[SPARK-42277][CORE] Use RocksDB for spark.history.store.hybridStore.diskBackend by default#39845
Closed
dongjoon-hyun wants to merge 1 commit intoapache:masterfrom
dongjoon-hyun:SPARK-42277
Closed
[SPARK-42277][CORE] Use RocksDB for spark.history.store.hybridStore.diskBackend by default#39845dongjoon-hyun wants to merge 1 commit intoapache:masterfrom dongjoon-hyun:SPARK-42277
spark.history.store.hybridStore.diskBackend by default#39845dongjoon-hyun wants to merge 1 commit intoapache:masterfrom
dongjoon-hyun:SPARK-42277
Conversation
…iskBackend by default
Member
Author
|
cc @mridulm and @tgravescs |
Member
Author
|
Thank you, @LuciferYang . |
Member
Author
|
Thank you, @sunchao ! |
viirya
approved these changes
Feb 1, 2023
Member
Author
|
Thank you, @viirya ! |
dongjoon-hyun
added a commit
that referenced
this pull request
Feb 1, 2023
…diskBackend` by default ### What changes were proposed in this pull request? This PR aims to use `RocksDB` for `spark.history.store.hybridStore.diskBackend` by default instead of `LevelDB`. The last update of `LevelDB` is about over 10 years. - `all except aarch64`: https://mvnrepository.com/artifact/org.fusesource.leveldbjni/leveldbjni-all/1.8 (Oct 17, 2013) - `aarch64-only`: https://mvnrepository.com/artifact/org.openlabtesting.leveldbjni/leveldbjni-all/1.8 (Aug 28, 2019) ### Why are the changes needed? RocksDB is maintained well and works on all environments where Apache Spark supports while LevelDB is not maintained at all in these days and doesn't work in some environment like the following. ``` $ SPARK_HISTORY_OPTS="-Dspark.history.fs.logDirectory=$HOME/data/history" sbin/start-history-server.sh starting org.apache.spark.deploy.history.HistoryServer, logging to /Users/dongjoon/APACHE/spark-merge/logs/spark-dongjoon-org.apache.spark.deploy.history.HistoryServer-1-coffee.local.out failed to launch: nice -n 0 /Users/dongjoon/APACHE/spark-merge/bin/spark-class org.apache.spark.deploy.history.HistoryServer at org.fusesource.hawtjni.runtime.Library.load(Library.java:140) at org.fusesource.leveldbjni.JniDBFactory.<clinit>(JniDBFactory.java:48) at org.apache.spark.util.kvstore.LevelDB.<init>(LevelDB.java:88) at org.apache.spark.status.KVUtils$.open(KVUtils.scala:98) at org.apache.spark.status.KVUtils$.$anonfun$createKVStore$1(KVUtils.scala:144) at scala.Option.map(Option.scala:230) at org.apache.spark.status.KVUtils$.createKVStore(KVUtils.scala:127) at org.apache.spark.deploy.history.FsHistoryProvider.<init>(FsHistoryProvider.scala:136) at org.apache.spark.deploy.history.FsHistoryProvider.<init>(FsHistoryProvider.scala:86) ... 7 more ``` ### Does this PR introduce _any_ user-facing change? Since `spark.history.store.hybridStore.enabled` is `false` by default. There is no behavior change. In case of `spark.history.store.hybridStore.enabled=true`, History Server will rebuild the new hybrid store simply. ### How was this patch tested? Pass the CIs. Closes #39845 from dongjoon-hyun/SPARK-42277. Authored-by: Dongjoon Hyun <dongjoon@apache.org> Signed-off-by: Dongjoon Hyun <dongjoon@apache.org> (cherry picked from commit 39c9945) Signed-off-by: Dongjoon Hyun <dongjoon@apache.org>
Member
Author
|
Merged to master/3.4. Thank you all! |
Contributor
|
so if user is upgrading and want to reuse the disk contents, they have to explicitly set to leveldb, correct? might be good to have release notes about this. |
Member
Author
|
Yes, that's right. Thank you, @tgravescs . I added |
