Hey Stack, I just uploaded a patch on 5945. The BufferChain stuff is still not incorporated. I couldn't yet find a good reason in the current codebase to incorporate it. But as you said, it will be useful in the context of KeyValue. I have left some TODOs in the patch to accomodate that (and explained in the jira). Thanks, Devaraj.
On Mon, Jan 7, 2013 at 11:32 PM, Stack <[email protected]> wrote: > Hey boss. Put some comments up on the issue. Maybe write up more in the > issue a high-level on where you would like to drive the issue. As you say, > there is overlap with hbase-6521 it seems. The doc linked to on the latter > issue suggests rpc is preamble, header, then the protobuf Message (with > optional encoded data block of KeyValues). Ditto for the response. > Looking at 5945 this seems to be where you are going to? > > Good on you DD, > St.Ack > > > On Mon, Jan 7, 2013 at 2:20 PM, Devaraj Das <[email protected]> wrote: > >> Hey Stack, Todd, and others, any chance you guys could take a quick >> look at the last patch on hbase-5495. Although the patch is not >> complete (and deserves a good cleanup and double-checks on whether it >> killed all the buffer copies), it probably gives a good idea on the >> overall direction. The RPC.proto is some thing that one could start >> with, and then drill deeper into the other areas. I have only tried to >> minimize the copies for the request/response bodies. The copy for the >> header bits would probably be noise and the current patch doesn't try >> to reduce copies for those.. >> 5945 is related to 6521 (versioning), but I think 5945 is already an >> improvement over what exists and if folks agree, I could spend some >> extra time to get it in shape sooner (within a day or two).. >> Thanks, >> Devaraj. >>
