Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

[Segment Replication] Prevent store read for clean up post segment replication and reader manager close operation #7044

Closed

Conversation

dreamer-89
Copy link
Member

Description

This PR is an improvement on changes introduced in #6765, where clean up operations relies on expensive disk reads (1. For loading store metadata map for in memory segment infos, 2. read latest committed segment infos). It introduces a new utility in store which allows clean up based on reference counts. This PR changes:

  1. OpenSearchReaderManager. During reader manager close operation, linked OpenSearchDirectoryReader references are closed where a clean up is performed for segment files whose reference count is 0. This change prevent this by simply relying on reference count.
  2. NRTReplicationEngine. It uses idea above and uses reference count on committed segment infos on replica engine and prevents un-necessary read operations. This comes in path of updating segment infos on replica.
  3. Revert changes in store.cleanupAndPreserveLatestCommitPoint added as part of [Segment Replication] Add PIT/Scroll compatibility with Segment Replication #6644 #6765

Issues Resolved

Resolves #7043

Check List

  • New functionality includes testing.
    • All tests pass
  • New functionality has been documented.
    • New functionality has javadoc added
  • Commits are signed per the DCO using --signoff
  • Commit changes are listed out in CHANGELOG.md file (See: Changelog)

By submitting this pull request, I confirm that my contribution is made under the terms of the Apache 2.0 license.
For more information on following Developer Certificate of Origin and signing off your commits, please check here.

@github-actions
Copy link
Contributor

github-actions bot commented Apr 7, 2023

Gradle Check (Jenkins) Run Completed with:

@github-actions
Copy link
Contributor

Gradle Check (Jenkins) Run Completed with:

@codecov-commenter

This comment was marked as outdated.

@github-actions

This comment was marked as outdated.

@github-actions

This comment was marked as outdated.

@github-actions

This comment was marked as outdated.

…plication and reader manager close operation

Signed-off-by: Suraj Singh <surajrider@gmail.com>
…icaTracker

Signed-off-by: Suraj Singh <surajrider@gmail.com>
…deletion

Signed-off-by: Suraj Singh <surajrider@gmail.com>
…ed file deletion"

This reverts commit b338828.

Signed-off-by: Suraj Singh <surajrider@gmail.com>
…used cleanupAndPreserveLatestCommitPoint

Signed-off-by: Suraj Singh <surajrider@gmail.com>
…ition

Signed-off-by: Suraj Singh <surajrider@gmail.com>
Signed-off-by: Suraj Singh <surajrider@gmail.com>
Signed-off-by: Suraj Singh <surajrider@gmail.com>
@github-actions

This comment was marked as outdated.

@github-actions

This comment was marked as outdated.

@github-actions

This comment was marked as outdated.

@github-actions
Copy link
Contributor

Gradle Check (Jenkins) Run Completed with:

@dreamer-89
Copy link
Member Author

dreamer-89 commented Apr 12, 2023

Performed benchmark and see perf degradation.

Settings

ubuntu@ip-172-31-49-145:~/OpenSearch-Benchmark$ cat ~/.benchmark/segrep_index_settings.json
{
    "index_settings":{
        "number_of_replicas": 1,
            "number_of_shards": 12,
              "index":{
                  "replication": {"type": "SEGMENT"}
                }
     }
}

Result

Baseline: main
Contender: main with changes in this PR.

ubuntu@ip-172-31-49-145:~/OpenSearch-Benchmark$ opensearch-benchmark compare --baseline=fa68fbe5-0cc4-4866-8eee-3abe803bc409  --contender=5bbecedc-6a9a-4712-8456-b25e2b366283

   ____                  _____                      __       ____                  __                         __
  / __ \____  ___  ____ / ___/___  ____ ___________/ /_     / __ )___  ____  _____/ /_  ____ ___  ____ ______/ /__
 / / / / __ \/ _ \/ __ \\__ \/ _ \/ __ `/ ___/ ___/ __ \   / __  / _ \/ __ \/ ___/ __ \/ __ `__ \/ __ `/ ___/ //_/
/ /_/ / /_/ /  __/ / / /__/ /  __/ /_/ / /  / /__/ / / /  / /_/ /  __/ / / / /__/ / / / / / / / / /_/ / /  / ,<
\____/ .___/\___/_/ /_/____/\___/\__,_/_/   \___/_/ /_/  /_____/\___/_/ /_/\___/_/ /_/_/ /_/ /_/\__,_/_/  /_/|_|
    /_/


