You are viewing a plain text version of this content. The canonical link for it is here.
Posted to reviews@spark.apache.org by "vicennial (via GitHub)" <gi...@apache.org> on 2023/08/03 13:09:33 UTC
[GitHub] [spark] vicennial opened a new pull request, #42321: [SPARK-44657] Fix incorrect limit handling in ArrowBatchWithSchemaIterator and config parsing of CONNECT_GRPC_ARROW_MAX_BATCH_SIZE
vicennial opened a new pull request, #42321:
URL: https://github.com/apache/spark/pull/42321
<!--
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://spark.apache.org/contributing.html
2. Ensure you have added or run the appropriate tests for your PR: https://spark.apache.org/developer-tools.html
3. If the PR is unfinished, add '[WIP]' in your PR title, e.g., '[WIP][SPARK-XXXX] Your PR title ...'.
4. Be sure to keep the PR description updated to reflect all changes.
5. Please write your PR title to summarize what this PR proposes.
6. If possible, provide a concise example to reproduce the issue for a faster review.
7. If you want to add a new configuration, please read the guideline first for naming configurations in
'core/src/main/scala/org/apache/spark/internal/config/ConfigEntry.scala'.
8. If you want to add or modify an error type or message, please read the guideline first in
'core/src/main/resources/error/README.md'.
-->
### What changes were proposed in this pull request?
<!--
Please clarify what changes you are proposing. The purpose of this section is to outline the changes and how this PR fixes the issue.
If possible, please consider writing useful notes for better and faster reviews in your PR. See the examples below.
1. If you refactor some codes with changing classes, showing the class hierarchy will help reviewers.
2. If you fix some SQL features, you can provide some references of other DBMSes.
3. If there is design documentation, please add the link.
4. If there is a discussion in the mailing list, please add the link.
-->
Fixes the limit checking of `maxEstimatedBatchSize` and `maxRecordsPerBatch` to respect the more restrictive limit and fixes the config parsing of `CONNECT_GRPC_ARROW_MAX_BATCH_SIZE` by converting the value to bytes.
### Why are the changes needed?
<!--
Please clarify why the changes are needed. For instance,
1. If you propose a new API, clarify the use case for a new API.
2. If you fix a bug, you can clarify why it is a bug.
-->
Bugfix.
In the arrow writer [code](https://github.com/apache/spark/blob/6161bf44f40f8146ea4c115c788fd4eaeb128769/sql/core/src/main/scala/org/apache/spark/sql/execution/arrow/ArrowConverters.scala#L154-L163) , the conditions don’t seem to hold what the documentation says regd "maxBatchSize and maxRecordsPerBatch, respect whatever smaller" since it seems to actually respect the conf which is "larger" (i.e less restrictive) due to || operator.
Further, when the `CONNECT_GRPC_ARROW_MAX_BATCH_SIZE` conf is read, the value is not converted to bytes from MiB ([example](https://github.com/apache/spark/blob/3e5203c64c06cc8a8560dfa0fb6f52e74589b583/connector/connect/server/src/main/scala/org/apache/spark/sql/connect/execution/SparkConnectPlanExecution.scala#L103)).
### Does this PR introduce _any_ user-facing change?
<!--
Note that it means *any* user-facing change including all aspects such as the documentation fix.
If yes, please clarify the previous behavior and the change this PR proposes - provide the console output, description and/or an example to show the behavior difference if possible.
If possible, please also clarify if this is a user-facing change compared to the released Spark versions or within the unreleased branches such as master.
If no, write 'No'.
-->
No.
### How was this patch tested?
<!--
If tests were added, say they were added here. Please make sure to add some test cases that check the changes thoroughly including negative and positive cases if possible.
If it was tested in a way different from regular unit tests, please clarify how you tested step by step, ideally copy and paste-able, so that other reviewers can test and check, and descendants can verify in the future.
If tests were not added, please describe why they were not added and/or why it was difficult to add.
If benchmark tests were added, please run the benchmarks in GitHub Actions for the consistent environment, and the instructions could accord to: https://spark.apache.org/developer-tools.html#github-workflow-benchmarks.
