You are viewing a plain text version of this content. The canonical link for it is here.
Posted to reviews@spark.apache.org by davies <gi...@git.apache.org> on 2016/08/17 22:06:25 UTC
[GitHub] spark pull request #14692: decrease the threshold when split expressions
GitHub user davies opened a pull request:
https://github.com/apache/spark/pull/14692
decrease the threshold when split expressions
## What changes were proposed in this pull request?
In 2.0, we change the threshold of splitting expressions from 16K to 64K, which cause very bad performance on wide table, because the generated method can't be JIT compiled by default (above the limit of 8K bytecode).
This PR will decrease it to 1K, based on the benchmark results for a wide table with 400 columns of LongType.
## How was this patch tested?
Added benchmark suite.
You can merge this pull request into a Git repository by running:
$ git pull https://github.com/davies/spark split_exprs
Alternatively you can review and apply these changes as the patch at:
https://github.com/apache/spark/pull/14692.patch
To close this pull request, make a commit to your master/trunk branch
with (at least) the following in the commit message:
This closes #14692
----
commit 2ffe46f81723367af3e23c11ee05fe175ba61f69
Author: Davies Liu <da...@databricks.com>
Date: 2016-08-17T21:40:41Z
decrease the threshold when split expressions
----
---
If your project is set up for it, you can reply to this email and have your
reply appear on GitHub as well. If your project does not have this feature
enabled and wishes so, or if the feature is enabled but not working, please
contact infrastructure at infrastructure@apache.org or file a JIRA ticket
with INFRA.
---
---------------------------------------------------------------------
To unsubscribe, e-mail: reviews-unsubscribe@spark.apache.org
For additional commands, e-mail: reviews-help@spark.apache.org
[GitHub] spark issue #14692: [SPARK-17115] [SQL] decrease the threshold when split ex...
Posted by AmplabJenkins <gi...@git.apache.org>.
Github user AmplabJenkins commented on the issue:
https://github.com/apache/spark/pull/14692
Test FAILed.
Refer to this link for build results (access rights to CI server needed):
https://amplab.cs.berkeley.edu/jenkins//job/SparkPullRequestBuilder/63944/
Test FAILed.
---
If your project is set up for it, you can reply to this email and have your
reply appear on GitHub as well. If your project does not have this feature
enabled and wishes so, or if the feature is enabled but not working, please
contact infrastructure at infrastructure@apache.org or file a JIRA ticket
with INFRA.
---
---------------------------------------------------------------------
To unsubscribe, e-mail: reviews-unsubscribe@spark.apache.org
For additional commands, e-mail: reviews-help@spark.apache.org
[GitHub] spark issue #14692: [SPARK-17115] [SQL] decrease the threshold when split ex...
Posted by SparkQA <gi...@git.apache.org>.
Github user SparkQA commented on the issue:
https://github.com/apache/spark/pull/14692
**[Test build #63948 has started](https://amplab.cs.berkeley.edu/jenkins/job/SparkPullRequestBuilder/63948/consoleFull)** for PR 14692 at commit [`ad1d83f`](https://github.com/apache/spark/commit/ad1d83f52525679a9f987bd5fd0c79dea04692a0).
---
If your project is set up for it, you can reply to this email and have your
reply appear on GitHub as well. If your project does not have this feature
enabled and wishes so, or if the feature is enabled but not working, please
contact infrastructure at infrastructure@apache.org or file a JIRA ticket
with INFRA.
---
---------------------------------------------------------------------
To unsubscribe, e-mail: reviews-unsubscribe@spark.apache.org
For additional commands, e-mail: reviews-help@spark.apache.org
[GitHub] spark issue #14692: [SPARK-17115] [SQL] decrease the threshold when split ex...
Posted by AmplabJenkins <gi...@git.apache.org>.
Github user AmplabJenkins commented on the issue:
https://github.com/apache/spark/pull/14692
Test FAILed.
Refer to this link for build results (access rights to CI server needed):
https://amplab.cs.berkeley.edu/jenkins//job/SparkPullRequestBuilder/63942/
Test FAILed.
---
If your project is set up for it, you can reply to this email and have your
reply appear on GitHub as well. If your project does not have this feature
enabled and wishes so, or if the feature is enabled but not working, please
contact infrastructure at infrastructure@apache.org or file a JIRA ticket
with INFRA.
