[
https://issues.apache.org/jira/browse/FLINK-8845?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16482298#comment-16482298
]
ASF GitHub Bot commented on FLINK-8845:
---------------------------------------
Github user sihuazhou commented on the issue:
https://github.com/apache/flink/pull/5650
A micro-benchmark for this: @StefanRRichter @StephanEwen
```
-----------------------> With disableWAL is false <-----------------
Number of values added | time for Put | time for WriteBach |
performance improvement of WriteBatch over Put
1000 10146397 ns 3546287 ns
2.86x
10000 118227077 ns 26040222 ns
4.54x
100000 1838593196 ns 375053755 ns
4.9x
1000000 8844612079 ns 2014077396 ns
4.39x
-----------------------> With disableWAL is true <-----------------
1000 3955204 ns 2429725 ns
1.62x
10000 25618237 ns 16440113 ns
1.55x
100000 289153346 ns 183712685 ns
1.57x
1000000 2886298967 ns 1768688571 ns
1.63x
```
> Use WriteBatch to improve performance for recovery in RocksDB backend
> ---------------------------------------------------------------------
>
> Key: FLINK-8845
> URL: https://issues.apache.org/jira/browse/FLINK-8845
> Project: Flink
> Issue Type: Improvement
> Components: State Backends, Checkpointing
> Affects Versions: 1.5.0
> Reporter: Sihua Zhou
> Assignee: Sihua Zhou
> Priority: Major
> Fix For: 1.6.0
>
>
> Base on {{WriteBatch}} we could get 30% ~ 50% performance lift when loading
> data into RocksDB.
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)