Comparing baseline
  TestExecution ID: fa68fbe5-0cc4-4866-8eee-3abe803bc409
  TestExecution timestamp: 2023-04-12 01:30:59
  TestProcedure: append-no-conflicts
  ProvisionConfigInstance: external

with contender
  TestExecution ID: 5bbecedc-6a9a-4712-8456-b25e2b366283
  TestExecution timestamp: 2023-04-12 04:11:51
  TestProcedure: append-no-conflicts
  ProvisionConfigInstance: external

------------------------------------------------------
    _______             __   _____
   / ____(_)___  ____ _/ /  / ___/_________  ________
  / /_  / / __ \/ __ `/ /   \__ \/ ___/ __ \/ ___/ _ \
 / __/ / / / / / /_/ / /   ___/ / /__/ /_/ / /  /  __/
/_/   /_/_/ /_/\__,_/_/   /____/\___/\____/_/   \___/
------------------------------------------------------

|                                                        Metric |                     Task |    Baseline |   Contender |     Diff |   Unit |
|--------------------------------------------------------------:|-------------------------:|------------:|------------:|---------:|-------:|
|                    Cumulative indexing time of primary shards |                          |     171.178 |     168.727 | -2.45072 |    min |
|             Min cumulative indexing time across primary shard |                          |     12.6983 |     11.4922 | -1.20615 |    min |
|          Median cumulative indexing time across primary shard |                          |     13.8645 |     12.0608 | -1.80367 |    min |
|             Max cumulative indexing time across primary shard |                          |     16.1073 |     18.4122 |  2.30487 |    min |
|           Cumulative indexing throttle time of primary shards |                          |           0 |           0 |        0 |    min |
|    Min cumulative indexing throttle time across primary shard |                          |           0 |           0 |        0 |    min |
| Median cumulative indexing throttle time across primary shard |                          |           0 |           0 |        0 |    min |
|    Max cumulative indexing throttle time across primary shard |                          |           0 |           0 |        0 |    min |
|                       Cumulative merge time of primary shards |                          |     167.214 |     184.826 |  17.6119 |    min |
|                      Cumulative merge count of primary shards |                          |        2132 |        2398 |      266 |        |
|                Min cumulative merge time across primary shard |                          |     11.8232 |     10.3992 | -1.42402 |    min |
|             Median cumulative merge time across primary shard |                          |     13.3929 |     15.7978 |   2.4049 |    min |
|                Max cumulative merge time across primary shard |                          |     16.4998 |     16.9185 |  0.41875 |    min |
|              Cumulative merge throttle time of primary shards |                          |     77.5246 |     94.3867 |  16.8621 |    min |
|       Min cumulative merge throttle time across primary shard |                          |     5.23053 |     3.65005 | -1.58048 |    min |
|    Median cumulative merge throttle time across primary shard |                          |     6.35068 |       8.379 |  2.02832 |    min |
|       Max cumulative merge throttle time across primary shard |                          |     8.69047 |     9.42582 |  0.73535 |    min |
|                     Cumulative refresh time of primary shards |                          |     31.4461 |      32.264 |  0.81793 |    min |
|                    Cumulative refresh count of primary shards |                          |       11092 |       12494 |     1402 |        |
|              Min cumulative refresh time across primary shard |                          |       2.393 |     2.33898 | -0.05402 |    min |
|           Median cumulative refresh time across primary shard |                          |     2.68242 |     2.75084 |  0.06842 |    min |
|              Max cumulative refresh time across primary shard |                          |     2.76505 |     2.99582 |  0.23077 |    min |
|                       Cumulative flush time of primary shards |                          |     2.37038 |     2.32432 | -0.04607 |    min |
|                      Cumulative flush count of primary shards |                          |         193 |         193 |        0 |        |
|                Min cumulative flush time across primary shard |                          |    0.120167 |   0.0780833 | -0.04208 |    min |
|             Median cumulative flush time across primary shard |                          |    0.192258 |    0.168575 | -0.02368 |    min |
|                Max cumulative flush time across primary shard |                          |    0.290083 |    0.338033 |  0.04795 |    min |
|                                       Total Young Gen GC time |                          |     267.903 |     254.241 |  -13.662 |      s |
|                                      Total Young Gen GC count |                          |       21227 |       21255 |       28 |        |
|                                         Total Old Gen GC time |                          |           0 |           0 |        0 |      s |
|                                        Total Old Gen GC count |                          |           0 |           0 |        0 |        |
|                                                    Store size |                          |     66.0178 |     66.0042 | -0.01357 |     GB |
|                                                 Translog size |                          | 1.22935e-06 | 1.22935e-06 |        0 |     GB |
|                                        Heap used for segments |                          |           0 |           0 |        0 |     MB |
|                                      Heap used for doc values |                          |           0 |           0 |        0 |     MB |
|                                           Heap used for terms |                          |           0 |           0 |        0 |     MB |
|                                           Heap used for norms |                          |           0 |           0 |        0 |     MB |
|                                          Heap used for points |                          |           0 |           0 |        0 |     MB |
