[jira] [Commented] (HBASE-9935) Slight perf improvement: Avoid KeyValue.getRowLength() at some places

[ https://issues.apache.org/jira/browse/HBASE-9935?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13818068#comment-13818068 ] Lars Hofhansl commented on HBASE-9935: -- There are some more cases like this in lazy s

[jira] [Commented] (HBASE-9935) Slight perf improvement: Avoid KeyValue.getRowLength() at some places

[ https://issues.apache.org/jira/browse/HBASE-9935?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13818117#comment-13818117 ] Jean-Marc Spaggiari commented on HBASE-9935: Nice one again. We should take mo

[jira] [Commented] (HBASE-9935) Slight perf improvement: Avoid KeyValue.getRowLength() at some places

[ https://issues.apache.org/jira/browse/HBASE-9935?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13818222#comment-13818222 ] Andrew Purtell commented on HBASE-9935: --- Nice, I've been perpetuating this antipatte

[jira] [Commented] (HBASE-9935) Slight perf improvement: Avoid KeyValue.getRowLength() at some places

[ https://issues.apache.org/jira/browse/HBASE-9935?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13818256#comment-13818256 ] Lars Hofhansl commented on HBASE-9935: -- [~jmspaggi], caching stuff because of a profi

[jira] [Commented] (HBASE-9935) Slight perf improvement: Avoid KeyValue.getRowLength() at some places

[ https://issues.apache.org/jira/browse/HBASE-9935?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13818272#comment-13818272 ] Lars Hofhansl commented on HBASE-9935: -- Meant to put a :) after "anti pattern" above.

[jira] [Commented] (HBASE-9935) Slight perf improvement: Avoid KeyValue.getRowLength() at some places

[ https://issues.apache.org/jira/browse/HBASE-9935?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13818282#comment-13818282 ] Lars Hofhansl commented on HBASE-9935: -- Looking at KeyValue. The overall static size

[jira] [Commented] (HBASE-9935) Slight perf improvement: Avoid KeyValue.getRowLength() at some places

[ https://issues.apache.org/jira/browse/HBASE-9935?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13818287#comment-13818287 ] Lars Hofhansl commented on HBASE-9935: -- Interestingly, when I remove the keyLength ca

[jira] [Commented] (HBASE-9935) Slight perf improvement: Avoid KeyValue.getRowLength() at some places

[ https://issues.apache.org/jira/browse/HBASE-9935?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13818605#comment-13818605 ] Lars Hofhansl commented on HBASE-9935: -- Did more some tests. The steady state of this

[jira] [Commented] (HBASE-9935) Slight perf improvement: Avoid KeyValue.getRowLength() at some places

[ https://issues.apache.org/jira/browse/HBASE-9935?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13818625#comment-13818625 ] Lars Hofhansl commented on HBASE-9935: -- I tried a big'ish patch. Didn't see any impro

[jira] [Commented] (HBASE-9935) Slight perf improvement: Avoid KeyValue.getRowLength() at some places

[ https://issues.apache.org/jira/browse/HBASE-9935?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13818957#comment-13818957 ] Jean-Marc Spaggiari commented on HBASE-9935: Might be good for the GC time. Al

[jira] [Commented] (HBASE-9935) Slight perf improvement: Avoid KeyValue.getRowLength() at some places

[ https://issues.apache.org/jira/browse/HBASE-9935?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13819172#comment-13819172 ] Lars Hofhansl commented on HBASE-9935: -- He yes can do :) I'll probably file another i

[jira] [Commented] (HBASE-9935) Slight perf improvement: Avoid KeyValue.getRowLength() at some places

[ https://issues.apache.org/jira/browse/HBASE-9935?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13819408#comment-13819408 ] stack commented on HBASE-9935: -- +1 on removing caching of a keylength that provides no discer

[jira] [Commented] (HBASE-9935) Slight perf improvement: Avoid KeyValue.getRowLength() at some places

[ https://issues.apache.org/jira/browse/HBASE-9935?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13819527#comment-13819527 ] Lars Hofhansl commented on HBASE-9935: -- Lemme do a bit more testing. I'll report back

[jira] [Commented] (HBASE-9935) Slight perf improvement: Avoid KeyValue.getRowLength() at some places

[ https://issues.apache.org/jira/browse/HBASE-9935?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13819682#comment-13819682 ] Lars Hofhansl commented on HBASE-9935: -- Tried on a 6 core (12 hw threads) machine. In

[jira] [Commented] (HBASE-9935) Slight perf improvement: Avoid KeyValue.getRowLength() at some places

[ https://issues.apache.org/jira/browse/HBASE-9935?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13819689#comment-13819689 ] Lars Hofhansl commented on HBASE-9935: -- Filed HBASE-9956. > Slight perf improvement: