∞ ¹ ~ (p=0.100 n=3) ²
Query/OrderByParallel-8 40.55m ± ∞ ¹ 44.51m ± ∞ ¹ ~ (p=0.100 n=3) ² 38.15m ± ∞ ¹ ~ (p=0.700 n=3) ² 66.22m ± ∞ ¹ ~ (p=0.100 n=3) ² 74.20m ± ∞ ¹ ~ (p=0.100 n=3) ² 65.41m ± ∞ ¹ ~ (p=0.100 n=3) ² 41.51m ± ∞ ¹ ~ (p=0.100 n=3) ² 76.77m ± ∞ ¹ ~ (p=0.100 n=3) ²
Query/OrderByParallel-16 51.16m ± ∞ ¹ 52.81m ± ∞ ¹ ~ (p=0.700 n=3) ² 48.87m ± ∞ ¹ ~ (p=0.200 n=3) ² 71.58m ± ∞ ¹ ~ (p=0.100 n=3) ² 97.17m ± ∞ ¹ ~ (p=0.100 n=3) ² 68.18m ± ∞ ¹ ~ (p=0.100 n=3) ² 51.36m ± ∞ ¹ ~ (p=0.700 n=3) ² 78.76m ± ∞ ¹ ~ (p=0.100 n=3) ²
geomean 48.60m 54.22m +11.58% 37.34m -23.15% 61.36m +26.27% 72.94m +50.09% 59.94m +23.33% 44.01m -9.44% 65.40m +34.57%
¹ need >= 6 samples for confidence interval at level 0.95
² need >= 4 samples to detect a difference at alpha level 0.05
│ bench_ncruces_direct.txt │ bench_ncruces_driver.txt │ bench_eatonphil_direct.txt │ bench_glebarez_driver.txt │ bench_mattn_driver.txt │ bench_modernc_driver.txt │ bench_tailscale_driver.txt │ bench_zombiezen_direct.txt │
│ B/op │ B/op vs base │ B/op vs base │ B/op vs base │ B/op vs base │ B/op vs base │ B/op vs base │ B/op vs base │
Query/OrderByParallel 55903.00 ± ∞ ¹ 6399279.00 ± ∞ ¹ ~ (p=0.100 n=3) ² 83.00 ± ∞ ¹ ~ (p=0.100 n=3) ² 6397704.00 ± ∞ ¹ ~ (p=0.100 n=3) ² 11999009.00 ± ∞ ¹ ~ (p=0.100 n=3) ² 6397711.00 ± ∞ ¹ ~ (p=0.100 n=3) ² 2798848.00 ± ∞ ¹ ~ (p=0.100 n=3) ² 6477.00 ± ∞ ¹ ~ (p=0.100 n=3) ²
Query/OrderByParallel-2 392.000 ± ∞ ¹ 6399274.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 9.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 6397723.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 11999045.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 6397741.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 2798844.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 1131.000 ± ∞ ¹ ~ (p=0.600 n=3) ²
Query/OrderByParallel-4 413.00 ± ∞ ¹ 6399273.00 ± ∞ ¹ ~ (p=0.100 n=3) ² 14.00 ± ∞ ¹ ~ (p=0.100 n=3) ² 6397889.00 ± ∞ ¹ ~ (p=0.100 n=3) ² 11999095.00 ± ∞ ¹ ~ (p=0.100 n=3) ² 6397862.00 ± ∞ ¹ ~ (p=0.100 n=3) ² 2798861.00 ± ∞ ¹ ~ (p=0.100 n=3) ² 1128.00 ± ∞ ¹ ~ (p=0.100 n=3) ²
Query/OrderByParallel-8 493.0 ± ∞ ¹ 6399360.0 ± ∞ ¹ ~ (p=0.100 n=3) ² 233.0 ± ∞ ¹ ~ (p=0.200 n=3) ² 6398016.0 ± ∞ ¹ ~ (p=0.100 n=3) ² 12000246.0 ± ∞ ¹ ~ (p=0.100 n=3) ² 6398052.0 ± ∞ ¹ ~ (p=0.100 n=3) ² 2798894.0 ± ∞ ¹ ~ (p=0.100 n=3) ² 1480.0 ± ∞ ¹ ~ (p=0.100 n=3) ²
Query/OrderByParallel-16 715.0 ± ∞ ¹ 6399513.0 ± ∞ ¹ ~ (p=0.100 n=3) ² 312.0 ± ∞ ¹ ~ (p=0.700 n=3) ² 6398773.0 ± ∞ ¹ ~ (p=0.100 n=3) ² 12001562.0 ± ∞ ¹ ~ (p=0.100 n=3) ² 6398608.0 ± ∞ ¹ ~ (p=0.100 n=3) ² 2799260.0 ± ∞ ¹ ~ (p=0.100 n=3) ² 1661.0 ± ∞ ¹ ~ (p=0.100 n=3) ²
geomean 1.232Ki 6.103Mi +507323.21% 59.73 -95.26% 6.102Mi +507218.64% 11.44Mi +951400.14% 6.102Mi +507216.57% 2.669Mi +221836.62% 1.783Ki +44.81%
¹ need >= 6 samples for confidence interval at level 0.95
² need >= 4 samples to detect a difference at alpha level 0.05
│ bench_ncruces_direct.txt │ bench_ncruces_driver.txt │ bench_eatonphil_direct.txt │ bench_glebarez_driver.txt │ bench_mattn_driver.txt │ bench_modernc_driver.txt │ bench_tailscale_driver.txt │ bench_zombiezen_direct.txt │
│ allocs/op │ allocs/op vs base │ allocs/op vs base │ allocs/op vs base │ allocs/op vs base │ allocs/op vs base │ allocs/op vs base │ allocs/op vs base │
Query/OrderByParallel 20.000 ± ∞ ¹ 349779.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 1.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 449780.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 349771.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 449780.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 149766.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 34.000 ± ∞ ¹ ~ (p=0.400 n=3) ²
Query/OrderByParallel-2 8.000 ± ∞ ¹ 349779.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 0.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 449780.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 349771.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 449780.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 149766.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 16.000 ± ∞ ¹ ~ (p=0.100 n=3) ²
Query/OrderByParallel-4 8.000 ± ∞ ¹ 349778.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 0.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 449782.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 349772.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 449781.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 149766.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 16.000 ± ∞ ¹ ~ (p=0.100 n=3) ²
Query/OrderByParallel-8 9.000 ± ∞ ¹ 349778.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 1.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 449783.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 349776.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 449784.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 149766.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 18.000 ± ∞ ¹ ~ (p=0.100 n=3) ²