|                                   Heap used for stored fields |                          |           0 |           0 |        0 |     MB |
|                                                 Segment count |                          |         316 |         318 |        2 |        |
|                                                Min Throughput |                    index |     82910.7 |     77548.6 | -5362.11 | docs/s |
|                                               Mean Throughput |                    index |       85943 |       80066 | -5877.04 | docs/s |
|                                             Median Throughput |                    index |     86303.5 |     80238.9 | -6064.62 | docs/s |
|                                                Max Throughput |                    index |       88506 |     80851.1 | -7654.87 | docs/s |
|                                       50th percentile latency |                    index |     936.815 |     986.197 |  49.3819 |     ms |
|                                       90th percentile latency |                    index |     1163.04 |     1213.48 |  50.4387 |     ms |
|                                       99th percentile latency |                    index |     1396.33 |     1487.16 |  90.8326 |     ms |
|                                     99.9th percentile latency |                    index |     1723.17 |     1817.92 |  94.7595 |     ms |
|                                    99.99th percentile latency |                    index |     1992.53 |     2349.42 |  356.886 |     ms |
|                                      100th percentile latency |                    index |     2170.53 |     2756.79 |  586.258 |     ms |
|                                  50th percentile service time |                    index |     936.815 |     986.197 |  49.3819 |     ms |
|                                  90th percentile service time |                    index |     1163.04 |     1213.48 |  50.4387 |     ms |
|                                  99th percentile service time |                    index |     1396.33 |     1487.16 |  90.8326 |     ms |
|                                99.9th percentile service time |                    index |     1723.17 |     1817.92 |  94.7595 |     ms |
|                               99.99th percentile service time |                    index |     1992.53 |     2349.42 |  356.886 |     ms |
|                                 100th percentile service time |                    index |     2170.53 |     2756.79 |  586.258 |     ms |
|                                                    error rate |                    index |           0 |           0 |        0 |      % |
|                                                Min Throughput | wait-until-merges-finish |  0.00715532 |  0.00575599 |  -0.0014 |  ops/s |
|                                               Mean Throughput | wait-until-merges-finish |  0.00715532 |  0.00575599 |  -0.0014 |  ops/s |
|                                             Median Throughput | wait-until-merges-finish |  0.00715532 |  0.00575599 |  -0.0014 |  ops/s |
|                                                Max Throughput | wait-until-merges-finish |  0.00715532 |  0.00575599 |  -0.0014 |  ops/s |
|                                      100th percentile latency | wait-until-merges-finish |      139756 |      173732 |  33975.8 |     ms |
|                                 100th percentile service time | wait-until-merges-finish |      139756 |      173732 |  33975.8 |     ms |
|                                                    error rate | wait-until-merges-finish |           0 |           0 |        0 |      % |
|                                                Min Throughput |                  default |     3.00885 |     3.00838 | -0.00047 |  ops/s |
|                                               Mean Throughput |                  default |     3.01437 |     3.01358 | -0.00079 |  ops/s |
|                                             Median Throughput |                  default |      3.0131 |     3.01233 | -0.00077 |  ops/s |
|                                                Max Throughput |                  default |     3.02531 |     3.02396 | -0.00135 |  ops/s |
|                                       50th percentile latency |                  default |     7.97209 |     7.09264 | -0.87945 |     ms |
|                                       90th percentile latency |                  default |     9.18624 |     8.12216 | -1.06408 |     ms |
|                                       99th percentile latency |                  default |     11.0728 |     9.42169 | -1.65113 |     ms |
|                                      100th percentile latency |                  default |     11.7656 |      10.445 | -1.32055 |     ms |
|                                  50th percentile service time |                  default |     6.60254 |     5.70597 | -0.89657 |     ms |
|                                  90th percentile service time |                  default |     8.00979 |      6.6891 | -1.32068 |     ms |
|                                  99th percentile service time |                  default |     10.0115 |     8.04105 | -1.97046 |     ms |
|                                 100th percentile service time |                  default |     10.2528 |     9.00918 | -1.24363 |     ms |
|                                                    error rate |                  default |           0 |           0 |        0 |      % |
|                                                Min Throughput |                    range |    0.703602 |      0.7037 |   0.0001 |  ops/s |
|                                               Mean Throughput |                    range |    0.705924 |    0.706091 |  0.00017 |  ops/s |
|                                             Median Throughput |                    range |    0.705385 |     0.70554 |  0.00015 |  ops/s |
|                                                Max Throughput |                    range |     0.71069 |    0.711007 |  0.00032 |  ops/s |
|                                       50th percentile latency |                    range |     30.9505 |     33.3392 |  2.38874 |     ms |
|                                       90th percentile latency |                    range |     38.9906 |     36.9521 | -2.03844 |     ms |
|                                       99th percentile latency |                    range |     44.7271 |      39.734 | -4.99306 |     ms |
