[ https://issues.apache.org/jira/browse/HBASE-14921?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15392497#comment-15392497 ]
stack commented on HBASE-14921: ------------------------------- Just did a reread of this whole issue. First, what is going on in here is wonderful. True, the issue is going on too long and starting to run away from us but it is a shining example of the best of collaboration; informed, data-based compares, accommodating, smart, respectful back-and-forth, detailed reviews, actual testing (and fixes) of posted patches, etc. You can't beat it. Second, all involved agree on the merit of these developments, their promise, and are trying to help land the patch. There is consensus that we should commit and then-address-outstanding-issues afterward but as I read it there seems to be a reluctance to take on the patch while it demonstrably slows down the default case -- i.e. when no duplicates -- and there is concern that we may not be able to recover the lost perf with the current approach. We could of course turn this feature 'off', by default, but most of us don't want to do that for reasons stated above (another is that [~ram_krish] and [~anoop.hbase] want to base some of their offheaping of write path on the work done here). Can I help in here? I can run some perf compares like [~ram_krish]'s? [~anoop.hbase] Mind repeating what your two concerns were here just so the discussion is contained (this issue is long now). > Memory optimizations > -------------------- > > Key: HBASE-14921 > URL: https://issues.apache.org/jira/browse/HBASE-14921 > Project: HBase > Issue Type: Sub-task > Affects Versions: 2.0.0 > Reporter: Eshcar Hillel > Assignee: Anastasia Braginsky > Attachments: CellBlocksSegmentInMemStore.pdf, > CellBlocksSegmentinthecontextofMemStore(1).pdf, HBASE-14921-V01.patch, > HBASE-14921-V02.patch, HBASE-14921-V03.patch, HBASE-14921-V04-CA-V02.patch, > HBASE-14921-V04-CA.patch, HBASE-14921-V05-CAO.patch, > HBASE-14921-V06-CAO.patch, InitialCellArrayMapEvaluation.pdf, > IntroductiontoNewFlatandCompactMemStore.pdf > > > Memory optimizations including compressed format representation and offheap > allocations -- This message was sent by Atlassian JIRA (v6.3.4#6332)