[
https://issues.apache.org/jira/browse/HBASE-12917?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14291706#comment-14291706
]
Vikas Vishwakarma commented on HBASE-12917:
-------------------------------------------
Locally all tests succeeded. One observation is that
HFilePerformanceEvaluation(2.0.0) looks to be running very slow compared to
HFilePerformanceEvaluation(0.98.10), comparison given below:
HFilePerformanceEvaluation(2.0.0)
===============================
SequentialWriteBenchmark for 1000000 rows took 1401ms (for 0.98.10 761ms)
UniformRandomSmallScan for 1000000 rows took 19123ms (for 0.98.10 2271ms)
UniformRandomReadBenchmark for 1000000 rows took 260927ms (for 0.98.10 71717ms)
GaussianRandomReadBenchmark for 1000000 rows took 299131ms (for 0.98.10 79260ms)
SequentialReadBenchmark for 1000000 rows took 250ms (for 0.98.10 227ms)
> HFilePerformanceEvaluation Scan tests fail with StackOverflowError due to
> recursive loop in createCell function
> ---------------------------------------------------------------------------------------------------------------
>
> Key: HBASE-12917
> URL: https://issues.apache.org/jira/browse/HBASE-12917
> Project: HBase
> Issue Type: Bug
> Components: hbase
> Affects Versions: 2.0.0
> Environment: Ubuntu
> Reporter: Vikas Vishwakarma
> Assignee: Vikas Vishwakarma
> Fix For: 2.0.0
>
> Attachments: HBASE-12917.patch
>
>
> While trying to run HFilePerformanceEvaluation with hbase-2.0.0
> UniformRandomSmallScan is failing with StackOverflowError. Lookes related to
> the byte array to Cell based HFile changes.
> 2015-01-26 12:42:09,551 WARN [main] util.NativeCodeLoader: Unable to load
> native-hadoop library for your platform... using builtin-java classes where
> applicable
> 2015-01-26 12:42:09,760 INFO [main] hbase.HFilePerformanceEvaluation:
> Running SequentialWriteBenchmark for 1000000 rows.
> 2015-01-26 12:42:10,007 INFO [main] hfile.CacheConfig: CacheConfig:disabled
> 2015-01-26 12:42:11,560 INFO [main] hbase.HFilePerformanceEvaluation:
> Running SequentialWriteBenchmark for 1000000 rows took 1496ms.
> 2015-01-26 12:42:11,561 INFO [0] hbase.HFilePerformanceEvaluation: Running
> UniformRandomSmallScan for 1000000 rows.
> 2015-01-26 12:42:11,561 INFO [0] hfile.CacheConfig: CacheConfig:disabled
> Exception in thread "0" java.lang.StackOverflowError
> at
> org.apache.hadoop.hbase.HFilePerformanceEvaluation.createCell(HFilePerformanceEvaluation.java:78)
> at
> org.apache.hadoop.hbase.HFilePerformanceEvaluation.createCell(HFilePerformanceEvaluation.java:78)
> at
> org.apache.hadoop.hbase.HFilePerformanceEvaluation.createCell(HFilePerformanceEvaluation.java:78)
> at
> org.apache.hadoop.hbase.HFilePerformanceEvaluation.createCell(HFilePerformanceEvaluation.java:78)
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