|                                      100th percentile latency |                    range |     45.9572 |     51.1524 |  5.19518 |     ms |
|                                  50th percentile service time |                    range |     27.8729 |     30.1123 |  2.23948 |     ms |
|                                  90th percentile service time |                    range |      36.196 |     34.0706 | -2.12545 |     ms |
|                                  99th percentile service time |                    range |     41.3931 |     37.6931 | -3.69994 |     ms |
|                                 100th percentile service time |                    range |     42.5744 |     47.2592 |  4.68481 |     ms |
|                                                    error rate |                    range |           0 |           0 |        0 |      % |
|                                                Min Throughput |      distance_amount_agg |     2.00837 |     2.00556 | -0.00281 |  ops/s |
|                                               Mean Throughput |      distance_amount_agg |     2.01375 |     2.00911 | -0.00464 |  ops/s |
|                                             Median Throughput |      distance_amount_agg |      2.0125 |     2.00829 |  -0.0042 |  ops/s |
|                                                Max Throughput |      distance_amount_agg |     2.02466 |     2.01632 | -0.00834 |  ops/s |
|                                       50th percentile latency |      distance_amount_agg |     6.84576 |     6.87393 |  0.02817 |     ms |
|                                       90th percentile latency |      distance_amount_agg |     7.59578 |     7.57937 | -0.01641 |     ms |
|                                       99th percentile latency |      distance_amount_agg |     10.1581 |      9.7171 | -0.44097 |     ms |
|                                      100th percentile latency |      distance_amount_agg |     17.5769 |     17.4243 |  -0.1526 |     ms |
|                                  50th percentile service time |      distance_amount_agg |     5.34751 |     5.31465 | -0.03286 |     ms |
|                                  90th percentile service time |      distance_amount_agg |      6.0108 |     5.82383 | -0.18697 |     ms |
|                                  99th percentile service time |      distance_amount_agg |     8.48351 |     7.70381 |  -0.7797 |     ms |
|                                 100th percentile service time |      distance_amount_agg |     15.7113 |     16.0232 |   0.3119 |     ms |
|                                                    error rate |      distance_amount_agg |           0 |           0 |        0 |      % |
|                                                Min Throughput |            autohisto_agg |     1.49635 |     1.49943 |  0.00307 |  ops/s |
|                                               Mean Throughput |            autohisto_agg |     1.49797 |     1.49967 |   0.0017 |  ops/s |
|                                             Median Throughput |            autohisto_agg |     1.49812 |      1.4997 |  0.00158 |  ops/s |
|                                                Max Throughput |            autohisto_agg |     1.49876 |      1.4998 |  0.00104 |  ops/s |
|                                       50th percentile latency |            autohisto_agg |     78.1529 |     78.6508 |  0.49792 |     ms |
|                                       90th percentile latency |            autohisto_agg |     91.0565 |     91.1628 |  0.10624 |     ms |
|                                       99th percentile latency |            autohisto_agg |     114.052 |     103.258 |  -10.794 |     ms |
|                                      100th percentile latency |            autohisto_agg |     120.757 |     110.535 | -10.2221 |     ms |
|                                  50th percentile service time |            autohisto_agg |     76.7718 |     77.1612 |  0.38943 |     ms |
|                                  90th percentile service time |            autohisto_agg |     89.3739 |     89.3345 | -0.03948 |     ms |
|                                  99th percentile service time |            autohisto_agg |     112.555 |     101.856 | -10.6994 |     ms |
|                                 100th percentile service time |            autohisto_agg |     119.382 |     109.272 | -10.1099 |     ms |
|                                                    error rate |            autohisto_agg |           0 |           0 |        0 |      % |
|                                                Min Throughput |       date_histogram_agg |     1.50646 |     1.50623 | -0.00023 |  ops/s |
|                                               Mean Throughput |       date_histogram_agg |     1.51067 |     1.51029 | -0.00038 |  ops/s |
|                                             Median Throughput |       date_histogram_agg |      1.5097 |     1.50938 | -0.00032 |  ops/s |
|                                                Max Throughput |       date_histogram_agg |     1.51915 |     1.51847 | -0.00068 |  ops/s |
|                                       50th percentile latency |       date_histogram_agg |     85.9077 |     86.4617 |  0.55393 |     ms |
|                                       90th percentile latency |       date_histogram_agg |     105.337 |     99.9942 | -5.34239 |     ms |
|                                       99th percentile latency |       date_histogram_agg |     113.145 |     103.191 | -9.95331 |     ms |
|                                      100th percentile latency |       date_histogram_agg |     120.109 |     108.578 | -11.5309 |     ms |
|                                  50th percentile service time |       date_histogram_agg |     84.3422 |     84.3296 | -0.01264 |     ms |
|                                  90th percentile service time |       date_histogram_agg |     103.613 |     98.0249 | -5.58851 |     ms |
|                                  99th percentile service time |       date_histogram_agg |     110.929 |     101.812 |  -9.1169 |     ms |
|                                 100th percentile service time |       date_histogram_agg |     118.724 |     107.082 | -11.6417 |     ms |
|                                                    error rate |       date_histogram_agg |           0 |           0 |        0 |      % |