Query/OrderByParallel-16 10.000 ± ∞ ¹ 349779.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 2.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 449791.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 349791.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 449789.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 149769.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 19.000 ± ∞ ¹ ~ (p=0.100 n=3) ²
geomean 10.29 349.8k +3400086.50% ? ³ ⁴ 449.8k +4372228.00% 349.8k +3400063.17% 449.8k +4372224.11% 149.8k +1455776.29% 19.71 +91.63%
¹ need >= 6 samples for confidence interval at level 0.95
² need >= 4 samples to detect a difference at alpha level 0.05
³ summaries must be >0 to compute geomean
⁴ ratios must be >0 to compute geomean
WITH RECURSIVE dates(day) AS (
SELECT date('now', '-30 days')
UNION ALL
SELECT date(day, '+1 day')
FROM dates
WHERE day < date('now')
)
SELECT day,
(SELECT COUNT(*) FROM posts WHERE date(created) = day) as day_total
FROM dates
ORDER BY daybenchstat bench_ncruces_direct.txt bench_ncruces_driver.txt bench_eatonphil_direct.txt bench_glebarez_driver.txt bench_mattn_driver.txt bench_modernc_driver.txt bench_tailscale_driver.txt bench_zombiezen_direct.txt
goos: darwin
goarch: arm64
pkg: github.com/michaellenaghan/go-sqlite-bench
cpu: Apple M2 Pro
│ bench_ncruces_direct.txt │ bench_ncruces_driver.txt │ bench_eatonphil_direct.txt │ bench_glebarez_driver.txt │ bench_mattn_driver.txt │ bench_modernc_driver.txt │ bench_tailscale_driver.txt │ bench_zombiezen_direct.txt │
│ sec/op │ sec/op vs base │ sec/op vs base │ sec/op vs base │ sec/op vs base │ sec/op vs base │ sec/op vs base │ sec/op vs base │
Query/RecursiveCTE-12 6.276m ± ∞ ¹ 6.256m ± ∞ ¹ ~ (p=0.400 n=3) ² 2.503m ± ∞ ¹ ~ (p=0.100 n=3) ² 5.287m ± ∞ ¹ ~ (p=0.100 n=3) ² 2.474m ± ∞ ¹ ~ (p=0.100 n=3) ² 5.313m ± ∞ ¹ ~ (p=0.100 n=3) ² 2.431m ± ∞ ¹ ~ (p=0.100 n=3) ² 5.268m ± ∞ ¹ ~ (p=0.100 n=3) ²
Query/RecursiveCTEParallel-12 670.0µ ± ∞ ¹ 687.9µ ± ∞ ¹ ~ (p=0.100 n=3) ² 276.9µ ± ∞ ¹ ~ (p=0.100 n=3) ² 657.3µ ± ∞ ¹ ~ (p=0.400 n=3) ² 275.0µ ± ∞ ¹ ~ (p=0.100 n=3) ² 633.6µ ± ∞ ¹ ~ (p=0.100 n=3) ² 272.0µ ± ∞ ¹ ~ (p=0.100 n=3) ² 655.8µ ± ∞ ¹ ~ (p=0.100 n=3) ²
geomean 2.051m 2.074m +1.16% 832.5µ -59.40% 1.864m -9.09% 824.8µ -59.78% 1.835m -10.53% 813.2µ -60.34% 1.859m -9.36%
¹ need >= 6 samples for confidence interval at level 0.95
² need >= 4 samples to detect a difference at alpha level 0.05
│ bench_ncruces_direct.txt │ bench_ncruces_driver.txt │ bench_eatonphil_direct.txt │ bench_glebarez_driver.txt │ bench_mattn_driver.txt │ bench_modernc_driver.txt │ bench_tailscale_driver.txt │ bench_zombiezen_direct.txt │
│ B/op │ B/op vs base │ B/op vs base │ B/op vs base │ B/op vs base │ B/op vs base │ B/op vs base │ B/op vs base │
Query/RecursiveCTE-12 1.000 ± ∞ ¹ 2488.000 ± ∞ ¹ ~ (p=0.600 n=3) ² 0.000 ± ∞ ¹ ~ (p=0.300 n=3) ² 2354.000 ± ∞ ¹ ~ (p=0.400 n=3) ² 6859.000 ± ∞ ¹ ~ (p=0.600 n=3) ² 2293.000 ± ∞ ¹ ~ (p=0.600 n=3) ² 1505.000 ± ∞ ¹ ~ (p=0.400 n=3) ² 389.000 ± ∞ ¹ ~ (p=0.600 n=3) ²
Query/RecursiveCTEParallel-12 2.000 ± ∞ ¹ 2483.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 1.000 ± ∞ ¹ ~ (p=0.800 n=3) ² 2358.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 6850.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 2286.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 1496.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 368.000 ± ∞ ¹ ~ (p=0.100 n=3) ²
geomean 1.414 2.427Ki +175651.30% ? ³ ⁴ 2.301Ki +166494.30% 6.694Ki +484586.24% 2.236Ki +161791.91% 1.465Ki +106000.90% 378.4 +26653.69%
¹ need >= 6 samples for confidence interval at level 0.95
² need >= 4 samples to detect a difference at alpha level 0.05
³ summaries must be >0 to compute geomean
⁴ ratios must be >0 to compute geomean
│ bench_ncruces_direct.txt │ bench_ncruces_driver.txt │ bench_eatonphil_direct.txt │ bench_glebarez_driver.txt │ bench_mattn_driver.txt │ bench_modernc_driver.txt │ bench_tailscale_driver.txt │ bench_zombiezen_direct.txt │
│ allocs/op │ allocs/op vs base │ allocs/op vs base │ allocs/op vs base │ allocs/op vs base │ allocs/op vs base │ allocs/op vs base │ allocs/op vs base │
Query/RecursiveCTE-12 0.000 ± ∞ ¹ 110.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 0.000 ± ∞ ¹ ~ (p=1.000 n=3) ² 113.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 143.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 112.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 49.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 6.000 ± ∞ ¹ ~ (p=0.100 n=3) ²
Query/RecursiveCTEParallel-12 0.000 ± ∞ ¹ 110.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 0.000 ± ∞ ¹ ~ (p=1.000 n=3) ³ 113.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 142.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 112.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 48.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 6.000 ± ∞ ¹ ~ (p=0.100 n=3) ²
geomean ⁴ 110.0 ? +0.00% ⁴ 113.0 ? 142.5 ? 112.0 ? 48.50 ? 6.000 ?
