[ https://issues.apache.org/jira/browse/HBASE-9969?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13823210#comment-13823210 ]
Vladimir Rodionov commented on HBASE-9969: ------------------------------------------ Matt, yes, this is exactly what needs to be done. Most of the time, KVs comes in batches from a single scanner and the optimal assumption in this case is to check next from a top scanner first and compare it against second scanner's top (I am talking about loser-winner-tree here). > Improve KeyValueHeap using loser tree > ------------------------------------- > > Key: HBASE-9969 > URL: https://issues.apache.org/jira/browse/HBASE-9969 > Project: HBase > Issue Type: Improvement > Components: Performance, regionserver > Reporter: Chao Shi > Assignee: Chao Shi > Fix For: 0.98.0, 0.96.1 > > Attachments: hbase-9969.patch, hbase-9969.patch, > kvheap-benchmark.png, kvheap-benchmark.txt > > > LoserTree is the better data structure than binary heap. It saves half of the > comparisons on each next(), though the time complexity is on O(logN). > Currently A scan or get will go through two KeyValueHeaps, one is merging KVs > read from multiple HFiles in a single store, the other is merging results > from multiple stores. This patch should improve the both cases whenever CPU > is the bottleneck (e.g. scan with filter over cached blocks, HBASE-9811). > All of the optimization work is done in KeyValueHeap and does not change its > public interfaces. The new code looks more cleaner and simpler to understand. -- This message was sent by Atlassian JIRA (v6.1#6144)