[ https://issues.apache.org/jira/browse/HBASE-16652?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15502190#comment-15502190 ]
Phil Yang edited comment on HBASE-16652 at 9/19/16 3:53 AM: ------------------------------------------------------------ In append we use arraycopy to merge the bytes and in increment we have to deserialize two Long (current and delta) and serialize the sum. Maybe this is the difference? And if the size of data in performance test is not large, all data is in memory, we have no disk/network io in reading and the difference is more obvious? was (Author: yangzhe1991): In append we use arraycopy to merge the bytes and in increment we have to deserialize two Long (current and delta) and serialize the sum. Maybe this is the difference? > Figure out performance difference between increment and append > -------------------------------------------------------------- > > Key: HBASE-16652 > URL: https://issues.apache.org/jira/browse/HBASE-16652 > Project: HBase > Issue Type: Sub-task > Reporter: Heng Chen > Labels: performance > Attachments: append.jfr, increment.jfr > > > When do performance test in HBASE-16625, i found it has the very big > difference between Append and Increment (append is about 37% faster than > increment). > As [~stack] mentioned in > https://issues.apache.org/jira/browse/HBASE-16610?focusedCommentId=15493166&page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel#comment-15493166, > append and increment has been unified in server-side, and they looks the > same in client-side. > This issue is to figure out why the performance looks different between them. -- This message was sent by Atlassian JIRA (v6.3.4#6332)