-
Notifications
You must be signed in to change notification settings - Fork 134
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
[ISSUE-342][Improvement] Check Spark Serializer type #344
Conversation
@jerqi Hi, would you mind to take a look? |
Please fix code style and compile error. |
Codecov Report
@@ Coverage Diff @@
## master #344 +/- ##
============================================
+ Coverage 58.17% 58.22% +0.04%
Complexity 1529 1529
============================================
Files 192 191 -1
Lines 10606 10597 -9
Branches 924 924
============================================
Hits 6170 6170
+ Misses 4068 4059 -9
Partials 368 368
📣 We’re building smart automated test selection to slash your CI/CD build times. Learn more |
client-spark/spark2/src/main/java/org/apache/spark/shuffle/RssShuffleManager.java
Show resolved
Hide resolved
you can click the failure pipeline https://github.com/apache/incubator-uniffle/actions/runs/3507622551/jobs/5875494928 and then you can see |
We should modify the test org.apache.uniffle.test.GetReaderTest, the test use the JavaSerializer. We should change it. https://github.com/apache/incubator-uniffle/actions/runs/3507622551/jobs/5875494954 |
You could import the uniffle codestyle into the IDEA, following this guide https://github.com/apache/incubator-uniffle/blob/master/CONTRIBUTING.md#code-style-guide @chong0929 |
5fd7907
to
ef7b303
Compare
ef7b303
to
c6f1ed6
Compare
Thanks for you review, add a default conf about spark.serializer which shoule be KryoSerializer when in PBS. |
Thanks you for your reminder and suggestion, I'm trying to make some changes. |
Spark3 also have org.apache.uniffle.test.GetReaderTest, you should also modify it. |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
LGTM, thanks @chong0929 @zuston . Great work!
What changes were proposed in this pull request?
Spark have multiple serializers. We support the spark serializer which supportsRelocationOfSerializedObjects.
You can see https://github.com/apache/spark/blob/25849684b78cca6651e25d6efc9644a576e7e20f/core/src/main/scala/org/apache/spark/serializer/Serializer.scala#L98
Spark have three kinds of serializer
org.apache.spark.serializer.JavaSerializer
org.apache.spark.sql.execution.UnsafeRowSerializer
org.apache.spark.serializer.KryoSerializer
Only org.apache.spark.serializer.JavaSerializer don't support RelocationOfSerializedObjects.
Why are the changes needed?
So when we find the parameters to use org.apache.spark.serializer.JavaSerializer, We should throw an exception.
Does this PR introduce any user-facing change?
No
How was this patch tested?
test locally