[ https://issues.apache.org/jira/browse/HBASE-15756?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15270186#comment-15270186 ]
stack commented on HBASE-15756: ------------------------------- I'd suggest you not waste your time making it pluggable [~aoxiang]. Just change our rpcserver be netty posting some numbers that show it has same basic performance and we'll get it in? One kink is what to do w/ the [~anoop.hbase] and [~ram_krish] work? These lads are busy making our write pipeline offheap. It starts with a pool of fixed-sized offheap buffers to take the request (and thereby avoid an alloccation by the socket). Going netty, IIRC, you get netty ByteBufs to work with as opposed to ByteBuffers. [~anoop.hbase] Any comment in here sir? > Pluggable RpcServer > ------------------- > > Key: HBASE-15756 > URL: https://issues.apache.org/jira/browse/HBASE-15756 > Project: HBase > Issue Type: Improvement > Reporter: binlijin > Assignee: binlijin > > Current we use a simple RpcServer, and can not configure and use other > implementation.This issue is to make the RpcServer pluggable, so we can make > other implementation for example netty rpc server. Patch will upload laterly -- This message was sent by Atlassian JIRA (v6.3.4#6332)