Improve block cache description, update default

src/current/v25.4/cockroach-start-single-node.md

@@ -41,7 +41,7 @@ Flag | Description
 -----|-----------
 `--attrs` | **Not relevant for single-node clusters.** Arbitrary strings, separated by colons, specifying node capability, which might include specialized hardware or number of cores, for example:<br><br>`--attrs=ram:64gb`<br><br>These can be used to influence the location of data replicas. See [Replication Controls]({% link {{ page.version.version }}/configure-replication-zones.md %}#replication-constraints) for full details.
 `--background` | Runs the node in the background. Control is returned to the shell only once the node is ready to accept requests, so this is recommended over appending `&` to the command. This flag is **not** available in Windows environments.<br><br>**Note:** `--background` is suitable for writing automated test suites or maintenance procedures that need a temporary server process running in the background. It is not intended to be used to start a long-running server, because it does not fully detach from the controlling terminal. Consider using a service manager or a tool like [daemon(8)](https://www.freebsd.org/cgi/man.cgi?query=daemon&sektion=8) instead. If you use `--background`, using `--pid-file` is also recommended. To gracefully stop the `cockroach` process, ssend the `SIGTERM` signal to the process ID in the PID file. To gracefully restart the process, send the `SIGHUP` signal.
-`--cache` | The total size for caches, shared evenly if there are multiple storage devices. This can be a percentage (notated as a decimal or with `%`) or any bytes-based unit, for example: <br><br>`--cache=.25`<br>`--cache=25%`<br>`--cache=1000000000 ----> 1000000000 bytes`<br>`--cache=1GB ----> 1000000000 bytes`<br>`--cache=1GiB ----> 1073741824 bytes` <br><br><strong>Note:</strong> If you use the `%` notation, you might need to escape the `%` sign, for instance, while configuring CockroachDB through `systemd` service files. For this reason, it's recommended to use the decimal notation instead.<br><br>**Note:** The sum of `--cache`, `--max-sql-memory`, and `--max-tsdb-memory` should not exceed 75% of the memory available to the `cockroach` process.<br><br>**Default:** `128MiB`<br><br>The default cache size is reasonable for local development clusters. For production deployments, this should be increased to 25% or higher. Increasing the cache size will generally improve the node's read performance. See [Recommended Production Settings]({% link {{ page.version.version }}/recommended-production-settings.md %}#cache-and-sql-memory-size) for more details.
+`--cache` | The total size for the [Pebble Storage engine]({% link {{ page.version.version }}/architecture/storage-layer.md %}#pebble)'s block cache, shared evenly if there are multiple storage devices. This can be a percentage (notated as a decimal or with `%`) or any bytes-based unit, for example: <br><br>`--cache=.25`<br>`--cache=25%`<br>`--cache=1000000000 ----> 1000000000 bytes`<br>`--cache=1GB ----> 1000000000 bytes`<br>`--cache=1GiB ----> 1073741824 bytes` <br><br><strong>Note:</strong> If you use the `%` notation, you might need to escape the `%` sign, for instance, while configuring CockroachDB through `systemd` service files. For this reason, it's recommended to use the decimal notation instead.<br><br>**Note:** The sum of `--cache`, `--max-sql-memory`, and `--max-tsdb-memory` should not exceed 75% of the memory available to the `cockroach` process.<br><br>**Default:** `256 MiB`<br><br>The default cache size is reasonable for local development clusters. For production deployments, this should be increased to 25% or higher. Increasing the cache size will generally improve the node's read performance.<br><br>The block cache holds uncompressed blocks of persisted [key-value data]({% link {{ page.version.version }}/architecture/distribution-layer.md %}#overview) in memory. If a read misses within the block cache, the storage engine reads the file via the operating system's page cache, which may hold the relevant block in-memory in its compressed form. Otherwise, the read is served from the storage device. The block cache fills to the configured size and is then recycled using a least-recently used (LRU) policy.<br><br>Refer to [Recommended Production Settings]({% link {{ page.version.version }}/recommended-production-settings.md %}#cache-and-sql-memory-size) for more details. Production systems should always configure this setting.
 `--external-io-dir` | The path of the external IO directory with which the local file access paths are prefixed while performing backup and restore operations using local node directories or NFS drives. If set to `disabled`, backups and restores using local node directories and NFS drives are disabled.<br><br>**Default:** `extern` subdirectory of the first configured [`store`](#store).<br><br>To set the `--external-io-dir` flag to the locations you want to use without needing to restart nodes, create symlinks to the desired locations from within the `extern` directory.
 `--listening-url-file` | The file to which the node's SQL connection URL will be written on successful startup, in addition to being printed to the [standard output](#standard-output).<br><br>This is particularly helpful in identifying the node's port when an unused port is assigned automatically (`--port=0`).
 `--locality` | **Not relevant for single-node clusters.** Arbitrary key-value pairs that describe the location of the node. Locality might include country, region, datacenter, rack, etc. For more details, see [Locality]({% link {{ page.version.version }}/cockroach-start.md %}#locality) below.

