[ https://issues.apache.org/jira/browse/HBASE-17932?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15971011#comment-15971011 ]
Allan Yang edited comment on HBASE-17932 at 4/17/17 12:24 PM: -------------------------------------------------------------- Have you take gc into count? Those troughs in Figure1 may result of GCs. Big updates(100k) will increase gc burden obviously. In my opinion, even if client 1 and client 2 fights over CPU and IO resources, the total throughput should be stable. The reason why figure 2 is so smooth may because you use only 3 handler for big updates, reliefs the gc problem. I think more tests are needed. 1. test big updates only(test client1 in your case) 2. use 150 handlers for client1 and 3 handler for client2 3.use HBase2.0 which have off-heap write path was (Author: allan163): Have you take gc into count? Those troughs in Figure1 may result of GCs. Big updates(100k) will increase gc burden obviously. In my opinion, even if client 1 and client 2 fights over CPU and IO resources, the total throughput should be stable. The reason why figure 2 is so smooth may because you use only 3 handler for big updates, reliefs the gc problem. I think more tests are needed. 1. test big updates only(test client1 in your case) 2. use 150 handlers for client1 and 3 handler for client2 2.use HBase2.0 which have off-heap write path > add manageable call queue > ------------------------- > > Key: HBASE-17932 > URL: https://issues.apache.org/jira/browse/HBASE-17932 > Project: HBase > Issue Type: Improvement > Components: IPC/RPC > Reporter: xinxin fan > Fix For: 1.1.2 > > Attachments: ManageableCallQueue.pdf > > > The Manageable Call Queue aims to provide a unified way to build: > 1. Administor can create queue with specified number handler using shell > command > 2. Administor can increase/decrease the handler number for the specified > queue using shell command > 3. Administor can assign the same type request load to one queue, for > example, administor can assign all the large object write requests to one > queue, while assign all the small object write requests to another queue. > the feature allow you to change the ipc queues/handlers at runtime and is a > improvements for running multiple workloads on a single hbase cluster in some > cases. -- This message was sent by Atlassian JIRA (v6.3.15#6346)