You are viewing a plain text version of this content. The canonical link for it is here.
Posted to issues@flink.apache.org by "Yu Li (Jira)" <ji...@apache.org> on 2020/01/08 08:17:00 UTC
[jira] [Commented] (FLINK-15512) Refactor the mechanism of how to
constructure the cache and write buffer manager shared across RocksDB
instances
[ https://issues.apache.org/jira/browse/FLINK-15512?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17010457#comment-17010457 ]
Yu Li commented on FLINK-15512:
-------------------------------
More details, please refer to the [discussion|https://issues.apache.org/jira/browse/FLINK-15368?focusedCommentId=17007221&page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel#comment-17007221] in FLINK-15368
> Refactor the mechanism of how to constructure the cache and write buffer manager shared across RocksDB instances
> -----------------------------------------------------------------------------------------------------------------
>
> Key: FLINK-15512
> URL: https://issues.apache.org/jira/browse/FLINK-15512
> Project: Flink
> Issue Type: Sub-task
> Components: Runtime / State Backends
> Reporter: Yun Tang
> Priority: Critical
> Fix For: 1.10.0
>
>
> FLINK-14484 introduce a {{LRUCache}} to share among RocksDB instances, so that the memory usage by RocksDB could be controlled well. However, due to the implementation and some bugs in RocksDB ([issue-6247|https://github.com/facebook/rocksdb/issues/6247]), we cannot limit the memory strictly.
> The way to walk around this issue is to consider the buffer which memtable would overuse (1/2 write buffer manager size). By introducing this, the actual cache size for user to share is not the same as the managed off-heap memory or user configured memory.
--
This message was sent by Atlassian Jira
(v8.3.4#803005)