-->
Existing tests.
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: reviews-unsubscribe@spark.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
---------------------------------------------------------------------
To unsubscribe, e-mail: reviews-unsubscribe@spark.apache.org
For additional commands, e-mail: reviews-help@spark.apache.org
[GitHub] [spark] heyihong commented on a diff in pull request #42321: [SPARK-44657][CONNECT] Fix incorrect limit handling in ArrowBatchWithSchemaIterator and config parsing of CONNECT_GRPC_ARROW_MAX_BATCH_SIZE
Posted by "heyihong (via GitHub)" <gi...@apache.org>.
heyihong commented on code in PR #42321:
URL: https://github.com/apache/spark/pull/42321#discussion_r1283195331
##########
connector/connect/server/src/main/scala/org/apache/spark/sql/connect/execution/SparkConnectPlanExecution.scala:
##########
@@ -100,7 +100,8 @@ private[execution] class SparkConnectPlanExecution(executeHolder: ExecuteHolder)
val maxRecordsPerBatch = spark.sessionState.conf.arrowMaxRecordsPerBatch
val timeZoneId = spark.sessionState.conf.sessionLocalTimeZone
// Conservatively sets it 70% because the size is not accurate but estimated.
- val maxBatchSize = (SparkEnv.get.conf.get(CONNECT_GRPC_ARROW_MAX_BATCH_SIZE) * 0.7).toLong
+ val maxBatchSize =
Review Comment:
Should we add some tests for preventing future regressions?
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: reviews-unsubscribe@spark.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
---------------------------------------------------------------------
To unsubscribe, e-mail: reviews-unsubscribe@spark.apache.org
For additional commands, e-mail: reviews-help@spark.apache.org
[GitHub] [spark] HyukjinKwon commented on pull request #42321: [SPARK-44657][CONNECT] Fix incorrect limit handling in ArrowBatchWithSchemaIterator and config parsing of CONNECT_GRPC_ARROW_MAX_BATCH_SIZE
Posted by "HyukjinKwon (via GitHub)" <gi...@apache.org>.
HyukjinKwon commented on PR #42321:
URL: https://github.com/apache/spark/pull/42321#issuecomment-1669160736
Merged to master and branch-3.5.
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: reviews-unsubscribe@spark.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
---------------------------------------------------------------------
To unsubscribe, e-mail: reviews-unsubscribe@spark.apache.org
For additional commands, e-mail: reviews-help@spark.apache.org
[GitHub] [spark] HyukjinKwon commented on pull request #42321: [SPARK-44657][CONNECT] Fix incorrect limit handling in ArrowBatchWithSchemaIterator and config parsing of CONNECT_GRPC_ARROW_MAX_BATCH_SIZE
Posted by "HyukjinKwon (via GitHub)" <gi...@apache.org>.
HyukjinKwon commented on PR #42321:
URL: https://github.com/apache/spark/pull/42321#issuecomment-1669163903
I also merged to branch-3.4 - there's a heavy conflict in tests so I remove the changes for now.
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: reviews-unsubscribe@spark.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
---------------------------------------------------------------------
To unsubscribe, e-mail: reviews-unsubscribe@spark.apache.org
For additional commands, e-mail: reviews-help@spark.apache.org
[GitHub] [spark] aakshintala commented on a diff in pull request #42321: [SPARK-44657][CONNECT] Fix incorrect limit handling in ArrowBatchWithSchemaIterator and config parsing of CONNECT_GRPC_ARROW_MAX_BATCH_SIZE
Posted by "aakshintala (via GitHub)" <gi...@apache.org>.
aakshintala commented on code in PR #42321:
URL: https://github.com/apache/spark/pull/42321#discussion_r1283327336
##########
connector/connect/server/src/main/scala/org/apache/spark/sql/connect/config/Connect.scala:
##########
@@ -49,9 +49,9 @@ object Connect {
val CONNECT_GRPC_ARROW_MAX_BATCH_SIZE =
ConfigBuilder("spark.connect.grpc.arrow.maxBatchSize")
.doc(
- "When using Apache Arrow, limit the maximum size of one arrow batch that " +
- "can be sent from server side to client side. Currently, we conservatively use 70% " +
- "of it because the size is not accurate but estimated.")
