You are viewing a plain text version of this content. The canonical link for it is here.
Posted to issues@hbase.apache.org by "stack (JIRA)" <ji...@apache.org> on 2018/07/12 02:55:00 UTC
[jira] [Commented] (HBASE-20875) MemStoreLABImp::copyIntoCell uses
7% CPU when writing
[ https://issues.apache.org/jira/browse/HBASE-20875?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16541066#comment-16541066 ]
stack commented on HBASE-20875:
-------------------------------
Let me upload a patch. Has the method now using 1.6% of CPU. Here is the top methods:
{code}
16.31% libjvm.so [.] ParNewGeneration::copy_to_survivor_space_avoiding_promotion_undo(ParScanThreadState*, oopDesc*, unsigned long, markOopDesc*)
11.97% perf-145414.map [.] Lorg/apache/hadoop/hbase/BBKVComparator;::compare
6.16% perf-145414.map [.] Lorg/apache/hadoop/hbase/BBKVComparator;::compare
4.41% perf-145414.map [.] Ljava/util/concurrent/ConcurrentSkipListMap;::doPut
3.93% perf-145414.map [.] Lorg/apache/hadoop/hbase/regionserver/AbstractMemStore;::add
2.86% perf-145414.map [.] Lorg/apache/hadoop/hbase/client/Put;::add
2.39% perf-145414.map [.] Lorg/apache/hadoop/hbase/regionserver/HRegion;::updateCellTimestamps
1.96% perf-145414.map [.] Lorg/apache/hadoop/hbase/regionserver/RSRpcServices;::checkCellSizeLimit
1.71% perf-145414.map [.] Lorg/apache/hadoop/hbase/regionserver/wal/AsyncProtobufLogWriter;::append
1.64% perf-145414.map [.] Lorg/apache/hadoop/hbase/regionserver/wal/AbstractFSWAL;::stampSequenceIdAndPublishToRingBuffer
1.61% perf-145414.map [.] Lorg/apache/hadoop/hbase/regionserver/MemStoreLABImpl;::copyBBECellInto
{code}
The new, purposed method name is copyBBECellInto, the equiv of the old copyCellInto.
> MemStoreLABImp::copyIntoCell uses 7% CPU when writing
> -----------------------------------------------------
>
> Key: HBASE-20875
> URL: https://issues.apache.org/jira/browse/HBASE-20875
> Project: HBase
> Issue Type: Sub-task
> Reporter: stack
> Priority: Major
> Attachments: Screen Shot 2018-07-11 at 9.52.46 PM.png
>
>
> Looks like this with a PE random write loading:
> {code}
> ./hbase/bin/hbase --config ~/conf_hbase org.apache.hadoop.hbase.PerformanceEvaluation --nomapred --presplit=40 --size=30 --columns=10 --valueSize=100 randomWrite 200
> {code}
> ... against a single server.
> {code}
> 12.47% perf-91935.map [.] Lorg/apache/hadoop/hbase/BBKVComparator;::compare
> 10.42% libjvm.so [.] ParNewGeneration::copy_to_survivor_space_avoiding_promotion_undo(ParScanThreadState*, oopDesc*, unsigned long, markOopDesc*)
> 6.78% perf-91935.map [.] Lorg/apache/hadoop/hbase/regionserver/MemStoreLABImpl;::copyCellInto
> ....
> {code}
> These are top CPU consumers using perf-map-agent ./bin/perf-java-top...
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)