[ https://issues.apache.org/jira/browse/HADOOP-13144?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16994164#comment-16994164 ]
Janus Chow commented on HADOOP-13144: ------------------------------------- Attached path named HADOOP-13144. Our test logic is : Start 500 threads to call getBlockLocations on 1 directory with 100,000 parquet files, we use random policy to split the throughput to 2 Routers, the result is as follows. |HADOOP-13144|inner_patch|processingAvg(ms)|proxyAvg(ms)|rpcProcessingTime(ms)| |off|off|2.01,1.31|2.55,2.54|4.86,4.33| |on|off|4,4|0.99,0.75|4.88,5.11| |on|on|0.023,0.025|1.88,1.92|2.1,2.11| HADOOP-13144 helps a lot to reduce proxyAvg. > Enhancing IPC client throughput via multiple connections per user > ----------------------------------------------------------------- > > Key: HADOOP-13144 > URL: https://issues.apache.org/jira/browse/HADOOP-13144 > Project: Hadoop Common > Issue Type: Improvement > Components: ipc > Reporter: Jason Kace > Assignee: Íñigo Goiri > Priority: Minor > Attachments: HADOOP-13144-performance.patch, HADOOP-13144.000.patch, > HADOOP-13144.001.patch, HADOOP-13144.002.patch, HADOOP-13144.003.patch > > > The generic IPC client ({{org.apache.hadoop.ipc.Client}}) utilizes a single > connection thread for each {{ConnectionId}}. The {{ConnectionId}} is unique > to the connection's remote address, ticket and protocol. Each ConnectionId > is 1:1 mapped to a connection thread by the client via a map cache. > The result is to serialize all IPC read/write activity through a single > thread for a each user/ticket + address. If a single user makes repeated > calls (1k-100k/sec) to the same destination, the IPC client becomes a > bottleneck. -- This message was sent by Atlassian Jira (v8.3.4#803005) --------------------------------------------------------------------- To unsubscribe, e-mail: common-issues-unsubscr...@hadoop.apache.org For additional commands, e-mail: common-issues-h...@hadoop.apache.org