Return-Path: X-Original-To: apmail-hbase-issues-archive@www.apache.org Delivered-To: apmail-hbase-issues-archive@www.apache.org Received: from mail.apache.org (hermes.apache.org [140.211.11.3]) by minotaur.apache.org (Postfix) with SMTP id B0DEC1008E for ; Wed, 20 Nov 2013 06:04:19 +0000 (UTC) Received: (qmail 17335 invoked by uid 500); 20 Nov 2013 06:03:59 -0000 Delivered-To: apmail-hbase-issues-archive@hbase.apache.org Received: (qmail 17131 invoked by uid 500); 20 Nov 2013 06:03:38 -0000 Mailing-List: contact issues-help@hbase.apache.org; run by ezmlm Precedence: bulk List-Help: List-Unsubscribe: List-Post: List-Id: Delivered-To: mailing list issues@hbase.apache.org Received: (qmail 17062 invoked by uid 99); 20 Nov 2013 06:03:30 -0000 Received: from arcas.apache.org (HELO arcas.apache.org) (140.211.11.28) by apache.org (qpsmtpd/0.29) with ESMTP; Wed, 20 Nov 2013 06:03:30 +0000 Date: Wed, 20 Nov 2013 06:03:30 +0000 (UTC) From: "Matt Corgan (JIRA)" To: issues@hbase.apache.org Message-ID: In-Reply-To: References: Subject: [jira] [Updated] (HBASE-9969) Improve KeyValueHeap using loser tree MIME-Version: 1.0 Content-Type: text/plain; charset=utf-8 Content-Transfer-Encoding: 7bit X-JIRA-FingerPrint: 30527f35849b9dde25b450d4833f0394 [ https://issues.apache.org/jira/browse/HBASE-9969?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Matt Corgan updated HBASE-9969: ------------------------------- Attachment: hbase-9969-pq-v2.patch KeyValueHeapBenchmark_v2.ods attaching hbase-9969-pq-v2.patch and KeyValueHeapBenchmark_v2.ods * this patch will hopefully apply (previous one accidentally had BenchmarkableKeyValueHeap.java in the src/test directory) * KeyValueScannerPriorityQueue is no longer based on PriorityQueue. it's now a custom implementation with class comments explaining * KeyValueScannerHeap (based on KeyValueScannerPriorityQueue) looks to beat the existing KeyValueHeap in all cases (tested up to 32 scanners) * LoserTree is usually faster with more scanners, but still slower when many consecutive KVs come from the same scanner > 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: 9969-0.94.txt, KeyValueHeapBenchmark_v1.ods, KeyValueHeapBenchmark_v2.ods, hbase-9969-pq-v1.patch, hbase-9969-pq-v2.patch, hbase-9969-v2.patch, hbase-9969-v3.patch, 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)