¹ need >= 6 samples for confidence interval at level 0.95
² need >= 4 samples to detect a difference at alpha level 0.05
³ all samples are equal
⁴ summaries must be >0 to compute geomean
benchstat bench_ncruces_direct.txt bench_ncruces_driver.txt bench_eatonphil_direct.txt bench_glebarez_driver.txt bench_mattn_driver.txt bench_modernc_driver.txt bench_tailscale_driver.txt bench_zombiezen_direct.txt
goos: darwin
goarch: arm64
pkg: github.com/michaellenaghan/go-sqlite-bench
cpu: Apple M2 Pro
│ bench_ncruces_direct.txt │ bench_ncruces_driver.txt │ bench_eatonphil_direct.txt │ bench_glebarez_driver.txt │ bench_mattn_driver.txt │ bench_modernc_driver.txt │ bench_tailscale_driver.txt │ bench_zombiezen_direct.txt │
│ sec/op │ sec/op vs base │ sec/op vs base │ sec/op vs base │ sec/op vs base │ sec/op vs base │ sec/op vs base │ sec/op vs base │
Query/RecursiveCTEParallel 6.152m ± ∞ ¹ 6.186m ± ∞ ¹ ~ (p=0.100 n=3) ² 2.512m ± ∞ ¹ ~ (p=0.100 n=3) ² 5.289m ± ∞ ¹ ~ (p=0.100 n=3) ² 2.464m ± ∞ ¹ ~ (p=0.100 n=3) ² 5.281m ± ∞ ¹ ~ (p=0.100 n=3) ² 2.422m ± ∞ ¹ ~ (p=0.100 n=3) ² 5.239m ± ∞ ¹ ~ (p=0.100 n=3) ²
Query/RecursiveCTEParallel-2 3.122m ± ∞ ¹ 3.117m ± ∞ ¹ ~ (p=0.700 n=3) ² 1.269m ± ∞ ¹ ~ (p=0.100 n=3) ² 2.672m ± ∞ ¹ ~ (p=0.100 n=3) ² 1.249m ± ∞ ¹ ~ (p=0.100 n=3) ² 2.667m ± ∞ ¹ ~ (p=0.100 n=3) ² 1.230m ± ∞ ¹ ~ (p=0.100 n=3) ² 2.650m ± ∞ ¹ ~ (p=0.100 n=3) ²
Query/RecursiveCTEParallel-4 1593.6µ ± ∞ ¹ 1597.5µ ± ∞ ¹ ~ (p=0.100 n=3) ² 645.3µ ± ∞ ¹ ~ (p=0.100 n=3) ² 1374.1µ ± ∞ ¹ ~ (p=0.100 n=3) ² 632.0µ ± ∞ ¹ ~ (p=0.100 n=3) ² 1368.1µ ± ∞ ¹ ~ (p=0.100 n=3) ² 623.8µ ± ∞ ¹ ~ (p=0.100 n=3) ² 1362.1µ ± ∞ ¹ ~ (p=0.100 n=3) ²
Query/RecursiveCTEParallel-8 799.0µ ± ∞ ¹ 805.5µ ± ∞ ¹ ~ (p=0.100 n=3) ² 324.3µ ± ∞ ¹ ~ (p=0.100 n=3) ² 749.6µ ± ∞ ¹ ~ (p=0.100 n=3) ² 324.2µ ± ∞ ¹ ~ (p=0.100 n=3) ² 742.0µ ± ∞ ¹ ~ (p=0.100 n=3) ² 317.4µ ± ∞ ¹ ~ (p=0.100 n=3) ² 746.6µ ± ∞ ¹ ~ (p=0.100 n=3) ²
Query/RecursiveCTEParallel-16 674.0µ ± ∞ ¹ 678.3µ ± ∞ ¹ ~ (p=0.400 n=3) ² 280.2µ ± ∞ ¹ ~ (p=0.100 n=3) ² 647.3µ ± ∞ ¹ ~ (p=0.100 n=3) ² 279.5µ ± ∞ ¹ ~ (p=0.100 n=3) ² 642.4µ ± ∞ ¹ ~ (p=0.100 n=3) ² 274.0µ ± ∞ ¹ ~ (p=0.100 n=3) ² 663.6µ ± ∞ ¹ ~ (p=0.100 n=3) ²
geomean 1.752m 1.759m +0.42% 715.1µ -59.17% 1.566m -10.58% 706.6µ -59.66% 1.558m -11.04% 694.6µ -60.34% 1.564m -10.68%
¹ need >= 6 samples for confidence interval at level 0.95
² need >= 4 samples to detect a difference at alpha level 0.05
│ bench_ncruces_direct.txt │ bench_ncruces_driver.txt │ bench_eatonphil_direct.txt │ bench_glebarez_driver.txt │ bench_mattn_driver.txt │ bench_modernc_driver.txt │ bench_tailscale_driver.txt │ bench_zombiezen_direct.txt │
│ B/op │ B/op vs base │ B/op vs base │ B/op vs base │ B/op vs base │ B/op vs base │ B/op vs base │ B/op vs base │
Query/RecursiveCTEParallel 1.000 ± ∞ ¹ 2482.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 0.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 2260.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 6857.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 2258.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 1504.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 362.000 ± ∞ ¹ ~ (p=0.100 n=3) ²
Query/RecursiveCTEParallel-2 1.000 ± ∞ ¹ 2481.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 0.000 ± ∞ ¹ ~ (p=0.400 n=3) ² 2263.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 6825.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 2258.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 1502.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 366.000 ± ∞ ¹ ~ (p=0.100 n=3) ²
Query/RecursiveCTEParallel-4 1.000 ± ∞ ¹ 2481.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 0.000 ± ∞ ¹ ~ (p=0.400 n=3) ² 2280.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 6842.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 2259.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 1502.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 373.000 ± ∞ ¹ ~ (p=0.100 n=3) ²
Query/RecursiveCTEParallel-8 0.0 ± ∞ ¹ 2482.0 ± ∞ ¹ ~ (p=0.100 n=3) ² 0.0 ± ∞ ¹ ~ (p=1.000 n=3) ² 2356.0 ± ∞ ¹ ~ (p=0.100 n=3) ² 6857.0 ± ∞ ¹ ~ (p=0.100 n=3) ² 2280.0 ± ∞ ¹ ~ (p=0.100 n=3) ² 1504.0 ± ∞ ¹ ~ (p=0.100 n=3) ² 363.0 ± ∞ ¹ ~ (p=0.100 n=3) ²
Query/RecursiveCTEParallel-16 0.000 ± ∞ ¹ 2489.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 1.000 ± ∞ ¹ ~ (p=0.700 n=3) ² 2361.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 6862.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 2284.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 1506.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 371.000 ± ∞ ¹ ~ (p=0.100 n=3) ²
geomean ³ 2.425Ki ? ? ³ 2.250Ki ? 6.688Ki ? 2.215Ki ? 1.468Ki ? 367.0 ?
¹ need >= 6 samples for confidence interval at level 0.95
² need >= 4 samples to detect a difference at alpha level 0.05
³ summaries must be >0 to compute geomean
│ bench_ncruces_direct.txt │ bench_ncruces_driver.txt │ bench_eatonphil_direct.txt │ bench_glebarez_driver.txt │ bench_mattn_driver.txt │ bench_modernc_driver.txt │ bench_tailscale_driver.txt │ bench_zombiezen_direct.txt │
│ allocs/op │ allocs/op vs base │ allocs/op vs base │ allocs/op vs base │ allocs/op vs base │ allocs/op vs base │ allocs/op vs base │ allocs/op vs base │
Query/RecursiveCTEParallel 0.000 ± ∞ ¹ 110.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 0.000 ± ∞ ¹ ~ (p=1.000 n=3) ³ 112.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 143.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 112.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 49.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 6.000 ± ∞ ¹ ~ (p=0.100 n=3) ²
Query/RecursiveCTEParallel-2 0.000 ± ∞ ¹ 110.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 0.000 ± ∞ ¹ ~ (p=1.000 n=3) ³ 112.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 142.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 112.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 48.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 6.000 ± ∞ ¹ ~ (p=0.100 n=3) ²
Query/RecursiveCTEParallel-4 0.000 ± ∞ ¹ 110.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 0.000 ± ∞ ¹ ~ (p=1.000 n=3) ³ 112.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 142.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 112.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 48.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 6.000 ± ∞ ¹ ~ (p=0.100 n=3) ²
Query/RecursiveCTEParallel-8 0.000 ± ∞ ¹ 110.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 0.000 ± ∞ ¹ ~ (p=1.000 n=3) ³ 113.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 143.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 112.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 49.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 6.000 ± ∞ ¹ ~ (p=0.100 n=3) ²
Query/RecursiveCTEParallel-16 0.000 ± ∞ ¹ 110.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 0.000 ± ∞ ¹ ~ (p=1.000 n=3) ³ 113.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 143.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 112.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 49.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 6.000 ± ∞ ¹ ~ (p=0.100 n=3) ²
geomean ⁴ 110.0 ? +0.00% ⁴ 112.4 ? 142.6 ? 112.0 ? 48.60 ? 6.000 ?
