You are viewing a plain text version of this content. The canonical link for it is here.
Posted to reviews@spark.apache.org by sameeragarwal <gi...@git.apache.org> on 2016/09/02 20:11:21 UTC
[GitHub] spark pull request #14941: [SPARK-16334] Reusing same dictionary column for ...
GitHub user sameeragarwal opened a pull request:
https://github.com/apache/spark/pull/14941
[SPARK-16334] Reusing same dictionary column for decoding consecutive row groups shouldn't throw an error
## What changes were proposed in this pull request?
This patch fixes a bug in the vectorized parquet reader that's caused by re-using the same dictionary column vector while reading consecutive row groups. Specifically, this issue manifests for a certain distribution of dictionary/plain encoded data while we read/populate the underlying bit packed dictionary data into a column-vector based data structure.
## How was this patch tested?
Manually tested on datasets provided by the community. Thanks to Chris Perluss and Keith Kraus for their invaluable help in tracking down this issue!
You can merge this pull request into a Git repository by running:
$ git pull https://github.com/sameeragarwal/spark parquet-exception-2
Alternatively you can review and apply these changes as the patch at:
https://github.com/apache/spark/pull/14941.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 #14941
----
commit efda29864506b4a9eb716652e0fcf5cd705c9b4c
Author: Sameer Agarwal <sa...@cs.berkeley.edu>
Date: 2016-09-02T19:03:36Z
Reusing dictionary column vectors for reading consecutive row groups shouldn't throw an error
----
---
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 #14941: [SPARK-16334] Reusing same dictionary column for decodin...
Posted by SparkQA <gi...@git.apache.org>.
Github user SparkQA commented on the issue:
https://github.com/apache/spark/pull/14941
**[Test build #64870 has started](https://amplab.cs.berkeley.edu/jenkins/job/SparkPullRequestBuilder/64870/consoleFull)** for PR 14941 at commit [`efda298`](https://github.com/apache/spark/commit/efda29864506b4a9eb716652e0fcf5cd705c9b4c).
---
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 #14941: [SPARK-16334] Reusing same dictionary column for decodin...
Posted by sameeragarwal <gi...@git.apache.org>.
Github user sameeragarwal commented on the issue:
https://github.com/apache/spark/pull/14941
@heroldus @davies I'll try to benchmark the worse case performance regression for this special case (while reading row batches of all but one dictionary encoded pages). I'll let you know if we find a substantial regression.
---
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 #14941: [SPARK-16334] Reusing same dictionary column for decodin...
Posted by sameeragarwal <gi...@git.apache.org>.
Github user sameeragarwal commented on the issue:
https://github.com/apache/spark/pull/14941
cc @davies
---
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 #14941: [SPARK-16334] Reusing same dictionary column for decodin...
Posted by davies <gi...@git.apache.org>.
Github user davies commented on the issue:
https://github.com/apache/spark/pull/14941
Merging this into master and 2.0 branch, 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 issue #14941: [SPARK-16334] Reusing same dictionary column for decodin...
Posted by AmplabJenkins <gi...@git.apache.org>.
Github user AmplabJenkins commented on the issue:
https://github.com/apache/spark/pull/14941
Test PASSed.
Refer to this link for build results (access rights to CI server needed):
https://amplab.cs.berkeley.edu/jenkins//job/SparkPullRequestBuilder/64870/
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 #14941: [SPARK-16334] Reusing same dictionary column for decodin...
Posted by SparkQA <gi...@git.apache.org>.
Github user SparkQA commented on the issue:
https://github.com/apache/spark/pull/14941
**[Test build #64870 has finished](https://amplab.cs.berkeley.edu/jenkins/job/SparkPullRequestBuilder/64870/consoleFull)** for PR 14941 at commit [`efda298`](https://github.com/apache/spark/commit/efda29864506b4a9eb716652e0fcf5cd705c9b4c).
* 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 #14941: [SPARK-16334] Reusing same dictionary column for ...
Posted by asfgit <gi...@git.apache.org>.
Github user asfgit closed the pull request at:
https://github.com/apache/spark/pull/14941
---
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 #14941: [SPARK-16334] Reusing same dictionary column for decodin...
Posted by davies <gi...@git.apache.org>.
Github user davies commented on the issue:
https://github.com/apache/spark/pull/14941
LGTM, pending jenkins.
---
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 #14941: [SPARK-16334] Reusing same dictionary column for decodin...
Posted by heroldus <gi...@git.apache.org>.
Github user heroldus commented on the issue:
https://github.com/apache/spark/pull/14941
@davies Fine, thx.
---
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 #14941: [SPARK-16334] Reusing same dictionary column for decodin...
Posted by heroldus <gi...@git.apache.org>.
Github user heroldus commented on the issue:
https://github.com/apache/spark/pull/14941
@sameeragarwal: Do you expect any performace impact of this commit? It's an additional `if (!column.isNullAt(i))` for every single value read.
---
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 #14941: [SPARK-16334] Reusing same dictionary column for decodin...
Posted by davies <gi...@git.apache.org>.
Github user davies commented on the issue:
https://github.com/apache/spark/pull/14941
@heroldus decodeDictionaryIds() is only used when a batch across pages with different encoding (dictionary or plain), so it's not in the hot pass, I think the performance impact should be fine.
---
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 #14941: [SPARK-16334] Reusing same dictionary column for decodin...
Posted by AmplabJenkins <gi...@git.apache.org>.
Github user AmplabJenkins commented on the issue:
https://github.com/apache/spark/pull/14941
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