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 2016/03/11 20:57:05 UTC
[GitHub] spark pull request: [SPARK-13830] prefer block manager than direct...
GitHub user davies opened a pull request:
https://github.com/apache/spark/pull/11659
[SPARK-13830] prefer block manager than direct result for large result
## What changes were proposed in this pull request?
The current RPC can't handle large blocks very well, it's very slow to fetch 100M block (about 1 minute). Once switch to block manager to fetch that, it took about 10 seconds (still could be improved).
## How was this patch tested?
existing unit tests.
You can merge this pull request into a Git repository by running:
$ git pull https://github.com/davies/spark direct_result
Alternatively you can review and apply these changes as the patch at:
https://github.com/apache/spark/pull/11659.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 #11659
----
commit e9826566b5f4b6c35941994fbc0f5c7aabe0b2e2
Author: Davies Liu <da...@databricks.com>
Date: 2016-03-11T19:53:43Z
prefer block manager than direct result for large result
----
---
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: [SPARK-13830] prefer block manager than direct...
Posted by andrewor14 <gi...@git.apache.org>.
Github user andrewor14 commented on the pull request:
https://github.com/apache/spark/pull/11659#issuecomment-195603451
LGTM
---
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: [SPARK-13830] prefer block manager than direct...
Posted by SparkQA <gi...@git.apache.org>.
Github user SparkQA commented on the pull request:
https://github.com/apache/spark/pull/11659#issuecomment-195606129
**[Test build #52950 has finished](https://amplab.cs.berkeley.edu/jenkins/job/SparkPullRequestBuilder/52950/consoleFull)** for PR 11659 at commit [`142b43e`](https://github.com/apache/spark/commit/142b43e56116af9b16899d16f7d375760c092ff7).
* This patch **fails PySpark 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: [SPARK-13830] prefer block manager than direct...
Posted by davies <gi...@git.apache.org>.
Github user davies commented on the pull request:
https://github.com/apache/spark/pull/11659#issuecomment-195522189
cc @zsxwing @andrewor14
---
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: [SPARK-13830] prefer block manager than direct...
Posted by asfgit <gi...@git.apache.org>.
Github user asfgit closed the pull request at:
https://github.com/apache/spark/pull/11659
---
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: [SPARK-13830] prefer block manager than direct...
Posted by AmplabJenkins <gi...@git.apache.org>.
Github user AmplabJenkins commented on the pull request:
https://github.com/apache/spark/pull/11659#issuecomment-195606316
Test FAILed.
Refer to this link for build results (access rights to CI server needed):
https://amplab.cs.berkeley.edu/jenkins//job/SparkPullRequestBuilder/52950/
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 pull request: [SPARK-13830] prefer block manager than direct...
Posted by SparkQA <gi...@git.apache.org>.
Github user SparkQA commented on the pull request:
https://github.com/apache/spark/pull/11659#issuecomment-195574451
**[Test build #52938 has finished](https://amplab.cs.berkeley.edu/jenkins/job/SparkPullRequestBuilder/52938/consoleFull)** for PR 11659 at commit [`e982656`](https://github.com/apache/spark/commit/e9826566b5f4b6c35941994fbc0f5c7aabe0b2e2).
* 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 pull request: [SPARK-13830] prefer block manager than direct...
Posted by davies <gi...@git.apache.org>.
Github user davies commented on the pull request:
https://github.com/apache/spark/pull/11659#issuecomment-195604813
Merged into master, thanks!
---
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: [SPARK-13830] prefer block manager than direct...
Posted by zsxwing <gi...@git.apache.org>.
Github user zsxwing commented on the pull request:
https://github.com/apache/spark/pull/11659#issuecomment-195604399
LGTM
---
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: [SPARK-13830] prefer block manager than direct...
Posted by SparkQA <gi...@git.apache.org>.
Github user SparkQA commented on the pull request:
https://github.com/apache/spark/pull/11659#issuecomment-195564685
**[Test build #52950 has started](https://amplab.cs.berkeley.edu/jenkins/job/SparkPullRequestBuilder/52950/consoleFull)** for PR 11659 at commit [`142b43e`](https://github.com/apache/spark/commit/142b43e56116af9b16899d16f7d375760c092ff7).
---
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: [SPARK-13830] prefer block manager than direct...
Posted by AmplabJenkins <gi...@git.apache.org>.
Github user AmplabJenkins commented on the pull request:
https://github.com/apache/spark/pull/11659#issuecomment-195606313
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 pull request: [SPARK-13830] prefer block manager than direct...
Posted by SparkQA <gi...@git.apache.org>.
Github user SparkQA commented on the pull request:
https://github.com/apache/spark/pull/11659#issuecomment-195522563
**[Test build #52938 has started](https://amplab.cs.berkeley.edu/jenkins/job/SparkPullRequestBuilder/52938/consoleFull)** for PR 11659 at commit [`e982656`](https://github.com/apache/spark/commit/e9826566b5f4b6c35941994fbc0f5c7aabe0b2e2).
---
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: [SPARK-13830] prefer block manager than direct...
Posted by AmplabJenkins <gi...@git.apache.org>.
Github user AmplabJenkins commented on the pull request:
https://github.com/apache/spark/pull/11659#issuecomment-195575304
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 pull request: [SPARK-13830] prefer block manager than direct...
Posted by andrewor14 <gi...@git.apache.org>.
Github user andrewor14 commented on a diff in the pull request:
https://github.com/apache/spark/pull/11659#discussion_r55889878
--- Diff: core/src/main/scala/org/apache/spark/executor/Executor.scala ---
@@ -97,9 +97,9 @@ private[spark] class Executor(
// Set the classloader for serializer
env.serializer.setDefaultClassLoader(replClassLoader)
- // Max RPC message size. If task result is bigger than this, we use the block manager
+ // Max size of direct result. If task result is bigger than this, we use the block manager
// to send the result back.
- private val maxRpcMessageSize = RpcUtils.maxMessageSizeBytes(conf)
+ private val maxDirectResultSize = Math.min(1L << 20, RpcUtils.maxMessageSizeBytes(conf))
--- End diff --
we should have another config for this instead. Also 1MB is very small, maybe 16 or 32MB?
---
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: [SPARK-13830] prefer block manager than direct...
Posted by AmplabJenkins <gi...@git.apache.org>.
Github user AmplabJenkins commented on the pull request:
https://github.com/apache/spark/pull/11659#issuecomment-195575309
Test PASSed.
Refer to this link for build results (access rights to CI server needed):
https://amplab.cs.berkeley.edu/jenkins//job/SparkPullRequestBuilder/52938/
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