¹ need >= 6 samples for confidence interval at level 0.95
² need >= 4 samples to detect a difference at alpha level 0.05
³ all samples are equal
⁴ summaries must be >0 to compute geomean
WITH day_totals AS (
SELECT date(created) as day, COUNT(*) as day_total
FROM posts
GROUP BY day
)
SELECT day, day_total,
SUM(day_total) OVER (ORDER BY day) as running_total
FROM day_totals
ORDER BY daybenchstat bench_ncruces_direct.txt bench_ncruces_driver.txt bench_eatonphil_direct.txt bench_glebarez_driver.txt bench_mattn_driver.txt bench_modernc_driver.txt bench_tailscale_driver.txt bench_zombiezen_direct.txt
goos: darwin
goarch: arm64
pkg: github.com/michaellenaghan/go-sqlite-bench
cpu: Apple M2 Pro
│ bench_ncruces_direct.txt │ bench_ncruces_driver.txt │ bench_eatonphil_direct.txt │ bench_glebarez_driver.txt │ bench_mattn_driver.txt │ bench_modernc_driver.txt │ bench_tailscale_driver.txt │ bench_zombiezen_direct.txt │
│ sec/op │ sec/op vs base │ sec/op vs base │ sec/op vs base │ sec/op vs base │ sec/op vs base │ sec/op vs base │ sec/op vs base │
Query/Window-12 2166.2µ ± ∞ ¹ 2448.0µ ± ∞ ¹ ~ (p=0.100 n=3) ² 917.2µ ± ∞ ¹ ~ (p=0.100 n=3) ² 1896.8µ ± ∞ ¹ ~ (p=0.100 n=3) ² 2078.7µ ± ∞ ¹ ~ (p=0.100 n=3) ² 1900.1µ ± ∞ ¹ ~ (p=0.100 n=3) ² 1055.3µ ± ∞ ¹ ~ (p=0.100 n=3) ² 1634.5µ ± ∞ ¹ ~ (p=0.100 n=3) ²
Query/WindowParallel-12 238.7µ ± ∞ ¹ 321.5µ ± ∞ ¹ ~ (p=0.100 n=3) ² 122.0µ ± ∞ ¹ ~ (p=0.100 n=3) ² 899.9µ ± ∞ ¹ ~ (p=0.100 n=3) ² 1069.2µ ± ∞ ¹ ~ (p=0.100 n=3) ² 908.9µ ± ∞ ¹ ~ (p=0.100 n=3) ² 133.3µ ± ∞ ¹ ~ (p=0.100 n=3) ² 864.8µ ± ∞ ¹ ~ (p=0.100 n=3) ²
geomean 719.1µ 887.2µ +23.37% 334.5µ -53.48% 1.307m +81.68% 1.491m +107.31% 1.314m +82.74% 375.1µ -47.84% 1.189m +65.33%
¹ need >= 6 samples for confidence interval at level 0.95
² need >= 4 samples to detect a difference at alpha level 0.05
│ bench_ncruces_direct.txt │ bench_ncruces_driver.txt │ bench_eatonphil_direct.txt │ bench_glebarez_driver.txt │ bench_mattn_driver.txt │ bench_modernc_driver.txt │ bench_tailscale_driver.txt │ bench_zombiezen_direct.txt │
│ B/op │ B/op vs base │ B/op vs base │ B/op vs base │ B/op vs base │ B/op vs base │ B/op vs base │ B/op vs base │
Query/Window-12 0.0 ± ∞ ¹ 62767.0 ± ∞ ¹ ~ (p=0.100 n=3) ² 0.0 ± ∞ ¹ ~ (p=1.000 n=3) ² 54897.0 ± ∞ ¹ ~ (p=0.100 n=3) ² 198937.0 ± ∞ ¹ ~ (p=0.100 n=3) ² 54911.0 ± ∞ ¹ ~ (p=0.100 n=3) ² 30798.0 ± ∞ ¹ ~ (p=0.100 n=3) ² 376.0 ± ∞ ¹ ~ (p=0.600 n=3) ²
Query/WindowParallel-12 1.000 ± ∞ ¹ 62763.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 1.000 ± ∞ ¹ ~ (p=1.000 n=3) ² 54886.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 198953.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 54907.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 30765.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 372.000 ± ∞ ¹ ~ (p=0.100 n=3) ²
geomean ³ 61.29Ki ? +0.00% ³ 53.60Ki ? 194.3Ki ? 53.62Ki ? 30.06Ki ? 374.0 ?
¹ need >= 6 samples for confidence interval at level 0.95
² need >= 4 samples to detect a difference at alpha level 0.05
³ summaries must be >0 to compute geomean
│ bench_ncruces_direct.txt │ bench_ncruces_driver.txt │ bench_eatonphil_direct.txt │ bench_glebarez_driver.txt │ bench_mattn_driver.txt │ bench_modernc_driver.txt │ bench_tailscale_driver.txt │ bench_zombiezen_direct.txt │
│ allocs/op │ allocs/op vs base │ allocs/op vs base │ allocs/op vs base │ allocs/op vs base │ allocs/op vs base │ allocs/op vs base │ allocs/op vs base │
Query/Window-12 0.000 ± ∞ ¹ 3763.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 0.000 ± ∞ ¹ ~ (p=1.000 n=3) ³ 3766.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 4765.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 3766.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 1764.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 6.000 ± ∞ ¹ ~ (p=0.100 n=3) ²
Query/WindowParallel-12 0.000 ± ∞ ¹ 3763.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 0.000 ± ∞ ¹ ~ (p=1.000 n=3) ³ 3766.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 4765.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 3766.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 1763.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 6.000 ± ∞ ¹ ~ (p=0.100 n=3) ²
geomean ⁴ 3.763k ? +0.00% ⁴ 3.766k ? 4.765k ? 3.766k ? 1.763k ? 6.000 ?
