[ https://issues.apache.org/jira/browse/HBASE-16417?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15903755#comment-15903755 ]
stack commented on HBASE-16417: ------------------------------- bq. Any chance of using larger values (e.g. 2KB) ? Please be explicit on what you are asking for [~ted_yu] Asking for 'larger' values w/o why or fit criteria you would see satisfied is like asking for values that are 'ethical' or values that smell sweet. [~eshcar] Nice writeup. Those numbers are looking good (I like the write amplification improvements and the less GC...). (Trying to channel [~ted_yu]), how you think the numbers would change if 'default' 1k values were used? Would we see the same benefit do you think? On the 90th percentile degradation when BASIC, how many segments we talking .... 2 or 3 or more than this? In another issue we might dig in on why the degradation... I'd think read from Segments in parallel should be nice and prompt. Probably something dumb we are doing. For another issue. You thinking the numbers good enough to enable BASIC as default? Merging means more memory churn, more CPU expended.... would be cool if we could do w/o merge... Nice writeup. > In-Memory MemStore Policy for Flattening and Compactions > -------------------------------------------------------- > > Key: HBASE-16417 > URL: https://issues.apache.org/jira/browse/HBASE-16417 > Project: HBase > Issue Type: Sub-task > Reporter: Anastasia Braginsky > Assignee: Eshcar Hillel > Fix For: 2.0.0 > > Attachments: HBASE-16417-benchmarkresults-20161101.pdf, > HBASE-16417-benchmarkresults-20161110.pdf, > HBASE-16417-benchmarkresults-20161123.pdf, > HBASE-16417-benchmarkresults-20161205.pdf, > HBASE-16417-benchmarkresults-20170309.pdf > > -- This message was sent by Atlassian JIRA (v6.3.15#6346)