src/current/v25.4/cockroach-start.md

@@ -56,7 +56,7 @@ Flag | Description
 -----|-----------
 `--attrs` | Arbitrary strings, separated by colons, specifying node capability, which might include specialized hardware or number of cores, for example:<br><br>`--attrs=ram:64gb`<br><br>These can be used to influence the location of data replicas. See [Replication Controls]({% link {{ page.version.version }}/configure-replication-zones.md %}#replication-constraints) for full details.
 `--background` | Runs the node in the background. Control is returned to the shell only once the node is ready to accept requests, so this is recommended over appending `&` to the command. This flag is **not** available in Windows environments.<br><br>**Note:** `--background` is suitable for writing automated test suites or maintenance procedures that need a temporary server process running in the background. It is not intended to be used to start a long-running server, because it does not fully detach from the controlling terminal. Consider using a service manager or a tool like [daemon(8)](https://www.freebsd.org/cgi/man.cgi?query=daemon&sektion=8) instead. If you use `--background`, using `--pid-file` is also recommended. To gracefully stop the `cockroach` process, send the `SIGTERM` signal to the process ID in the PID file. To gracefully restart the process, send the `SIGHUP` signal.
-`--cache` | The total size for caches, shared evenly if there are multiple storage devices. This can be a percentage (notated as a decimal or with `%`) or any bytes-based unit; for example: <br><br>`--cache=.25`<br>`--cache=25%`<br>`--cache=1000000000 ----> 1000000000 bytes`<br>`--cache=1GB ----> 1000000000 bytes`<br>`--cache=1GiB ----> 1073741824 bytes` <br><br>**Note:** If you use the `%` notation, you might need to escape the `%` sign (for instance, while configuring CockroachDB through `systemd` service files). For this reason, it's recommended to use the decimal notation instead.<br><br>**Note:** The sum of `--cache`, `--max-sql-memory`, and `--max-tsdb-memory` should not exceed 75% of the memory available to the `cockroach` process.<br><br>**Default:** `128MiB`<br><br>The default cache size is reasonable for local development clusters. For production deployments, this should be increased to 25% or higher. Increasing the cache size will generally improve the node's read performance. For more details, see [Recommended Production Settings]({% link {{ page.version.version }}/recommended-production-settings.md %}#cache-and-sql-memory-size).
+`--cache` | The total size for the [Pebble Storage engine]({% link {{ page.version.version }}/architecture/storage-layer.md %}#pebble)'s block cache, shared evenly if there are multiple storage devices. This can be a percentage (notated as a decimal or with `%`) or any bytes-based unit, for example: <br><br>`--cache=.25`<br>`--cache=25%`<br>`--cache=1000000000 ----> 1000000000 bytes`<br>`--cache=1GB ----> 1000000000 bytes`<br>`--cache=1GiB ----> 1073741824 bytes` <br><br>**Note:** If you use the `%` notation, you might need to escape the `%` sign, for instance, while configuring CockroachDB through `systemd` service files. For this reason, it's recommended to use the decimal notation instead.<br><br>**Note:** The sum of `--cache`, `--max-sql-memory`, and `--max-tsdb-memory` should not exceed 75% of the memory available to the `cockroach` process.<br><br>**Default:** `256 MiB`<br><br>The default cache size is reasonable for local development clusters. For production deployments, this should be increased to 25% or higher. Increasing the cache size will generally improve the node's read performance.<br><br>The block cache holds uncompressed blocks of persisted [key-value data]({% link {{ page.version.version }}/architecture/distribution-layer.md %}#overview) in memory. If a read misses within the block cache, the storage engine reads the file via the operating system's page cache, which may hold the relevant block in-memory in its compressed form. Otherwise, the read is served from the storage device. The block cache fills to the configured size and is then recycled using a least-recently used (LRU) policy.<br><br>Refer to [Recommended Production Settings]({% link {{ page.version.version }}/recommended-production-settings.md %}#cache-and-sql-memory-size) for more details. Production systems should always configure this setting.
 `--clock-device` | Enable CockroachDB to use a [PTP hardware clock](https://www.kernel.org/doc/html/latest/driver-api/ptp.html) when querying the current time. The value is a string that specifies the clock device to use. For example: `--clock-device=/dev/ptp0`<br><br>**Note:** This is supported on Linux only and may be needed in cases where the host clock is unreliable or prone to large jumps (e.g., when using vMotion).
 `--cluster-name` | <a name="flags-cluster-name"></a> A string that specifies a cluster name. This is used together with `--join` to ensure that all newly created nodes join the intended cluster when you are running multiple clusters.<br><br>**Note:** If this is set, [`cockroach init`]({% link {{ page.version.version }}/cockroach-init.md %}), [`cockroach node decommission`]({% link {{ page.version.version }}/cockroach-node.md %}), [`cockroach node recommission`]({% link {{ page.version.version }}/cockroach-node.md %}), and the `cockroach debug` commands must specify either `--cluster-name` or `--disable-cluster-name-verification` in order to work.
 `--disable-cluster-name-verification` | On clusters for which a cluster name has been set, this flag paired with `--cluster-name` disables the cluster name check for the command. This is necessary on existing clusters, when setting a cluster name or changing the cluster name: Perform a rolling restart of all nodes and include both the new `--cluster-name` value and `--disable-cluster-name-verification`, then a second rolling restart with `--cluster-name` and without `--disable-cluster-name-verification`.