¹ need >= 6 samples for confidence interval at level 0.95
² need >= 4 samples to detect a difference at alpha level 0.05
³ all samples are equal
⁴ summaries must be >0 to compute geomean
benchstat bench_ncruces_direct.txt bench_ncruces_driver.txt bench_eatonphil_direct.txt bench_glebarez_driver.txt bench_mattn_driver.txt bench_modernc_driver.txt bench_tailscale_driver.txt bench_zombiezen_direct.txt
goos: darwin
goarch: arm64
pkg: github.com/michaellenaghan/go-sqlite-bench
cpu: Apple M2 Pro
│ bench_ncruces_direct.txt │ bench_ncruces_driver.txt │ bench_eatonphil_direct.txt │ bench_glebarez_driver.txt │ bench_mattn_driver.txt │ bench_modernc_driver.txt │ bench_tailscale_driver.txt │ bench_zombiezen_direct.txt │
│ sec/op │ sec/op vs base │ sec/op vs base │ sec/op vs base │ sec/op vs base │ sec/op vs base │ sec/op vs base │ sec/op vs base │
Query/WindowParallel 2119.8µ ± ∞ ¹ 2406.5µ ± ∞ ¹ ~ (p=0.100 n=3) ² 930.8µ ± ∞ ¹ ~ (p=0.100 n=3) ² 1884.7µ ± ∞ ¹ ~ (p=0.100 n=3) ² 1568.3µ ± ∞ ¹ ~ (p=0.100 n=3) ² 1876.7µ ± ∞ ¹ ~ (p=0.100 n=3) ² 1060.8µ ± ∞ ¹ ~ (p=0.100 n=3) ² 1614.3µ ± ∞ ¹ ~ (p=0.100 n=3) ²
Query/WindowParallel-2 1070.4µ ± ∞ ¹ 1234.2µ ± ∞ ¹ ~ (p=0.100 n=3) ² 476.6µ ± ∞ ¹ ~ (p=0.100 n=3) ² 993.3µ ± ∞ ¹ ~ (p=0.100 n=3) ² 849.7µ ± ∞ ¹ ~ (p=0.100 n=3) ² 977.6µ ± ∞ ¹ ~ (p=0.100 n=3) ² 544.2µ ± ∞ ¹ ~ (p=0.100 n=3) ² 846.4µ ± ∞ ¹ ~ (p=0.100 n=3) ²
Query/WindowParallel-4 544.7µ ± ∞ ¹ 643.9µ ± ∞ ¹ ~ (p=0.100 n=3) ² 254.5µ ± ∞ ¹ ~ (p=0.100 n=3) ² 541.9µ ± ∞ ¹ ~ (p=0.700 n=3) ² 503.4µ ± ∞ ¹ ~ (p=0.700 n=3) ² 539.5µ ± ∞ ¹ ~ (p=0.100 n=3) ² 283.7µ ± ∞ ¹ ~ (p=0.100 n=3) ² 465.0µ ± ∞ ¹ ~ (p=0.100 n=3) ²
Query/WindowParallel-8 277.1µ ± ∞ ¹ 352.7µ ± ∞ ¹ ~ (p=0.100 n=3) ² 142.6µ ± ∞ ¹ ~ (p=0.100 n=3) ² 864.2µ ± ∞ ¹ ~ (p=0.100 n=3) ² 1071.9µ ± ∞ ¹ ~ (p=0.100 n=3) ² 879.3µ ± ∞ ¹ ~ (p=0.100 n=3) ² 149.3µ ± ∞ ¹ ~ (p=0.100 n=3) ² 854.9µ ± ∞ ¹ ~ (p=0.100 n=3) ²
Query/WindowParallel-16 236.8µ ± ∞ ¹ 314.5µ ± ∞ ¹ ~ (p=0.100 n=3) ² 122.1µ ± ∞ ¹ ~ (p=0.100 n=3) ² 915.8µ ± ∞ ¹ ~ (p=0.100 n=3) ² 1112.4µ ± ∞ ¹ ~ (p=0.100 n=3) ² 913.5µ ± ∞ ¹ ~ (p=0.100 n=3) ² 144.0µ ± ∞ ¹ ~ (p=0.100 n=3) ² 874.8µ ± ∞ ¹ ~ (p=0.100 n=3) ²
geomean 605.1µ 733.4µ +21.20% 287.5µ -52.48% 957.0µ +58.17% 956.3µ +58.05% 955.2µ +57.86% 323.1µ -46.60% 861.7µ +42.42%
¹ need >= 6 samples for confidence interval at level 0.95
² need >= 4 samples to detect a difference at alpha level 0.05
│ bench_ncruces_direct.txt │ bench_ncruces_driver.txt │ bench_eatonphil_direct.txt │ bench_glebarez_driver.txt │ bench_mattn_driver.txt │ bench_modernc_driver.txt │ bench_tailscale_driver.txt │ bench_zombiezen_direct.txt │
│ B/op │ B/op vs base │ B/op vs base │ B/op vs base │ B/op vs base │ B/op vs base │ B/op vs base │ B/op vs base │
Query/WindowParallel 0.0 ± ∞ ¹ 62762.0 ± ∞ ¹ ~ (p=0.100 n=3) ² 0.0 ± ∞ ¹ ~ (p=1.000 n=3) ³ 54775.0 ± ∞ ¹ ~ (p=0.100 n=3) ² 198936.0 ± ∞ ¹ ~ (p=0.100 n=3) ² 54789.0 ± ∞ ¹ ~ (p=0.100 n=3) ² 30792.0 ± ∞ ¹ ~ (p=0.100 n=3) ² 360.0 ± ∞ ¹ ~ (p=0.100 n=3) ²
Query/WindowParallel-2 0.0 ± ∞ ¹ 62761.0 ± ∞ ¹ ~ (p=0.100 n=3) ² 0.0 ± ∞ ¹ ~ (p=1.000 n=3) ³ 54831.0 ± ∞ ¹ ~ (p=0.100 n=3) ² 198863.0 ± ∞ ¹ ~ (p=0.100 n=3) ² 54815.0 ± ∞ ¹ ~ (p=0.100 n=3) ² 30793.0 ± ∞ ¹ ~ (p=0.100 n=3) ² 365.0 ± ∞ ¹ ~ (p=0.100 n=3) ²
Query/WindowParallel-4 0.0 ± ∞ ¹ 62761.0 ± ∞ ¹ ~ (p=0.100 n=3) ² 0.0 ± ∞ ¹ ~ (p=1.000 n=3) ³ 54882.0 ± ∞ ¹ ~ (p=0.100 n=3) ² 198853.0 ± ∞ ¹ ~ (p=0.100 n=3) ² 54883.0 ± ∞ ¹ ~ (p=0.100 n=3) ² 30794.0 ± ∞ ¹ ~ (p=0.100 n=3) ² 363.0 ± ∞ ¹ ~ (p=0.100 n=3) ²
Query/WindowParallel-8 0.0 ± ∞ ¹ 62763.0 ± ∞ ¹ ~ (p=0.100 n=3) ² 0.0 ± ∞ ¹ ~ (p=1.000 n=3) ³ 54884.0 ± ∞ ¹ ~ (p=0.100 n=3) ² 198906.0 ± ∞ ¹ ~ (p=0.100 n=3) ² 54904.0 ± ∞ ¹ ~ (p=0.100 n=3) ² 30793.0 ± ∞ ¹ ~ (p=0.100 n=3) ² 365.0 ± ∞ ¹ ~ (p=0.100 n=3) ²
Query/WindowParallel-16 1.000 ± ∞ ¹ 62766.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 1.000 ± ∞ ¹ ~ (p=1.000 n=3) ² 54894.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 198976.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 54917.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 30773.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 375.000 ± ∞ ¹ ~ (p=0.100 n=3) ²
geomean ⁴ 61.29Ki ? +0.00% ⁴ 53.57Ki ? 194.2Ki ? 53.58Ki ? 30.07Ki ? 365.6 ?
¹ need >= 6 samples for confidence interval at level 0.95
² need >= 4 samples to detect a difference at alpha level 0.05
³ all samples are equal
⁴ summaries must be >0 to compute geomean
│ bench_ncruces_direct.txt │ bench_ncruces_driver.txt │ bench_eatonphil_direct.txt │ bench_glebarez_driver.txt │ bench_mattn_driver.txt │ bench_modernc_driver.txt │ bench_tailscale_driver.txt │ bench_zombiezen_direct.txt │
│ allocs/op │ allocs/op vs base │ allocs/op vs base │ allocs/op vs base │ allocs/op vs base │ allocs/op vs base │ allocs/op vs base │ allocs/op vs base │
Query/WindowParallel 0.000 ± ∞ ¹ 3763.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 0.000 ± ∞ ¹ ~ (p=1.000 n=3) ³ 3765.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 4765.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 3765.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 1764.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 6.000 ± ∞ ¹ ~ (p=0.100 n=3) ²
Query/WindowParallel-2 0.000 ± ∞ ¹ 3763.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 0.000 ± ∞ ¹ ~ (p=1.000 n=3) ³ 3765.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 4764.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 3765.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 1764.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 6.000 ± ∞ ¹ ~ (p=0.100 n=3) ²
Query/WindowParallel-4 0.000 ± ∞ ¹ 3763.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 0.000 ± ∞ ¹ ~ (p=1.000 n=3) ³ 3766.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 4764.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 3765.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 1764.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 6.000 ± ∞ ¹ ~ (p=0.100 n=3) ²
Query/WindowParallel-8 0.000 ± ∞ ¹ 3763.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 0.000 ± ∞ ¹ ~ (p=1.000 n=3) ³ 3766.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 4764.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 3766.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 1764.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 6.000 ± ∞ ¹ ~ (p=0.100 n=3) ²
Query/WindowParallel-16 0.000 ± ∞ ¹ 3763.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 0.000 ± ∞ ¹ ~ (p=1.000 n=3) ³ 3766.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 4765.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 3766.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 1763.000 ± ∞ ¹ ~ (p=0.100 n=3) ² 6.000 ± ∞ ¹ ~ (p=0.100 n=3) ²
geomean ⁴ 3.763k ? +0.00% ⁴ 3.766k ? 4.764k ? 3.765k ? 1.764k ? 6.000 ?