snmvaughan
pushed a commit
to snmvaughan/spark
that referenced
this pull request
Jun 20, 2023
…diskBackend` by default ### What changes were proposed in this pull request? This PR aims to use `RocksDB` for `spark.history.store.hybridStore.diskBackend` by default instead of `LevelDB`. The last update of `LevelDB` is about over 10 years. - `all except aarch64`: https://mvnrepository.com/artifact/org.fusesource.leveldbjni/leveldbjni-all/1.8 (Oct 17, 2013) - `aarch64-only`: https://mvnrepository.com/artifact/org.openlabtesting.leveldbjni/leveldbjni-all/1.8 (Aug 28, 2019) ### Why are the changes needed? RocksDB is maintained well and works on all environments where Apache Spark supports while LevelDB is not maintained at all in these days and doesn't work in some environment like the following. ``` $ SPARK_HISTORY_OPTS="-Dspark.history.fs.logDirectory=$HOME/data/history" sbin/start-history-server.sh starting org.apache.spark.deploy.history.HistoryServer, logging to /Users/dongjoon/APACHE/spark-merge/logs/spark-dongjoon-org.apache.spark.deploy.history.HistoryServer-1-coffee.local.out failed to launch: nice -n 0 /Users/dongjoon/APACHE/spark-merge/bin/spark-class org.apache.spark.deploy.history.HistoryServer at org.fusesource.hawtjni.runtime.Library.load(Library.java:140) at org.fusesource.leveldbjni.JniDBFactory.<clinit>(JniDBFactory.java:48) at org.apache.spark.util.kvstore.LevelDB.<init>(LevelDB.java:88) at org.apache.spark.status.KVUtils$.open(KVUtils.scala:98) at org.apache.spark.status.KVUtils$.$anonfun$createKVStore$1(KVUtils.scala:144) at scala.Option.map(Option.scala:230) at org.apache.spark.status.KVUtils$.createKVStore(KVUtils.scala:127) at org.apache.spark.deploy.history.FsHistoryProvider.<init>(FsHistoryProvider.scala:136) at org.apache.spark.deploy.history.FsHistoryProvider.<init>(FsHistoryProvider.scala:86) ... 7 more ``` ### Does this PR introduce _any_ user-facing change? Since `spark.history.store.hybridStore.enabled` is `false` by default. There is no behavior change. In case of `spark.history.store.hybridStore.enabled=true`, History Server will rebuild the new hybrid store simply. ### How was this patch tested? Pass the CIs. Closes apache#39845 from dongjoon-hyun/SPARK-42277. Authored-by: Dongjoon Hyun <dongjoon@apache.org> Signed-off-by: Dongjoon Hyun <dongjoon@apache.org> (cherry picked from commit 39c9945) Signed-off-by: Dongjoon Hyun <dongjoon@apache.org>
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
5 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
What changes were proposed in this pull request?
This PR aims to use
RocksDBforspark.history.store.hybridStore.diskBackendby default instead ofLevelDB.The last update of
LevelDBis about over 10 years.all except aarch64: https://mvnrepository.com/artifact/org.fusesource.leveldbjni/leveldbjni-all/1.8 (Oct 17, 2013)aarch64-only: https://mvnrepository.com/artifact/org.openlabtesting.leveldbjni/leveldbjni-all/1.8 (Aug 28, 2019)Why are the changes needed?
RocksDB is maintained well and works on all environments where Apache Spark supports while LevelDB is not maintained at all in these days and doesn't work in some environment like the following.
Does this PR introduce any user-facing change?
Since
spark.history.store.hybridStore.enabledisfalseby default. There is no behavior change.In case of
spark.history.store.hybridStore.enabled=true, History Server will rebuild the new hybrid store simply.How was this patch tested?
Pass the CIs.