kyuubi/docs/deployment/spark/shuffle.md
Kent Yao 183e18f1bc
[KYUUBI #404] How To Use Spark Dynamic Resource Allocation (DRA) in Kyuubi
![yaooqinn](https://badgen.net/badge/Hello/yaooqinn/green) [![Closes #404](https://badgen.net/badge/Preview/Closes%20%23404/blue)](https://github.com/yaooqinn/kyuubi/pull/404) ![518](https://badgen.net/badge/%2B/518/red) ![443](https://badgen.net/badge/-/443/green) ![3](https://badgen.net/badge/commits/3/yellow) ![Target Issue](https://badgen.net/badge/Missing/Target%20Issue/ff0000) [&#10088;?&#10089;](https://pullrequestbadge.com/?utm_medium=github&utm_source=yaooqinn&utm_campaign=badge_info)<!-- PR-BADGE: PLEASE DO NOT REMOVE THIS COMMENT -->

<!--
Thanks for sending a pull request!

Here are some tips for you:
  1. If this is your first time, please read our contributor guidelines: https://kyuubi.readthedocs.io/en/latest/community/contributions.html
  2. If the PR is related to an issue in https://github.com/yaooqinn/kyuubi/issues, add '[KYUUBI #XXXX]' in your PR title, e.g., '[KYUUBI #XXXX] Your PR title ...'.
  3. If the PR is unfinished, add '[WIP]' in your PR title, e.g., '[WIP][KYUUBI #XXXX] Your PR title ...'.
-->

### _Why are the changes needed?_
<!--
Please clarify why the changes are needed. For instance,
  1. If you add a feature, you can talk about the use case of it.
  2. If you fix a bug, you can clarify why it is a bug.
-->

### _How was this patch tested?_
- [ ] Add some test cases that check the changes thoroughly including negative and positive cases if possible

- [ ] Add screenshots for manual tests if appropriate

- [x] [Run test](https://kyuubi.readthedocs.io/en/latest/tools/testing.html#running-tests) locally before make a pull request

Closes #404 from yaooqinn/dra.

b8b755c [Kent Yao] How To Use Spark Dynamic Resource Allocation (DRA) in Kyuubi
9bb47b1 [Kent Yao] How To Use Spark Dynamic Resource Allocation (DRA) in Kyuubi
e786787 [Kent Yao] How To Use Spark Dynamic Resource Allocation (DRA) in Kyuubi

Authored-by: Kent Yao <yao@apache.org>
Signed-off-by: Kent Yao <yao@apache.org>
2021-03-08 18:27:49 +08:00

1.7 KiB


## Shuffle Behavior
# spark.shuffle.compress                                    true
# spark.shuffle.detectCorrupt                               true
# spark.shuffle.detectCorrupt.useExtraMemory                true
# spark.shuffle.file.buffer                                 64k
# spark.shuffle.unsafe.file.output.buffer                   64k
# spark.shuffle.spill.diskWriteBufferSize                   8k
# spark.shuffle.spill.compress                              true
# spark.shuffle.mapOutput.dispatcher.numThreads             12
# spark.shuffle.mapOutput.parallelAggregationThreshold      5000
# spark.shuffle.readHostLocalDisk                           true
# spark.shuffle.io.maxRetries                               10
# spark.shuffle.io.retryWait                                6s
# spark.shuffle.io.preferDirectBufs                         false
# spark.shuffle.io.serverThreads                            8
# spark.shuffle.io.clientThreads                            8
# spark.shuffle.io.connectionTimeout                        240s
# spark.shuffle.registration.timeout                        6000
# spark.shuffle.registration.maxAttempts                    10
# spark.shuffle.sync                                        false
# spark.shuffle.useOldFetchProtocol                         true
# spark.shuffle.unsafe.fastMergeEnabled                     true
# spark.shuffle.minNumPartitionsToHighlyCompress            100
# spark.network.maxRemoteBlockSizeFetchToMem                128m
# spark.reducer.maxSizeInFlight                             48m
# spark.reducer.maxReqsInFlight                             256
# spark.reducer.maxBlocksInFlightPerAddress                 256