[jira] [Commented] (HBASE-16408) Handle On heap BucketCache size when HeapMemoryManager tunes memory
[ https://issues.apache.org/jira/browse/HBASE-16408?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15501156#comment-15501156 ] Anoop Sam John commented on HBASE-16408: As per the tests so far, even the new off heap BC itself will be faster than L1 cache specially on high load. This is mainly due to the fact that we can work with very less xmx then and as there are no copies, the GC activity will be much lesser. In case of on heap BC, the issue is our working set will be always high. So configuring the IHOP for G1GC will be tricky. Working set being so high will be GC problematic as per the tests. So what I feel is on heap BC should not be used at all. Specially now we have an off heap version which can be almost equal to L1 cache in lower workload and faster when high load. Ya mmaped is considered file mode way only. We will do a copy on the HFile block on hit. We have to do this also.. This block bytes will be read so many time in the read stack (Cell compares etc etc).. If we get more page faults, we will have horrible results. Thoughts boss? I plan to open a jira to remove on heap BC support. > Handle On heap BucketCache size when HeapMemoryManager tunes memory > --- > > Key: HBASE-16408 > URL: https://issues.apache.org/jira/browse/HBASE-16408 > Project: HBase > Issue Type: Sub-task >Reporter: Anoop Sam John >Assignee: Anoop Sam John > Fix For: 2.0.0 > > -- This message was sent by Atlassian JIRA (v6.3.4#6332)
[jira] [Commented] (HBASE-16408) Handle On heap BucketCache size when HeapMemoryManager tunes memory
[ https://issues.apache.org/jira/browse/HBASE-16408?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15496270#comment-15496270 ] stack commented on HBASE-16408: --- Will an on-heap bucketcache that does not make copies be faster than our lrublockcache? (Seems like it would be?) I'd say we can forbid it though to keep things simple (and given we don't have much practice doing onheap bucket cache). mmap'd is considered file-based? Its not looked on as onheap? > Handle On heap BucketCache size when HeapMemoryManager tunes memory > --- > > Key: HBASE-16408 > URL: https://issues.apache.org/jira/browse/HBASE-16408 > Project: HBase > Issue Type: Sub-task >Reporter: Anoop Sam John >Assignee: Anoop Sam John > Fix For: 2.0.0 > > -- This message was sent by Atlassian JIRA (v6.3.4#6332)
[jira] [Commented] (HBASE-16408) Handle On heap BucketCache size when HeapMemoryManager tunes memory
[ https://issues.apache.org/jira/browse/HBASE-16408?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15495603#comment-15495603 ] Anoop Sam John commented on HBASE-16408: As of now there is no problem (I was wrong).. Even if BucketCache is onheap, we dont consider that size at all in size calc for HeapMemory tuner.Now we have optimized off heap read so am wondering why we need on heap version of bucket cache. File and off heap mode is enough? Do some one really using BucketCache on heap? Off heap will perform equally or even better now. [~stack] how abt removing on heap Bucket cache mode? > Handle On heap BucketCache size when HeapMemoryManager tunes memory > --- > > Key: HBASE-16408 > URL: https://issues.apache.org/jira/browse/HBASE-16408 > Project: HBase > Issue Type: Sub-task >Reporter: Anoop Sam John >Assignee: Anoop Sam John > Fix For: 2.0.0 > > -- This message was sent by Atlassian JIRA (v6.3.4#6332)
[jira] [Commented] (HBASE-16408) Handle On heap BucketCache size when HeapMemoryManager tunes memory
[ https://issues.apache.org/jira/browse/HBASE-16408?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15422875#comment-15422875 ] stack commented on HBASE-16408: --- I see (I misread BucketCache as BlockCache) > Handle On heap BucketCache size when HeapMemoryManager tunes memory > --- > > Key: HBASE-16408 > URL: https://issues.apache.org/jira/browse/HBASE-16408 > Project: HBase > Issue Type: Sub-task >Reporter: Anoop Sam John >Assignee: Anoop Sam John > Fix For: 2.0.0 > > -- This message was sent by Atlassian JIRA (v6.3.4#6332)
[jira] [Commented] (HBASE-16408) Handle On heap BucketCache size when HeapMemoryManager tunes memory
[ https://issues.apache.org/jira/browse/HBASE-16408?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15422734#comment-15422734 ] Anoop Sam John commented on HBASE-16408: Ya. The HeapMemoryManager handle L1 cache only as of now. When we e have L2 cache as off heap, we might need to change size as per size change from tuner. > Handle On heap BucketCache size when HeapMemoryManager tunes memory > --- > > Key: HBASE-16408 > URL: https://issues.apache.org/jira/browse/HBASE-16408 > Project: HBase > Issue Type: Sub-task >Reporter: Anoop Sam John >Assignee: Anoop Sam John > Fix For: 2.0.0 > > -- This message was sent by Atlassian JIRA (v6.3.4#6332)
[jira] [Commented] (HBASE-16408) Handle On heap BucketCache size when HeapMemoryManager tunes memory
[ https://issues.apache.org/jira/browse/HBASE-16408?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15422629#comment-15422629 ] stack commented on HBASE-16408: --- I don't understand what this one is about [~anoop.hbase]? Doesn't HeapMemoryManager do this already? > Handle On heap BucketCache size when HeapMemoryManager tunes memory > --- > > Key: HBASE-16408 > URL: https://issues.apache.org/jira/browse/HBASE-16408 > Project: HBase > Issue Type: Sub-task >Reporter: Anoop Sam John >Assignee: Anoop Sam John > Fix For: 2.0.0 > > -- This message was sent by Atlassian JIRA (v6.3.4#6332)