---
---------------------------------------------------------------------
To unsubscribe, e-mail: reviews-unsubscribe@spark.apache.org
For additional commands, e-mail: reviews-help@spark.apache.org
[GitHub] spark issue #14692: [SPARK-17115] [SQL] decrease the threshold when split ex...
Posted by SparkQA <gi...@git.apache.org>.
Github user SparkQA commented on the issue:
https://github.com/apache/spark/pull/14692
**[Test build #63942 has finished](https://amplab.cs.berkeley.edu/jenkins/job/SparkPullRequestBuilder/63942/consoleFull)** for PR 14692 at commit [`2ffe46f`](https://github.com/apache/spark/commit/2ffe46f81723367af3e23c11ee05fe175ba61f69).
* This patch **fails Spark unit tests**.
* This patch merges cleanly.
* This patch adds no public classes.
---
If your project is set up for it, you can reply to this email and have your
reply appear on GitHub as well. If your project does not have this feature
enabled and wishes so, or if the feature is enabled but not working, please
contact infrastructure at infrastructure@apache.org or file a JIRA ticket
with INFRA.
---
---------------------------------------------------------------------
To unsubscribe, e-mail: reviews-unsubscribe@spark.apache.org
For additional commands, e-mail: reviews-help@spark.apache.org
[GitHub] spark pull request #14692: [SPARK-17115] [SQL] decrease the threshold when s...
Posted by cloud-fan <gi...@git.apache.org>.
Github user cloud-fan commented on a diff in the pull request:
https://github.com/apache/spark/pull/14692#discussion_r75572921
--- Diff: sql/catalyst/src/main/scala/org/apache/spark/sql/catalyst/expressions/codegen/CodeGenerator.scala ---
@@ -584,15 +584,18 @@ class CodegenContext {
* @param expressions the codes to evaluate expressions.
*/
def splitExpressions(row: String, expressions: Seq[String]): String = {
- if (row == null) {
+ if (row == null || currentVars != null) {
--- End diff --
When will `row == null`? I understand `currentVars != null` means we are in whole stage codegen.
---
If your project is set up for it, you can reply to this email and have your
reply appear on GitHub as well. If your project does not have this feature
enabled and wishes so, or if the feature is enabled but not working, please
contact infrastructure at infrastructure@apache.org or file a JIRA ticket
with INFRA.
---
---------------------------------------------------------------------
To unsubscribe, e-mail: reviews-unsubscribe@spark.apache.org
For additional commands, e-mail: reviews-help@spark.apache.org
[GitHub] spark issue #14692: [SPARK-17115] [SQL] decrease the threshold when split ex...
Posted by SparkQA <gi...@git.apache.org>.
Github user SparkQA commented on the issue:
https://github.com/apache/spark/pull/14692
**[Test build #63948 has finished](https://amplab.cs.berkeley.edu/jenkins/job/SparkPullRequestBuilder/63948/consoleFull)** for PR 14692 at commit [`ad1d83f`](https://github.com/apache/spark/commit/ad1d83f52525679a9f987bd5fd0c79dea04692a0).
* This patch passes all tests.
* This patch merges cleanly.
* This patch adds no public classes.
---
If your project is set up for it, you can reply to this email and have your
reply appear on GitHub as well. If your project does not have this feature
enabled and wishes so, or if the feature is enabled but not working, please
contact infrastructure at infrastructure@apache.org or file a JIRA ticket
with INFRA.
---
---------------------------------------------------------------------
To unsubscribe, e-mail: reviews-unsubscribe@spark.apache.org
For additional commands, e-mail: reviews-help@spark.apache.org
[GitHub] spark issue #14692: [SPARK-17115] [SQL] decrease the threshold when split ex...
Posted by SparkQA <gi...@git.apache.org>.
Github user SparkQA commented on the issue:
https://github.com/apache/spark/pull/14692
**[Test build #63942 has started](https://amplab.cs.berkeley.edu/jenkins/job/SparkPullRequestBuilder/63942/consoleFull)** for PR 14692 at commit [`2ffe46f`](https://github.com/apache/spark/commit/2ffe46f81723367af3e23c11ee05fe175ba61f69).
---
If your project is set up for it, you can reply to this email and have your
reply appear on GitHub as well. If your project does not have this feature
enabled and wishes so, or if the feature is enabled but not working, please
contact infrastructure at infrastructure@apache.org or file a JIRA ticket
with INFRA.
---
---------------------------------------------------------------------
To unsubscribe, e-mail: reviews-unsubscribe@spark.apache.org
For additional commands, e-mail: reviews-help@spark.apache.org
[GitHub] spark pull request #14692: [SPARK-17115] [SQL] decrease the threshold when s...
Posted by davies <gi...@git.apache.org>.
Github user davies commented on a diff in the pull request:
https://github.com/apache/spark/pull/14692#discussion_r75615959
--- Diff: sql/catalyst/src/main/scala/org/apache/spark/sql/catalyst/expressions/codegen/CodeGenerator.scala ---
@@ -584,15 +584,18 @@ class CodegenContext {
* @param expressions the codes to evaluate expressions.
*/
def splitExpressions(row: String, expressions: Seq[String]): String = {
- if (row == null) {
+ if (row == null || currentVars != null) {
--- End diff --
Mostly they have same meaning, but sometimes we forget to set INPUT as null in whole stage codegen (currentVars has higher priority)
---
If your project is set up for it, you can reply to this email and have your
reply appear on GitHub as well. If your project does not have this feature
enabled and wishes so, or if the feature is enabled but not working, please
contact infrastructure at infrastructure@apache.org or file a JIRA ticket
with INFRA.
---
---------------------------------------------------------------------
To unsubscribe, e-mail: reviews-unsubscribe@spark.apache.org
For additional commands, e-mail: reviews-help@spark.apache.org
[GitHub] spark issue #14692: [SPARK-17115] [SQL] decrease the threshold when split ex...
Posted by AmplabJenkins <gi...@git.apache.org>.
Github user AmplabJenkins commented on the issue:
https://github.com/apache/spark/pull/14692
Merged build finished. Test FAILed.
---
If your project is set up for it, you can reply to this email and have your
reply appear on GitHub as well. If your project does not have this feature
enabled and wishes so, or if the feature is enabled but not working, please
contact infrastructure at infrastructure@apache.org or file a JIRA ticket
with INFRA.
---
---------------------------------------------------------------------
To unsubscribe, e-mail: reviews-unsubscribe@spark.apache.org
For additional commands, e-mail: reviews-help@spark.apache.org
[GitHub] spark issue #14692: [SPARK-17115] [SQL] decrease the threshold when split ex...
Posted by SparkQA <gi...@git.apache.org>.
Github user SparkQA commented on the issue:
https://github.com/apache/spark/pull/14692
**[Test build #63944 has started](https://amplab.cs.berkeley.edu/jenkins/job/SparkPullRequestBuilder/63944/consoleFull)** for PR 14692 at commit [`a42a734`](https://github.com/apache/spark/commit/a42a73453e4c660d765192442f36d5595f8fc123).
---
If your project is set up for it, you can reply to this email and have your
reply appear on GitHub as well. If your project does not have this feature
enabled and wishes so, or if the feature is enabled but not working, please
contact infrastructure at infrastructure@apache.org or file a JIRA ticket
with INFRA.
---
---------------------------------------------------------------------
To unsubscribe, e-mail: reviews-unsubscribe@spark.apache.org
For additional commands, e-mail: reviews-help@spark.apache.org
[GitHub] spark issue #14692: [SPARK-17115] [SQL] decrease the threshold when split ex...
Posted by AmplabJenkins <gi...@git.apache.org>.
Github user AmplabJenkins commented on the issue:
https://github.com/apache/spark/pull/14692
Merged build finished. Test PASSed.
---
If your project is set up for it, you can reply to this email and have your
reply appear on GitHub as well. If your project does not have this feature
enabled and wishes so, or if the feature is enabled but not working, please
contact infrastructure at infrastructure@apache.org or file a JIRA ticket
with INFRA.
---
---------------------------------------------------------------------
To unsubscribe, e-mail: reviews-unsubscribe@spark.apache.org
For additional commands, e-mail: reviews-help@spark.apache.org
[GitHub] spark issue #14692: [SPARK-17115] [SQL] decrease the threshold when split ex...
Posted by AmplabJenkins <gi...@git.apache.org>.
Github user AmplabJenkins commented on the issue:
https://github.com/apache/spark/pull/14692
Merged build finished. Test FAILed.
---
If your project is set up for it, you can reply to this email and have your
reply appear on GitHub as well. If your project does not have this feature
enabled and wishes so, or if the feature is enabled but not working, please
contact infrastructure at infrastructure@apache.org or file a JIRA ticket
with INFRA.