+ "When using Apache Arrow, limit the maximum size of one arrow batch, in MiB unless " +
+ "otherwise specified, that can be sent from server side to client side. Currently, we " +
+ "conservatively use 70% of it because the size is not accurate but estimated.")
.version("3.4.0")
.bytesConf(ByteUnit.MiB)
Review Comment:
Bytes would just be better. No need to convert later. This confit returning 4 is a fairly surprising sharp edge (although it makes sense in hindsight).
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: reviews-unsubscribe@spark.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
---------------------------------------------------------------------
To unsubscribe, e-mail: reviews-unsubscribe@spark.apache.org
For additional commands, e-mail: reviews-help@spark.apache.org
[GitHub] [spark] HyukjinKwon closed pull request #42321: [SPARK-44657][CONNECT] Fix incorrect limit handling in ArrowBatchWithSchemaIterator and config parsing of CONNECT_GRPC_ARROW_MAX_BATCH_SIZE
Posted by "HyukjinKwon (via GitHub)" <gi...@apache.org>.
HyukjinKwon closed pull request #42321: [SPARK-44657][CONNECT] Fix incorrect limit handling in ArrowBatchWithSchemaIterator and config parsing of CONNECT_GRPC_ARROW_MAX_BATCH_SIZE
URL: https://github.com/apache/spark/pull/42321
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: reviews-unsubscribe@spark.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
---------------------------------------------------------------------
To unsubscribe, e-mail: reviews-unsubscribe@spark.apache.org
For additional commands, e-mail: reviews-help@spark.apache.org
[GitHub] [spark] HyukjinKwon commented on pull request #42321: [SPARK-44657][CONNECT] Fix incorrect limit handling in ArrowBatchWithSchemaIterator and config parsing of CONNECT_GRPC_ARROW_MAX_BATCH_SIZE
Posted by "HyukjinKwon (via GitHub)" <gi...@apache.org>.
HyukjinKwon commented on PR #42321:
URL: https://github.com/apache/spark/pull/42321#issuecomment-1668686189
Let's fix up https://github.com/vicennial/spark/actions/runs/5789336753/job/15690218462 the linter. otherwise should be good to go.
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: reviews-unsubscribe@spark.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
---------------------------------------------------------------------
To unsubscribe, e-mail: reviews-unsubscribe@spark.apache.org
For additional commands, e-mail: reviews-help@spark.apache.org
[GitHub] [spark] aakshintala commented on a diff in pull request #42321: [SPARK-44657][CONNECT] Fix incorrect limit handling in ArrowBatchWithSchemaIterator and config parsing of CONNECT_GRPC_ARROW_MAX_BATCH_SIZE
Posted by "aakshintala (via GitHub)" <gi...@apache.org>.
aakshintala commented on code in PR #42321:
URL: https://github.com/apache/spark/pull/42321#discussion_r1283328536
##########
connector/connect/server/src/main/scala/org/apache/spark/sql/connect/execution/SparkConnectPlanExecution.scala:
##########
@@ -100,7 +100,8 @@ private[execution] class SparkConnectPlanExecution(executeHolder: ExecuteHolder)
val maxRecordsPerBatch = spark.sessionState.conf.arrowMaxRecordsPerBatch
val timeZoneId = spark.sessionState.conf.sessionLocalTimeZone
// Conservatively sets it 70% because the size is not accurate but estimated.
- val maxBatchSize = (SparkEnv.get.conf.get(CONNECT_GRPC_ARROW_MAX_BATCH_SIZE) * 0.7).toLong
+ val maxBatchSize =
Review Comment:
A large range query should be sufficient. That's what we use in most of the arrow tests.