-------------------------------
[INFO] SUCCESS (took 0 seconds)
-------------------------------

The nodes were running with 1gb jvm heap. Trying again with 4gb jvm heap

[2023-04-12T18:59:09,751][INFO ][o.o.n.Node               ] [ip-10-0-5-135.us-west-2.compute.internal] JVM arguments [-Xshare:auto, ,... -Xms1g, -Xmx1g, -XX:+UseG1GC, ...

@dreamer-89
Copy link
Member Author

dreamer-89 commented Apr 12, 2023

See improvements with latest run, used 4gb jvm heap size. @mch2 : I think we can go ahead with this change ?

Benchmark comparision

There is improvement in the cumulative indexing time, throughput and latency numbers for index task.

ubuntu@ip-172-31-49-145:~/OpenSearch-Benchmark$ opensearch-benchmark compare --baseline=e0018550-ef50-48ea-bc64-ae3addf2e4a0  --contender=91044e5c-c136-4fc6-a296-3e7425578693

   ____                  _____                      __       ____                  __                         __
  / __ \____  ___  ____ / ___/___  ____ ___________/ /_     / __ )___  ____  _____/ /_  ____ ___  ____ ______/ /__
 / / / / __ \/ _ \/ __ \\__ \/ _ \/ __ `/ ___/ ___/ __ \   / __  / _ \/ __ \/ ___/ __ \/ __ `__ \/ __ `/ ___/ //_/
/ /_/ / /_/ /  __/ / / /__/ /  __/ /_/ / /  / /__/ / / /  / /_/ /  __/ / / / /__/ / / / / / / / / /_/ / /  / ,<
\____/ .___/\___/_/ /_/____/\___/\__,_/_/   \___/_/ /_/  /_____/\___/_/ /_/\___/_/ /_/_/ /_/ /_/\__,_/_/  /_/|_|
    /_/


Comparing baseline
  TestExecution ID: e0018550-ef50-48ea-bc64-ae3addf2e4a0
  TestExecution timestamp: 2023-04-12 19:01:36
  TestProcedure: append-no-conflicts
  ProvisionConfigInstance: external

with contender
  TestExecution ID: 91044e5c-c136-4fc6-a296-3e7425578693
  TestExecution timestamp: 2023-04-12 20:01:04
  TestProcedure: append-no-conflicts
  ProvisionConfigInstance: external

------------------------------------------------------
    _______             __   _____
   / ____(_)___  ____ _/ /  / ___/_________  ________
  / /_  / / __ \/ __ `/ /   \__ \/ ___/ __ \/ ___/ _ \
 / __/ / / / / / /_/ / /   ___/ / /__/ /_/ / /  /  __/
/_/   /_/_/ /_/\__,_/_/   /____/\___/\____/_/   \___/
------------------------------------------------------