¹ need >= 6 samples for confidence interval at level 0.95
² need >= 4 samples to detect a difference at alpha level 0.05
³ all samples are equal
⁴ summaries must be >0 to compute geomean
Please note that the benchmark framework sets the busy_timeout, foreign_keys, journal_mode and synchronous pragmas; what you're seeing below isn't their default values.
ATOMIC_INTRINSICS=1
COMPILER=clang-17.0.0
DEFAULT_AUTOVACUUM
DEFAULT_CACHE_SIZE=-2000
DEFAULT_FILE_FORMAT=4
DEFAULT_JOURNAL_SIZE_LIMIT=-1
DEFAULT_MEMSTATUS=0
DEFAULT_MMAP_SIZE=0
DEFAULT_PAGE_SIZE=4096
DEFAULT_PCACHE_INITSZ=20
DEFAULT_RECURSIVE_TRIGGERS
DEFAULT_SECTOR_SIZE=4096
DEFAULT_SYNCHRONOUS=2
DEFAULT_WAL_AUTOCHECKPOINT=1000
DEFAULT_WAL_SYNCHRONOUS=2
DEFAULT_WORKER_THREADS=0
DIRECT_OVERFLOW_READ
ENABLE_API_ARMOR
ENABLE_FTS3
ENABLE_FTS3_PARENTHESIS
ENABLE_FTS4
ENABLE_FTS5
ENABLE_GEOPOLY
ENABLE_PREUPDATE_HOOK
ENABLE_RTREE
ENABLE_SESSION
ENABLE_STAT4
ENABLE_UNLOCK_NOTIFY
ENABLE_UPDATE_DELETE_LIMIT
MALLOC_SOFT_LIMIT=1024
MAX_ATTACHED=10
MAX_COLUMN=2000
MAX_COMPOUND_SELECT=500
MAX_DEFAULT_PAGE_SIZE=8192
MAX_EXPR_DEPTH=1000
MAX_FUNCTION_ARG=127
MAX_LENGTH=1000000000
MAX_LIKE_PATTERN_LENGTH=50000
MAX_MMAP_SIZE=0x7fff0000
MAX_PAGE_COUNT=0xfffffffe
MAX_PAGE_SIZE=65536
MAX_SQL_LENGTH=1000000000
MAX_TRIGGER_DEPTH=1000
MAX_VARIABLE_NUMBER=32766
MAX_VDBE_OP=250000000
MAX_WORKER_THREADS=8
MUTEX_PTHREADS
OMIT_AUTOINIT
OMIT_DEPRECATED
OMIT_LOAD_EXTENSION
OMIT_PROGRESS_CALLBACK
OMIT_UTF16
SOUNDEX
SYSTEM_MALLOC
TEMP_STORE=2
THREADSAFE=2
USE_ALLOCA
USE_URI
auto_vacuum=0
automatic_index=1
busy_timeout=5000
cache_size=-2000
cache_spill=483
cell_size_check=0
checkpoint_fullfsync=0
defer_foreign_keys=0
foreign_keys=1
fullfsync=0
hard_heap_limit=0
journal_mode=wal
journal_size_limit=-1
locking_mode=normal
mmap_size=0
page_size=4096
query_only=0
read_uncommitted=0
recursive_triggers=0
reverse_unordered_selects=0
secure_delete=0
soft_heap_limit=0
synchronous=1
temp_store=0
threads=0
wal_autocheckpoint=1000
3.46.0
-
Configuration
glebarez uses some additional configuration:
_time_format=sqlite(default is none)_txlock=immediate(default isdeferred)
-
Notes
If a transaction is
ReadOnlyit's alwaysdeferred. If it isn't, it uses the value of_txlock.
ATOMIC_INTRINSICS=0
COMPILER=clang-16.0.0
DEFAULT_AUTOVACUUM
DEFAULT_CACHE_SIZE=-2000
DEFAULT_FILE_FORMAT=4
DEFAULT_JOURNAL_SIZE_LIMIT=-1
DEFAULT_MEMSTATUS=0
DEFAULT_MMAP_SIZE=0
DEFAULT_PAGE_SIZE=4096
DEFAULT_PCACHE_INITSZ=20
DEFAULT_RECURSIVE_TRIGGERS
DEFAULT_SECTOR_SIZE=4096
DEFAULT_SYNCHRONOUS=2
DEFAULT_WAL_AUTOCHECKPOINT=1000
DEFAULT_WAL_SYNCHRONOUS=2
DEFAULT_WORKER_THREADS=0
DIRECT_OVERFLOW_READ
ENABLE_COLUMN_METADATA
ENABLE_DBSTAT_VTAB
ENABLE_FTS5
ENABLE_GEOPOLY
ENABLE_MATH_FUNCTIONS
ENABLE_MEMORY_MANAGEMENT
ENABLE_OFFSET_SQL_FUNC
ENABLE_PREUPDATE_HOOK
ENABLE_RBU
ENABLE_RTREE
ENABLE_SESSION
ENABLE_SNAPSHOT
ENABLE_STAT4
ENABLE_UNLOCK_NOTIFY
LIKE_DOESNT_MATCH_BLOBS
MALLOC_SOFT_LIMIT=1024
MAX_ATTACHED=10
MAX_COLUMN=2000
MAX_COMPOUND_SELECT=500
MAX_DEFAULT_PAGE_SIZE=8192
MAX_EXPR_DEPTH=1000
MAX_FUNCTION_ARG=1000
MAX_LENGTH=1000000000
MAX_LIKE_PATTERN_LENGTH=50000
MAX_MMAP_SIZE=0x7fff0000
MAX_PAGE_COUNT=0xfffffffe
MAX_PAGE_SIZE=65536
MAX_SQL_LENGTH=1000000000
MAX_TRIGGER_DEPTH=1000
MAX_VARIABLE_NUMBER=32766
MAX_VDBE_OP=250000000
MAX_WORKER_THREADS=8
MUTEX_NOOP
SOUNDEX
SYSTEM_MALLOC
TEMP_STORE=1
THREADSAFE=1
auto_vacuum=0
automatic_index=1
busy_timeout=5000
cache_size=-2000
cache_spill=483
cell_size_check=0
checkpoint_fullfsync=0
defer_foreign_keys=0
encoding=UTF-8
foreign_keys=1
fullfsync=0
hard_heap_limit=0
journal_mode=wal
journal_size_limit=-1
locking_mode=normal
mmap_size=0
page_size=4096
query_only=0
read_uncommitted=0
recursive_triggers=0
reverse_unordered_selects=0
secure_delete=0
soft_heap_limit=0
synchronous=1
temp_store=0
threads=0
wal_autocheckpoint=1000
3.49.1
-
Configuration
mattn uses some additional configuration:
_mutex=no(default isfull)
-
Notes
mattn doesn't currently support read and write transactions on the same connection because it ignores
TxOptionscompletely. Instead, mattn uses the value of_txlockwhenever it starts a transaction. If you change the value of_txlocktoimmediateyou make all transactions write transactions. If you leave the value of_txlockat its default,deferred, you risk immediateSQLITE_BUSYerrors when a read transaction is upgraded to a write transaction. See more here.
