You are viewing a plain text version of this content. The canonical link for it is here.
Posted to issues@spark.apache.org by "Jungtaek Lim (Jira)" <ji...@apache.org> on 2023/10/11 23:47:00 UTC
[jira] [Resolved] (SPARK-45415) RocksDB consumes excessive disk space when many concurrent streaming queries are using dedup
[ https://issues.apache.org/jira/browse/SPARK-45415?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]
Jungtaek Lim resolved SPARK-45415.
----------------------------------
Fix Version/s: 4.0.0
Resolution: Fixed
Issue resolved by pull request 43202
[https://github.com/apache/spark/pull/43202]
> RocksDB consumes excessive disk space when many concurrent streaming queries are using dedup
> --------------------------------------------------------------------------------------------
>
> Key: SPARK-45415
> URL: https://issues.apache.org/jira/browse/SPARK-45415
> Project: Spark
> Issue Type: Bug
> Components: Structured Streaming
> Affects Versions: 3.3.2
> Environment: Apache spark on AWS EMR, local spark on a laptop on Linux. Does not impact MacOS due to missing support in RocksDB for pre-allocation (MacOS does not support the fallocate system call).
> Reporter: Scott Schenkein
> Assignee: Scott Schenkein
> Priority: Minor
> Labels: pull-request-available
> Fix For: 4.0.0
>
> Original Estimate: 4h
> Remaining Estimate: 4h
>
> Our spark environment features a number of parallel structured streaming jobs, many of which use state store. Most use state store for dropDuplicates and work with a tiny amount of information, but a few have a substantially large state store requiring use of RocksDB. In such a configuration, spark allocates a minimum of {{spark.sql.shuffle.partitions * queryCount}} partitions, each of which pre-allocate about 74mb (observed on EMR/Hadoop) disk storage for RocksDB. This allocation is due to pre-allocation of log files space using [fallocate|https://github.com/facebook/rocksdb/blob/main/include/rocksdb/options.h#L871-L880], requiring users to either unnaturally reduce shuffle partitions, split running spark instances, or allocate a large amount of wasted storage.
--
This message was sent by Atlassian Jira
(v8.20.10#820010)
---------------------------------------------------------------------
To unsubscribe, e-mail: issues-unsubscribe@spark.apache.org
For additional commands, e-mail: issues-help@spark.apache.org