You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@parquet.apache.org by "Juliet Hougland (JIRA)" <ji...@apache.org> on 2016/02/05 01:45:39 UTC
[jira] [Commented] (PARQUET-118) Provide option to use on-heap
buffers for Snappy compression/decompression
[ https://issues.apache.org/jira/browse/PARQUET-118?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15133426#comment-15133426 ]
Juliet Hougland commented on PARQUET-118:
-----------------------------------------
I've been working on [a patch|https://github.com/jhlch/parquet-mr/tree/bytebuffers] to allow on heap byte buffers for easier spark configuration. I've realized the [xerial-snappy itself explicitly disallows on heaps byte buffers.| https://github.com/xerial/snappy-java/blob/develop/src/main/java/org/xerial/snappy/Snappy.java#L136 ]The best I've got for handling this is trying to make this change there too. Any thoughts? any idea how receptive they may be?
> Provide option to use on-heap buffers for Snappy compression/decompression
> --------------------------------------------------------------------------
>
> Key: PARQUET-118
> URL: https://issues.apache.org/jira/browse/PARQUET-118
> Project: Parquet
> Issue Type: Improvement
> Components: parquet-mr
> Affects Versions: 1.6.0
> Reporter: Patrick Wendell
>
> The current code uses direct off-heap buffers for decompression. If many decompressors are instantiated across multiple threads, and/or the objects being decompressed are large, this can lead to a huge amount of off-heap allocation by the JVM. This can be exacerbated if overall, there is not heap contention, since no GC will be performed to reclaim the space used by these buffers.
> It would be nice if there was a flag we cold use to simply allocate on-heap buffers here:
> https://github.com/apache/incubator-parquet-mr/blob/master/parquet-hadoop/src/main/java/parquet/hadoop/codec/SnappyDecompressor.java#L28
> We ran into an issue today where these buffers totaled a very large amount of storage and caused our Java processes (running within containers) to be terminated by the kernel OOM-killer.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)