ATOMIC_INTRINSICS=1
COMPILER=clang-17.0.0
DEFAULT_AUTOVACUUM
DEFAULT_CACHE_SIZE=-2000
DEFAULT_FILE_FORMAT=4
DEFAULT_JOURNAL_SIZE_LIMIT=-1
DEFAULT_MMAP_SIZE=0
DEFAULT_PAGE_SIZE=4096
DEFAULT_PCACHE_INITSZ=20
DEFAULT_RECURSIVE_TRIGGERS
DEFAULT_SECTOR_SIZE=4096
DEFAULT_SYNCHRONOUS=2
DEFAULT_WAL_AUTOCHECKPOINT=1000
DEFAULT_WAL_SYNCHRONOUS=1
DEFAULT_WORKER_THREADS=0
DIRECT_OVERFLOW_READ
ENABLE_FTS3
ENABLE_FTS3_PARENTHESIS
ENABLE_RTREE
ENABLE_UPDATE_DELETE_LIMIT
MALLOC_SOFT_LIMIT=1024
MAX_ATTACHED=10
MAX_COLUMN=2000
MAX_COMPOUND_SELECT=500
MAX_DEFAULT_PAGE_SIZE=8192
MAX_EXPR_DEPTH=1000
MAX_FUNCTION_ARG=1000
MAX_LENGTH=1000000000
MAX_LIKE_PATTERN_LENGTH=50000
MAX_MMAP_SIZE=0x7fff0000
MAX_PAGE_COUNT=0xfffffffe
MAX_PAGE_SIZE=65536
MAX_SQL_LENGTH=1000000000
MAX_TRIGGER_DEPTH=1000
MAX_VARIABLE_NUMBER=32766
MAX_VDBE_OP=250000000
MAX_WORKER_THREADS=8
MUTEX_PTHREADS
OMIT_DEPRECATED
SYSTEM_MALLOC
TEMP_STORE=1
THREADSAFE=1
auto_vacuum=0
automatic_index=1
busy_timeout=5000
cache_size=-2000
cache_spill=483
cell_size_check=0
checkpoint_fullfsync=0
defer_foreign_keys=0
encoding=UTF-8
foreign_keys=1
fullfsync=0
hard_heap_limit=0
journal_mode=wal
journal_size_limit=-1
locking_mode=normal
mmap_size=0
page_size=4096
query_only=0
read_uncommitted=0
recursive_triggers=0
reverse_unordered_selects=0
secure_delete=0
soft_heap_limit=0
synchronous=1
temp_store=0
threads=0
wal_autocheckpoint=1000
3.49.1
-
Configuration
modernc uses some additional configuration:
_time_format=sqlite(default is none)_txlock=immediate(default isdeferred)
-
Notes
If a transaction is
ReadOnlyit's alwaysdeferred. If it isn't, it uses the value of_txlock.
ATOMIC_INTRINSICS=0
COMPILER=clang-16.0.0
DEFAULT_AUTOVACUUM
DEFAULT_CACHE_SIZE=-2000
DEFAULT_FILE_FORMAT=4
DEFAULT_JOURNAL_SIZE_LIMIT=-1
DEFAULT_MEMSTATUS=0
DEFAULT_MMAP_SIZE=0
DEFAULT_PAGE_SIZE=4096
DEFAULT_PCACHE_INITSZ=20
DEFAULT_RECURSIVE_TRIGGERS
DEFAULT_SECTOR_SIZE=4096
DEFAULT_SYNCHRONOUS=2
DEFAULT_WAL_AUTOCHECKPOINT=1000
DEFAULT_WAL_SYNCHRONOUS=2
DEFAULT_WORKER_THREADS=0
DIRECT_OVERFLOW_READ
ENABLE_COLUMN_METADATA
ENABLE_DBSTAT_VTAB
ENABLE_FTS5
ENABLE_GEOPOLY
ENABLE_MATH_FUNCTIONS
ENABLE_MEMORY_MANAGEMENT
ENABLE_OFFSET_SQL_FUNC
ENABLE_PREUPDATE_HOOK
ENABLE_RBU
ENABLE_RTREE
ENABLE_SESSION
ENABLE_SNAPSHOT
ENABLE_STAT4
ENABLE_UNLOCK_NOTIFY
LIKE_DOESNT_MATCH_BLOBS
MALLOC_SOFT_LIMIT=1024
MAX_ATTACHED=10
MAX_COLUMN=2000
MAX_COMPOUND_SELECT=500
MAX_DEFAULT_PAGE_SIZE=8192
MAX_EXPR_DEPTH=1000
MAX_FUNCTION_ARG=1000
MAX_LENGTH=1000000000
MAX_LIKE_PATTERN_LENGTH=50000
MAX_MMAP_SIZE=0x7fff0000
MAX_PAGE_COUNT=0xfffffffe
MAX_PAGE_SIZE=65536
MAX_SQL_LENGTH=1000000000
MAX_TRIGGER_DEPTH=1000
MAX_VARIABLE_NUMBER=32766
MAX_VDBE_OP=250000000
MAX_WORKER_THREADS=8
MUTEX_NOOP
SOUNDEX
SYSTEM_MALLOC
TEMP_STORE=1
THREADSAFE=1
auto_vacuum=0
automatic_index=1
busy_timeout=5000
cache_size=-2000
cache_spill=483
cell_size_check=0
checkpoint_fullfsync=0
defer_foreign_keys=0
encoding=UTF-8
foreign_keys=1
fullfsync=0
hard_heap_limit=0
journal_mode=wal
journal_size_limit=-1
locking_mode=normal
mmap_size=0
page_size=4096
query_only=0
read_uncommitted=0
recursive_triggers=0
reverse_unordered_selects=0
secure_delete=0
soft_heap_limit=0
synchronous=1
temp_store=0
threads=0
wal_autocheckpoint=1000
3.49.1
ALLOW_URI_AUTHORITY
ATOMIC_INTRINSICS=1
COMPILER=clang-19.1.5
DEFAULT_AUTOVACUUM
DEFAULT_CACHE_SIZE=-2000
DEFAULT_FILE_FORMAT=4
DEFAULT_FOREIGN_KEYS
DEFAULT_JOURNAL_SIZE_LIMIT=-1
DEFAULT_MMAP_SIZE=0
DEFAULT_PAGE_SIZE=4096
DEFAULT_PCACHE_INITSZ=20
DEFAULT_RECURSIVE_TRIGGERS
DEFAULT_SECTOR_SIZE=4096
DEFAULT_SYNCHRONOUS=2
DEFAULT_WAL_AUTOCHECKPOINT=1000
DEFAULT_WAL_SYNCHRONOUS=1
DEFAULT_WORKER_THREADS=0
DIRECT_OVERFLOW_READ
DQS=0
ENABLE_API_ARMOR
ENABLE_ATOMIC_WRITE
ENABLE_BATCH_ATOMIC_WRITE
ENABLE_COLUMN_METADATA
ENABLE_FTS5
ENABLE_GEOPOLY
ENABLE_MATH_FUNCTIONS
ENABLE_RTREE
ENABLE_STAT4
HAVE_ISNAN
LIKE_DOESNT_MATCH_BLOBS
MALLOC_SOFT_LIMIT=1024
MAX_ATTACHED=10
MAX_COLUMN=2000
MAX_COMPOUND_SELECT=500
MAX_DEFAULT_PAGE_SIZE=8192
MAX_EXPR_DEPTH=1000
MAX_FUNCTION_ARG=1000
MAX_LENGTH=1000000000
