[Doc] Add doc for lake_enable_vertical_compaction_fill_data_cache (St…

…arRocks#39004)

Signed-off-by: 絵空事スピリット <wanglichen@starrocks.com>

docs/en/administration/BE_configuration.md

@@ -361,6 +361,18 @@ BE dynamic parameters are as follows.
 - **Default:** 10 (Number of Threads)
 - **Description:** The thread pool size allowed on each BE for interacting with Kafka. Currently, the FE responsible for processing Routine Load requests depends on BEs to interact with Kafka, and each BE in StarRocks has its own thread pool for interactions with Kafka. If a large number of Routine Load tasks are distributed to a BE, the BE's thread pool for interactions with Kafka may be too busy to process all tasks in a timely manner. In this situation, you can adjust the value of this parameter to suit your needs.
 
+#### lake_enable_vertical_compaction_fill_data_cache
+
+- **Default:** false
+- **Description:** Whether to allow compaction tasks to cache data on local disks in a shared-data cluster.
+- **Introduced in:** v3.1.7, v3.2.3
+
+#### compact_threads
+
+- **Default:** 4
+- **Description:** The maximum number of threads used for concurrent compaction tasks. This configuration is changed to dynamic from v3.1.7 and v3.2.2 onwards.
+- **Introduced in:** v3.0.0
+
 #### update_compaction_ratio_threshold
 
 - **Default:** 0.5

docs/zh/administration/BE_configuration.md

@@ -421,6 +421,18 @@ curl -XPOST http://be_host:http_port/api/update_config?configuration_item=value
 - 含义：单个 BE 上与 Kafka 交互的线程池大小。当前 Routine Load FE 与 Kafka 的交互需经由 BE 完成，而每个 BE 上实际执行操作的是一个单独的线程池。当 Routine Load 任务较多时，可能会出现线程池线程繁忙的情况，可以调整该配置。
 - 默认值：10
 
+#### lake_enable_vertical_compaction_fill_data_cache
+
+- 含义：存算分离集群下，是否允许 Compaction 任务在执行时缓存数据到本地磁盘上。
+- 默认值：false
+- 引入版本：v3.1.7、v3.2.3
+
+#### compact_threads
+
+- 含义：并发 Compaction 任务的最大线程数。自 v3.1.7，v3.2.2 起变为动态参数。
+- 默认值：4
+- 引入版本：v3.0.0
+
 #### update_compaction_ratio_threshold
 
 - 含义：存算分离集群下主键模型表单次 Compaction 可以合并的最大数据比例。如果单个 Tablet 过大，建议适当调小该配置项取值。自 v3.1.5 起支持。