|                                                        Metric |                     Task |    Baseline |   Contender |     Diff |   Unit |
|--------------------------------------------------------------:|-------------------------:|------------:|------------:|---------:|-------:|
|                    Cumulative indexing time of primary shards |                          |     160.789 |     155.984 | -4.80557 |    min |
|             Min cumulative indexing time across primary shard |                          |     12.1672 |     11.6527 | -0.51453 |    min |
|          Median cumulative indexing time across primary shard |                          |     13.0152 |     12.3864 | -0.62878 |    min |
|             Max cumulative indexing time across primary shard |                          |     15.0026 |     14.9267 | -0.07595 |    min |
|           Cumulative indexing throttle time of primary shards |                          |           0 |           0 |        0 |    min |
|    Min cumulative indexing throttle time across primary shard |                          |           0 |           0 |        0 |    min |
| Median cumulative indexing throttle time across primary shard |                          |           0 |           0 |        0 |    min |
|    Max cumulative indexing throttle time across primary shard |                          |           0 |           0 |        0 |    min |
|                       Cumulative merge time of primary shards |                          |     164.878 |     163.775 | -1.10318 |    min |
|                      Cumulative merge count of primary shards |                          |        1918 |        1993 |       75 |        |
|                Min cumulative merge time across primary shard |                          |     12.5966 |     11.9167 |  -0.6799 |    min |
|             Median cumulative merge time across primary shard |                          |     13.9217 |     13.5281 | -0.39353 |    min |
|                Max cumulative merge time across primary shard |                          |     14.2696 |     15.8343 |   1.5647 |    min |
|              Cumulative merge throttle time of primary shards |                          |     74.1968 |     74.1208 |  -0.0761 |    min |
|       Min cumulative merge throttle time across primary shard |                          |     5.36398 |      5.4107 |  0.04672 |    min |
|    Median cumulative merge throttle time across primary shard |                          |     5.87859 |     5.80227 | -0.07633 |    min |
|       Max cumulative merge throttle time across primary shard |                          |      7.6324 |      8.0326 |   0.4002 |    min |
|                     Cumulative refresh time of primary shards |                          |     29.0306 |     28.7828 |  -0.2478 |    min |
|                    Cumulative refresh count of primary shards |                          |        9612 |       10010 |      398 |        |
|              Min cumulative refresh time across primary shard |                          |     2.10473 |     2.07305 | -0.03168 |    min |
|           Median cumulative refresh time across primary shard |                          |     2.44788 |     2.41253 | -0.03536 |    min |
|              Max cumulative refresh time across primary shard |                          |     2.68998 |     2.74652 |  0.05653 |    min |
|                       Cumulative flush time of primary shards |                          |     2.60152 |     2.20973 | -0.39178 |    min |
|                      Cumulative flush count of primary shards |                          |         192 |         192 |        0 |        |
|                Min cumulative flush time across primary shard |                          |     0.13045 |    0.125367 | -0.00508 |    min |
|             Median cumulative flush time across primary shard |                          |    0.204325 |    0.179067 | -0.02526 |    min |
|                Max cumulative flush time across primary shard |                          |    0.321633 |     0.26095 | -0.06068 |    min |
|                                       Total Young Gen GC time |                          |      45.614 |      45.459 |   -0.155 |      s |
|                                      Total Young Gen GC count |                          |        2734 |        2734 |        0 |        |
|                                         Total Old Gen GC time |                          |           0 |           0 |        0 |      s |
|                                        Total Old Gen GC count |                          |           0 |           0 |        0 |        |
|                                                    Store size |                          |      65.974 |     65.9319 | -0.04206 |     GB |
|                                                 Translog size |                          | 1.22935e-06 | 1.22935e-06 |        0 |     GB |
|                                        Heap used for segments |                          |           0 |           0 |        0 |     MB |
|                                      Heap used for doc values |                          |           0 |           0 |        0 |     MB |
|                                           Heap used for terms |                          |           0 |           0 |        0 |     MB |
|                                           Heap used for norms |                          |           0 |           0 |        0 |     MB |
|                                          Heap used for points |                          |           0 |           0 |        0 |     MB |
|                                   Heap used for stored fields |                          |           0 |           0 |        0 |     MB |
|                                                 Segment count |                          |         325 |         313 |      -12 |        |
|                                                Min Throughput |                    index |     90397.4 |     90973.7 |  576.234 | docs/s |
|                                               Mean Throughput |                    index |       94156 |     94633.6 |  477.523 | docs/s |
|                                             Median Throughput |                    index |     94785.5 |     94637.9 | -147.628 | docs/s |
|                                                Max Throughput |                    index |       97011 |     97262.8 |  251.757 | docs/s |
|                                       50th percentile latency |                    index |     831.803 |     841.246 |  9.44312 |     ms |
|                                       90th percentile latency |                    index |     1079.76 |      1046.9 | -32.8635 |     ms |
|                                       99th percentile latency |                    index |     1380.09 |     1281.62 | -98.4745 |     ms |
|                                     99.9th percentile latency |                    index |     1887.08 |     1575.18 |   -311.9 |     ms |
|                                    99.99th percentile latency |                    index |     2378.27 |      1807.4 | -570.868 |     ms |
|                                      100th percentile latency |                    index |     2552.62 |     1912.01 | -640.614 |     ms |
|                                  50th percentile service time |                    index |     831.803 |     841.246 |  9.44312 |     ms |
|                                  90th percentile service time |                    index |     1079.76 |      1046.9 | -32.8635 |     ms |
|                                  99th percentile service time |                    index |     1380.09 |     1281.62 | -98.4745 |     ms |
|                                99.9th percentile service time |                    index |     1887.08 |     1575.18 |   -311.9 |     ms |
|                               99.99th percentile service time |                    index |     2378.27 |      1807.4 | -570.868 |     ms |
|                                 100th percentile service time |                    index |     2552.62 |     1912.01 | -640.614 |     ms |
|                                                    error rate |                    index |           0 |           0 |        0 |      % |
|                                                Min Throughput | wait-until-merges-finish |   0.0531398 |  0.00538084 | -0.04776 |  ops/s |
|                                               Mean Throughput | wait-until-merges-finish |   0.0531398 |  0.00538084 | -0.04776 |  ops/s |
|                                             Median Throughput | wait-until-merges-finish |   0.0531398 |  0.00538084 | -0.04776 |  ops/s |
|                                                Max Throughput | wait-until-merges-finish |   0.0531398 |  0.00538084 | -0.04776 |  ops/s |
|                                      100th percentile latency | wait-until-merges-finish |       18818 |      185844 |   167026 |     ms |
|                                 100th percentile service time | wait-until-merges-finish |       18818 |      185844 |   167026 |     ms |
|                                                    error rate | wait-until-merges-finish |           0 |           0 |        0 |      % |
|                                                Min Throughput |                  default |     3.00525 |     3.00739 |  0.00214 |  ops/s |
|                                               Mean Throughput |                  default |      3.0085 |     3.01206 |  0.00357 |  ops/s |
|                                             Median Throughput |                  default |     3.00777 |     3.01099 |  0.00322 |  ops/s |
|                                                Max Throughput |                  default |     3.01491 |     3.02128 |  0.00637 |  ops/s |
|                                       50th percentile latency |                  default |     7.34868 |     7.50122 |  0.15253 |     ms |
|                                       90th percentile latency |                  default |     8.23766 |     8.68279 |  0.44513 |     ms |
|                                       99th percentile latency |                  default |     9.31858 |     9.56066 |  0.24208 |     ms |
|                                      100th percentile latency |                  default |     9.55781 |     10.2706 |  0.71275 |     ms |
|                                  50th percentile service time |                  default |     5.96266 |     6.11449 |  0.15183 |     ms |
|                                  90th percentile service time |                  default |     6.88244 |     7.12381 |  0.24137 |     ms |
|                                  99th percentile service time |                  default |     8.10346 |     8.02687 | -0.07658 |     ms |
|                                 100th percentile service time |                  default |     8.11502 |     8.99206 |  0.87705 |     ms |
|                                                    error rate |                  default |           0 |           0 |        0 |      % |
|                                                Min Throughput |                    range |    0.703352 |    0.703302 |   -5e-05 |  ops/s |
|                                               Mean Throughput |                    range |    0.705515 |    0.705426 |   -9e-05 |  ops/s |
|                                             Median Throughput |                    range |    0.705018 |    0.704937 |   -8e-05 |  ops/s |
|                                                Max Throughput |                    range |    0.709965 |    0.709786 | -0.00018 |  ops/s |
|                                       50th percentile latency |                    range |      32.742 |     31.1131 | -1.62886 |     ms |
|                                       90th percentile latency |                    range |     41.7257 |     36.6342 | -5.09142 |     ms |
|                                       99th percentile latency |                    range |     46.1819 |     38.5113 | -7.67059 |     ms |
|                                      100th percentile latency |                    range |      105.84 |     41.9368 | -63.9029 |     ms |
|                                  50th percentile service time |                    range |     29.3692 |     28.3179 |  -1.0513 |     ms |
|                                  90th percentile service time |                    range |     38.9015 |     33.3912 | -5.51028 |     ms |
|                                  99th percentile service time |                    range |     42.6105 |     36.9551 | -5.65545 |     ms |
|                                 100th percentile service time |                    range |     103.659 |     38.1672 |  -65.492 |     ms |
|                                                    error rate |                    range |           0 |           0 |        0 |      % |