MAX_LIKE_PATTERN_LENGTH=50000
MAX_MMAP_SIZE=0
MAX_PAGE_COUNT=0xfffffffe
MAX_PAGE_SIZE=65536
MAX_SQL_LENGTH=1000000000
MAX_TRIGGER_DEPTH=1000
MAX_VARIABLE_NUMBER=32766
MAX_VDBE_OP=250000000
MAX_WORKER_THREADS=0
MUTEX_OMIT
OMIT_AUTOINIT
OMIT_DEPRECATED
OMIT_DESERIALIZE
OMIT_LOAD_EXTENSION
OMIT_SHARED_CACHE
SOUNDEX
SYSTEM_MALLOC
TEMP_STORE=1
THREADSAFE=0
UNTESTABLE
auto_vacuum=0
automatic_index=1
busy_timeout=5000
cache_size=-2000
cache_spill=489
cell_size_check=0
checkpoint_fullfsync=0
defer_foreign_keys=0
encoding=UTF-8
foreign_keys=1
fullfsync=0
hard_heap_limit=0
journal_mode=wal
journal_size_limit=-1
locking_mode=normal
mmap_size=0
page_size=4096
query_only=0
read_uncommitted=0
recursive_triggers=0
reverse_unordered_selects=0
secure_delete=0
soft_heap_limit=0
synchronous=1
temp_store=0
threads=0
wal_autocheckpoint=1000
3.49.1
ATOMIC_INTRINSICS=1
COMPILER=clang-17.0.0
DEFAULT_AUTOVACUUM
DEFAULT_CACHE_SIZE=-2000
DEFAULT_FILE_FORMAT=4
DEFAULT_JOURNAL_SIZE_LIMIT=-1
DEFAULT_MEMSTATUS=0
DEFAULT_MMAP_SIZE=0
DEFAULT_PAGE_SIZE=4096
DEFAULT_PCACHE_INITSZ=20
DEFAULT_RECURSIVE_TRIGGERS
DEFAULT_SECTOR_SIZE=4096
DEFAULT_SYNCHRONOUS=2
DEFAULT_WAL_AUTOCHECKPOINT=1000
DEFAULT_WAL_SYNCHRONOUS=1
DEFAULT_WORKER_THREADS=0
DIRECT_OVERFLOW_READ
DQS=0
ENABLE_COLUMN_METADATA
ENABLE_DBSTAT_VTAB
ENABLE_FTS5
ENABLE_PREUPDATE_HOOK
ENABLE_RTREE
ENABLE_SESSION
ENABLE_SNAPSHOT
ENABLE_STAT4
LIKE_DOESNT_MATCH_BLOBS
MALLOC_SOFT_LIMIT=1024
MAX_ATTACHED=10
MAX_COLUMN=2000
MAX_COMPOUND_SELECT=500
MAX_DEFAULT_PAGE_SIZE=8192
MAX_EXPR_DEPTH=0
MAX_FUNCTION_ARG=127
MAX_LENGTH=1000000000
MAX_LIKE_PATTERN_LENGTH=50000
MAX_MMAP_SIZE=0x7fff0000
MAX_PAGE_COUNT=0xfffffffe
MAX_PAGE_SIZE=65536
MAX_SQL_LENGTH=1000000000
MAX_TRIGGER_DEPTH=1000
MAX_VARIABLE_NUMBER=32766
MAX_VDBE_OP=250000000
MAX_WORKER_THREADS=8
MUTEX_PTHREADS
OMIT_AUTOINIT
OMIT_DEPRECATED
OMIT_LOAD_EXTENSION
OMIT_PROGRESS_CALLBACK
OMIT_SHARED_CACHE
SYSTEM_MALLOC
TEMP_STORE=1
THREADSAFE=2
USE_ALLOCA
auto_vacuum=0
automatic_index=1
busy_timeout=5000
cache_size=-2000
cache_spill=483
cell_size_check=0
checkpoint_fullfsync=0
defer_foreign_keys=0
encoding=UTF-8
foreign_keys=1
fullfsync=0
hard_heap_limit=0
journal_mode=wal
journal_size_limit=-1
locking_mode=normal
mmap_size=0
page_size=4096
query_only=0
read_uncommitted=0
recursive_triggers=0
reverse_unordered_selects=0
secure_delete=0
soft_heap_limit=0
synchronous=1
temp_store=0
threads=0
wal_autocheckpoint=1000
3.46.1
ATOMIC_INTRINSICS=0
COMPILER=clang-16.0.0
DEFAULT_AUTOVACUUM
DEFAULT_CACHE_SIZE=-2000
DEFAULT_FILE_FORMAT=4
DEFAULT_JOURNAL_SIZE_LIMIT=-1
DEFAULT_MEMSTATUS=0
DEFAULT_MMAP_SIZE=0
DEFAULT_PAGE_SIZE=4096
DEFAULT_PCACHE_INITSZ=20
DEFAULT_RECURSIVE_TRIGGERS
DEFAULT_SECTOR_SIZE=4096
DEFAULT_SYNCHRONOUS=2
DEFAULT_WAL_AUTOCHECKPOINT=1000
DEFAULT_WAL_SYNCHRONOUS=2
DEFAULT_WORKER_THREADS=0
DIRECT_OVERFLOW_READ
ENABLE_COLUMN_METADATA
ENABLE_DBSTAT_VTAB
ENABLE_FTS5
ENABLE_GEOPOLY
ENABLE_MATH_FUNCTIONS
ENABLE_MEMORY_MANAGEMENT
ENABLE_OFFSET_SQL_FUNC
ENABLE_PREUPDATE_HOOK
ENABLE_RBU
ENABLE_RTREE
ENABLE_SESSION
ENABLE_SNAPSHOT
ENABLE_STAT4
ENABLE_UNLOCK_NOTIFY
LIKE_DOESNT_MATCH_BLOBS
MALLOC_SOFT_LIMIT=1024
MAX_ATTACHED=10
MAX_COLUMN=2000
MAX_COMPOUND_SELECT=500
MAX_DEFAULT_PAGE_SIZE=8192
MAX_EXPR_DEPTH=1000
MAX_FUNCTION_ARG=1000
MAX_LENGTH=1000000000
MAX_LIKE_PATTERN_LENGTH=50000
MAX_MMAP_SIZE=0x7fff0000
MAX_PAGE_COUNT=0xfffffffe
MAX_PAGE_SIZE=65536
MAX_SQL_LENGTH=1000000000
MAX_TRIGGER_DEPTH=1000
MAX_VARIABLE_NUMBER=32766
MAX_VDBE_OP=250000000
MAX_WORKER_THREADS=8
MUTEX_NOOP
SOUNDEX
SYSTEM_MALLOC
TEMP_STORE=1
THREADSAFE=1
auto_vacuum=0
automatic_index=1
busy_timeout=5000
cache_size=-2000
cache_spill=483
cell_size_check=0
checkpoint_fullfsync=0
defer_foreign_keys=0
encoding=UTF-8
foreign_keys=1
fullfsync=0
hard_heap_limit=0
journal_mode=wal
journal_size_limit=-1
locking_mode=normal
mmap_size=0
page_size=4096
query_only=0
read_uncommitted=0
recursive_triggers=0
reverse_unordered_selects=0
secure_delete=0
soft_heap_limit=0
synchronous=1
temp_store=0
threads=0
wal_autocheckpoint=1000
3.49.1