[
https://issues.apache.org/jira/browse/HBASE-19282?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16289140#comment-16289140
]
Anoop Sam John commented on HBASE-19282:
----------------------------------------
My intention for asking that was to do in a PoC way and measure the diff.. Ya
when MSLAB is on heap all are on heap only. The diff is whether u want to
waste the 2 MB chunks for much smaller sized index. The index is going to stick
there for some more time until flushes. Or else the MSLAB should have been
gives more space total considering the index chunk needs also.
Say for index we need a size of 0.5 MB. We use the 2 MB chunk for this. Any
way the total chunks in MSLAB pool is fixed. So this might cause a request for
chunk for storing the cells, to create an on demand one. So will end up making
a 2 MB on heap chunk instead we would have just created a 0.5 MB earlier.
On the sizing (RegionServerAccounting) we will have an issue. This is there
before also. I added that item to my TODO list. Just saying here.
By default we have enabled MSLAB pool now and the max size is 100%. Means the
pool can grow upto a size equal to the global memstore upper barrier. We have
the forced flushes there at this point. So just assume the workload is not
append/increment and there are no jumbo cells. All goes to MSLAB. Also the
CCM index takes the BBs from pool too. Now in accounting, we will consider
the actual index size ie 0.5 MB as per above eg:. The chunk been used is a 2
MB one and we can not really use the remaining 1.5 MB. So in sizing we have to
ideally increase by 2 MB once we create this index. I did not check patches in
detail how this is done. I know we have lots of thing to check an confirm in
this sizing part. Will do as my next task. Just saying here. If I did not
explain it well pls ask.
> CellChunkMap Benchmarking and User Interface
> --------------------------------------------
>
> Key: HBASE-19282
> URL: https://issues.apache.org/jira/browse/HBASE-19282
> Project: HBase
> Issue Type: Sub-task
> Reporter: Anastasia Braginsky
> Attachments: CCM Benchmarking.pdf
>
>
> We have made some experiments how working with CellChunkMap (CCM) influences
> the performance when running on-heap and off-heap. Based on those results it
> is suggested to tie the MSLAB usage (off-heap or on-heap) with CCM index
> usage.
--
This message was sent by Atlassian JIRA
(v6.4.14#64029)