src/current/v25.4/recommended-production-settings.md

@@ -342,7 +342,9 @@ For details about available backup and restore types in CockroachDB, see [Backup
 
 CockroachDB manages its own memory caches, independently of the operating system. These are configured via the [`--cache`]({% link {{ page.version.version }}/cockroach-start.md %}#flags) and [`--max-sql-memory`]({% link {{ page.version.version }}/cockroach-start.md %}#flags) flags.
 
-Each node has a default cache size of `128MiB` that is passively consumed. The default was chosen to facilitate development and testing, where users are likely to run multiple CockroachDB nodes on a single machine. Increasing the cache size will generally improve the node's read performance.
+The default cache size is per-node and is passively consumed; it was chosen to facilitate development and testing, where users are likely to run multiple CockroachDB nodes on a single machine. Increasing the cache size will generally improve the node's read performance. Production systems should always configure this setting.
+
+The [`--cache`]({% link {{ page.version.version }}/cockroach-start.md %}#flags) flag controls the [Pebble storage engine]({% link {{ page.version.version }}/architecture/storage-layer.md %}#pebble)'s block cache, which holds uncompressed blocks of persisted [key-value data]({% link {{ page.version.version }}/architecture/distribution-layer.md %}#overview) in memory. If a read misses within the block cache, the storage engine reads the file via the operating system's page cache, which may hold the relevant block in-memory in its compressed form. Otherwise, the read is served from the storage device. The block cache fills to the configured size and is then recycled using a least-recently used (LRU) policy.
 
 Each node has a default SQL memory size of `25%`. This memory is used as-needed by active operations to store temporary data for SQL queries.