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: reviews-unsubscribe@spark.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
---------------------------------------------------------------------
To unsubscribe, e-mail: reviews-unsubscribe@spark.apache.org
For additional commands, e-mail: reviews-help@spark.apache.org
[GitHub] [spark] vicennial commented on a diff in pull request #42321: [SPARK-44657] Fix incorrect limit handling in ArrowBatchWithSchemaIterator and config parsing of CONNECT_GRPC_ARROW_MAX_BATCH_SIZE
Posted by "vicennial (via GitHub)" <gi...@apache.org>.
vicennial commented on code in PR #42321:
URL: https://github.com/apache/spark/pull/42321#discussion_r1283183440
##########
connector/connect/server/src/main/scala/org/apache/spark/sql/connect/config/Connect.scala:
##########
@@ -49,9 +49,9 @@ object Connect {
val CONNECT_GRPC_ARROW_MAX_BATCH_SIZE =
ConfigBuilder("spark.connect.grpc.arrow.maxBatchSize")
.doc(
- "When using Apache Arrow, limit the maximum size of one arrow batch that " +
- "can be sent from server side to client side. Currently, we conservatively use 70% " +
- "of it because the size is not accurate but estimated.")
+ "When using Apache Arrow, limit the maximum size of one arrow batch, in MiB unless " +
+ "otherwise specified, that can be sent from server side to client side. Currently, we " +
+ "conservatively use 70% of it because the size is not accurate but estimated.")
.version("3.4.0")
.bytesConf(ByteUnit.MiB)
Review Comment:
We could also set this to accept values in bytes instead. No strong opinion here
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: reviews-unsubscribe@spark.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
---------------------------------------------------------------------
To unsubscribe, e-mail: reviews-unsubscribe@spark.apache.org
For additional commands, e-mail: reviews-help@spark.apache.org
[GitHub] [spark] vicennial commented on a diff in pull request #42321: [SPARK-44657][CONNECT] Fix incorrect limit handling in ArrowBatchWithSchemaIterator and config parsing of CONNECT_GRPC_ARROW_MAX_BATCH_SIZE
Posted by "vicennial (via GitHub)" <gi...@apache.org>.
vicennial commented on code in PR #42321:
URL: https://github.com/apache/spark/pull/42321#discussion_r1286332936
##########
connector/connect/server/src/main/scala/org/apache/spark/sql/connect/config/Connect.scala:
##########
@@ -48,13 +48,12 @@ object Connect {
val CONNECT_GRPC_ARROW_MAX_BATCH_SIZE =
ConfigBuilder("spark.connect.grpc.arrow.maxBatchSize")
- .doc(
- "When using Apache Arrow, limit the maximum size of one arrow batch that " +
- "can be sent from server side to client side. Currently, we conservatively use 70% " +
- "of it because the size is not accurate but estimated.")
+ .doc("When using Apache Arrow, limit the maximum size of one arrow batch, in bytes unless " +
+ "otherwise specified, that can be sent from server side to client side. Currently, we " +
+ "conservatively use 70% of it because the size is not accurate but estimated.")
.version("3.4.0")
- .bytesConf(ByteUnit.MiB)
- .createWithDefaultString("4m")
+ .bytesConf(ByteUnit.BYTE)
Review Comment:
@HyukjinKwon I've updated the conf to be Byte based instead of Mib (since the current impl actually assumed it was byte based)
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: reviews-unsubscribe@spark.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
---------------------------------------------------------------------
To unsubscribe, e-mail: reviews-unsubscribe@spark.apache.org
For additional commands, e-mail: reviews-help@spark.apache.org
[GitHub] [spark] vicennial commented on a diff in pull request #42321: [SPARK-44657][CONNECT] Fix incorrect limit handling in ArrowBatchWithSchemaIterator and config parsing of CONNECT_GRPC_ARROW_MAX_BATCH_SIZE
Posted by "vicennial (via GitHub)" <gi...@apache.org>.
vicennial commented on code in PR #42321:
URL: https://github.com/apache/spark/pull/42321#discussion_r1283200814
##########
connector/connect/server/src/main/scala/org/apache/spark/sql/connect/execution/SparkConnectPlanExecution.scala:
##########
@@ -100,7 +100,8 @@ private[execution] class SparkConnectPlanExecution(executeHolder: ExecuteHolder)
val maxRecordsPerBatch = spark.sessionState.conf.arrowMaxRecordsPerBatch
val timeZoneId = spark.sessionState.conf.sessionLocalTimeZone
// Conservatively sets it 70% because the size is not accurate but estimated.