|                                                Min Throughput |      distance_amount_agg |      2.0092 |     2.00598 | -0.00322 |  ops/s |
|                                               Mean Throughput |      distance_amount_agg |     2.01512 |     2.00984 | -0.00528 |  ops/s |
|                                             Median Throughput |      distance_amount_agg |     2.01375 |     2.00894 | -0.00481 |  ops/s |
|                                                Max Throughput |      distance_amount_agg |     2.02711 |     2.01765 | -0.00946 |  ops/s |
|                                       50th percentile latency |      distance_amount_agg |       6.641 |     6.78062 |  0.13962 |     ms |
|                                       90th percentile latency |      distance_amount_agg |     7.34796 |     7.36822 |  0.02027 |     ms |
|                                       99th percentile latency |      distance_amount_agg |     8.04209 |     7.99801 | -0.04408 |     ms |
|                                      100th percentile latency |      distance_amount_agg |     8.78268 |     8.35211 | -0.43057 |     ms |
|                                  50th percentile service time |      distance_amount_agg |     4.98163 |     5.25042 |  0.26879 |     ms |
|                                  90th percentile service time |      distance_amount_agg |      5.5378 |     5.83439 |  0.29659 |     ms |
|                                  99th percentile service time |      distance_amount_agg |     6.16771 |      6.8727 |  0.70498 |     ms |
|                                 100th percentile service time |      distance_amount_agg |     6.95583 |     7.11845 |  0.16262 |     ms |
|                                                    error rate |      distance_amount_agg |           0 |           0 |        0 |      % |
|                                                Min Throughput |            autohisto_agg |      1.4995 |     1.49816 | -0.00134 |  ops/s |
|                                               Mean Throughput |            autohisto_agg |     1.49972 |     1.49896 | -0.00077 |  ops/s |
|                                             Median Throughput |            autohisto_agg |     1.49975 |     1.49906 | -0.00069 |  ops/s |
|                                                Max Throughput |            autohisto_agg |     1.49983 |     1.49935 | -0.00048 |  ops/s |
|                                       50th percentile latency |            autohisto_agg |     77.1848 |      78.852 |  1.66721 |     ms |
|                                       90th percentile latency |            autohisto_agg |     95.4324 |      99.192 |  3.75967 |     ms |
|                                       99th percentile latency |            autohisto_agg |     100.909 |     107.024 |  6.11503 |     ms |
|                                      100th percentile latency |            autohisto_agg |     101.252 |     131.467 |  30.2146 |     ms |
|                                  50th percentile service time |            autohisto_agg |     75.5076 |     77.2307 |  1.72303 |     ms |
|                                  90th percentile service time |            autohisto_agg |     93.8644 |     97.7362 |  3.87184 |     ms |
|                                  99th percentile service time |            autohisto_agg |     99.4097 |     104.096 |  4.68611 |     ms |
|                                 100th percentile service time |            autohisto_agg |     99.6838 |     128.953 |  29.2689 |     ms |
|                                                    error rate |            autohisto_agg |           0 |           0 |        0 |      % |
|                                                Min Throughput |       date_histogram_agg |     1.50829 |     1.50822 |   -7e-05 |  ops/s |
|                                               Mean Throughput |       date_histogram_agg |     1.51369 |     1.51356 | -0.00013 |  ops/s |
|                                             Median Throughput |       date_histogram_agg |     1.51244 |     1.51234 |  -0.0001 |  ops/s |
|                                                Max Throughput |       date_histogram_agg |     1.52467 |     1.52435 | -0.00032 |  ops/s |
|                                       50th percentile latency |       date_histogram_agg |     81.3622 |     81.4776 |  0.11536 |     ms |
|                                       90th percentile latency |       date_histogram_agg |      101.51 |     101.179 | -0.33125 |     ms |
|                                       99th percentile latency |       date_histogram_agg |     107.572 |     112.943 |  5.37122 |     ms |
|                                      100th percentile latency |       date_histogram_agg |     108.563 |     114.006 |  5.44328 |     ms |
|                                  50th percentile service time |       date_histogram_agg |     79.2952 |     79.7108 |  0.41553 |     ms |
|                                  90th percentile service time |       date_histogram_agg |     99.6272 |     100.025 |  0.39791 |     ms |
|                                  99th percentile service time |       date_histogram_agg |      106.17 |     110.977 |  4.80745 |     ms |
|                                 100th percentile service time |       date_histogram_agg |     106.983 |     111.168 |  4.18562 |     ms |
|                                                    error rate |       date_histogram_agg |           0 |           0 |        0 |      % |


-------------------------------
[INFO] SUCCESS (took 0 seconds)
-------------------------------

EBS Read/Write

There is no significant drop in EBS reads/writes both count/bytes.

Baseline: Started 12pm PDT
Contender: Started 1pm PDT

Metric Bytes

Screenshot 2023-04-12 at 2 05 25 PM

Metric Count

Screenshot 2023-04-12 at 2 04 58 PM

@opensearch-trigger-bot
Copy link
Contributor

This PR is stalled because it has been open for 30 days with no activity. Remove stalled label or comment or this will be closed in 7 days.

@opensearch-trigger-bot opensearch-trigger-bot bot added the stalled Issues that have stalled label Jun 24, 2023
@opensearch-trigger-bot
Copy link
Contributor

This PR was closed because it has been stalled for 7 days with no activity.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
skip-changelog stalled Issues that have stalled
Projects
None yet
Development

Successfully merging this pull request may close these issues.

[Segment Replication] Prevent un-necessary disk reads on replica
3 participants