---
---------------------------------------------------------------------
To unsubscribe, e-mail: reviews-unsubscribe@spark.apache.org
For additional commands, e-mail: reviews-help@spark.apache.org
[GitHub] spark issue #14692: [SPARK-17115] [SQL] decrease the threshold when split ex...
Posted by davies <gi...@git.apache.org>.
Github user davies commented on the issue:
https://github.com/apache/spark/pull/14692
@cloud-fan Could you review this one?
---
If your project is set up for it, you can reply to this email and have your
reply appear on GitHub as well. If your project does not have this feature
enabled and wishes so, or if the feature is enabled but not working, please
contact infrastructure at infrastructure@apache.org or file a JIRA ticket
with INFRA.
---
---------------------------------------------------------------------
To unsubscribe, e-mail: reviews-unsubscribe@spark.apache.org
For additional commands, e-mail: reviews-help@spark.apache.org
[GitHub] spark pull request #14692: [SPARK-17115] [SQL] decrease the threshold when s...
Posted by asfgit <gi...@git.apache.org>.
Github user asfgit closed the pull request at:
https://github.com/apache/spark/pull/14692
---
If your project is set up for it, you can reply to this email and have your
reply appear on GitHub as well. If your project does not have this feature
enabled and wishes so, or if the feature is enabled but not working, please
contact infrastructure at infrastructure@apache.org or file a JIRA ticket
with INFRA.
---
---------------------------------------------------------------------
To unsubscribe, e-mail: reviews-unsubscribe@spark.apache.org
For additional commands, e-mail: reviews-help@spark.apache.org
[GitHub] spark issue #14692: [SPARK-17115] [SQL] decrease the threshold when split ex...
Posted by cloud-fan <gi...@git.apache.org>.
Github user cloud-fan commented on the issue:
https://github.com/apache/spark/pull/14692
thanks, merging to master/2.0
---
If your project is set up for it, you can reply to this email and have your
reply appear on GitHub as well. If your project does not have this feature
enabled and wishes so, or if the feature is enabled but not working, please
contact infrastructure at infrastructure@apache.org or file a JIRA ticket
with INFRA.
---
---------------------------------------------------------------------
To unsubscribe, e-mail: reviews-unsubscribe@spark.apache.org
For additional commands, e-mail: reviews-help@spark.apache.org
[GitHub] spark issue #14692: [SPARK-17115] [SQL] decrease the threshold when split ex...
Posted by AmplabJenkins <gi...@git.apache.org>.
Github user AmplabJenkins commented on the issue:
https://github.com/apache/spark/pull/14692
Test PASSed.
Refer to this link for build results (access rights to CI server needed):
https://amplab.cs.berkeley.edu/jenkins//job/SparkPullRequestBuilder/63948/
Test PASSed.
---
If your project is set up for it, you can reply to this email and have your
reply appear on GitHub as well. If your project does not have this feature
enabled and wishes so, or if the feature is enabled but not working, please
contact infrastructure at infrastructure@apache.org or file a JIRA ticket
with INFRA.
---
---------------------------------------------------------------------
To unsubscribe, e-mail: reviews-unsubscribe@spark.apache.org
For additional commands, e-mail: reviews-help@spark.apache.org
[GitHub] spark issue #14692: [SPARK-17115] [SQL] decrease the threshold when split ex...
Posted by SparkQA <gi...@git.apache.org>.
Github user SparkQA commented on the issue:
https://github.com/apache/spark/pull/14692
**[Test build #63944 has finished](https://amplab.cs.berkeley.edu/jenkins/job/SparkPullRequestBuilder/63944/consoleFull)** for PR 14692 at commit [`a42a734`](https://github.com/apache/spark/commit/a42a73453e4c660d765192442f36d5595f8fc123).
* This patch **fails Spark unit tests**.
* This patch merges cleanly.
* This patch adds no public classes.
---
If your project is set up for it, you can reply to this email and have your
reply appear on GitHub as well. If your project does not have this feature
enabled and wishes so, or if the feature is enabled but not working, please
contact infrastructure at infrastructure@apache.org or file a JIRA ticket
with INFRA.
---
---------------------------------------------------------------------
To unsubscribe, e-mail: reviews-unsubscribe@spark.apache.org
For additional commands, e-mail: reviews-help@spark.apache.org