- val maxBatchSize = (SparkEnv.get.conf.get(CONNECT_GRPC_ARROW_MAX_BATCH_SIZE) * 0.7).toLong
+ val maxBatchSize =
Review Comment:
That makes sense, yes. However, I am not super familiar with testing the arrow code path particularly with generating batches of a specific size (in order to trigger the limit). Perhaps, just a large range query should be sufficient? I'll try this our but pointers are appreciated
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: reviews-unsubscribe@spark.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
---------------------------------------------------------------------
To unsubscribe, e-mail: reviews-unsubscribe@spark.apache.org
For additional commands, e-mail: reviews-help@spark.apache.org
[GitHub] [spark] vicennial commented on a diff in pull request #42321: [SPARK-44657][CONNECT] Fix incorrect limit handling in ArrowBatchWithSchemaIterator and config parsing of CONNECT_GRPC_ARROW_MAX_BATCH_SIZE
Posted by "vicennial (via GitHub)" <gi...@apache.org>.
vicennial commented on code in PR #42321:
URL: https://github.com/apache/spark/pull/42321#discussion_r1283200814
##########
connector/connect/server/src/main/scala/org/apache/spark/sql/connect/execution/SparkConnectPlanExecution.scala:
##########
@@ -100,7 +100,8 @@ private[execution] class SparkConnectPlanExecution(executeHolder: ExecuteHolder)
val maxRecordsPerBatch = spark.sessionState.conf.arrowMaxRecordsPerBatch
val timeZoneId = spark.sessionState.conf.sessionLocalTimeZone
// Conservatively sets it 70% because the size is not accurate but estimated.
- val maxBatchSize = (SparkEnv.get.conf.get(CONNECT_GRPC_ARROW_MAX_BATCH_SIZE) * 0.7).toLong
+ val maxBatchSize =
Review Comment:
That makes sense, yes. However, I am not super familiar with testing the arrow code path particularly with generating batches of a specific size (in order to trigger the limit). Perhaps, just a large range query should be sufficient? I'll try this out but pointers are appreciated
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: reviews-unsubscribe@spark.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
---------------------------------------------------------------------
To unsubscribe, e-mail: reviews-unsubscribe@spark.apache.org
For additional commands, e-mail: reviews-help@spark.apache.org
[GitHub] [spark] vicennial commented on a diff in pull request #42321: [SPARK-44657][CONNECT] Fix incorrect limit handling in ArrowBatchWithSchemaIterator and config parsing of CONNECT_GRPC_ARROW_MAX_BATCH_SIZE
Posted by "vicennial (via GitHub)" <gi...@apache.org>.
vicennial commented on code in PR #42321:
URL: https://github.com/apache/spark/pull/42321#discussion_r1286333247
##########
connector/connect/server/src/main/scala/org/apache/spark/sql/connect/execution/SparkConnectPlanExecution.scala:
##########
@@ -100,7 +100,8 @@ private[execution] class SparkConnectPlanExecution(executeHolder: ExecuteHolder)
val maxRecordsPerBatch = spark.sessionState.conf.arrowMaxRecordsPerBatch
val timeZoneId = spark.sessionState.conf.sessionLocalTimeZone
// Conservatively sets it 70% because the size is not accurate but estimated.
- val maxBatchSize = (SparkEnv.get.conf.get(CONNECT_GRPC_ARROW_MAX_BATCH_SIZE) * 0.7).toLong
+ val maxBatchSize =
Review Comment:
Added!
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: reviews-unsubscribe@spark.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
---------------------------------------------------------------------
To unsubscribe, e-mail: reviews-unsubscribe@spark.apache.org
For additional commands, e-mail: reviews-help@spark.apache.org