You are viewing a plain text version of this content. The canonical link for it is here.
Posted to reviews@kudu.apache.org by "Alexey Serbin (Code Review)" <ge...@cloudera.org> on 2018/02/21 23:17:20 UTC
[kudu-CR] [consensus peers] micro-optimization on controller status
Alexey Serbin has uploaded this change for review. ( http://gerrit.cloudera.org:8080/9387
Change subject: [consensus_peers] micro-optimization on controller status
......................................................................
[consensus_peers] micro-optimization on controller status
While running raft_consensus_stress-itest I found the following stats
for operator new[](unsigned long) in tserver with regard of
Status::CopyState():
- 1.13% kudu-tserver libtcmalloc.so.4.5.1 [.] operator new[](unsigned long)
- operator new[](unsigned long)
...
+ 3.21% kudu::Status::CopyState(char const*)
kudu::Status::Status(kudu::Status const&)
kudu::rpc::OutboundCall::status() const
kudu::rpc::RpcController::status() const
kudu::consensus::Peer::ProcessResponse()
...
This tiny patch adds micro-optimization to avoid calling copying
Status multiple times in consensus::Peer::ProcessResponse().
Change-Id: I25ac21db191a21f3bdfb3378e32fbe366d98297c
---
M src/kudu/consensus/consensus_peers.cc
1 file changed, 9 insertions(+), 7 deletions(-)
git pull ssh://gerrit.cloudera.org:29418/kudu refs/changes/87/9387/1
--
To view, visit http://gerrit.cloudera.org:8080/9387
To unsubscribe, visit http://gerrit.cloudera.org:8080/settings
Gerrit-Project: kudu
Gerrit-Branch: master
Gerrit-MessageType: newchange
Gerrit-Change-Id: I25ac21db191a21f3bdfb3378e32fbe366d98297c
Gerrit-Change-Number: 9387
Gerrit-PatchSet: 1
Gerrit-Owner: Alexey Serbin <as...@cloudera.com>
[kudu-CR] [consensus peers] micro-optimization on controller status
Posted by "Todd Lipcon (Code Review)" <ge...@cloudera.org>.
Todd Lipcon has posted comments on this change. ( http://gerrit.cloudera.org:8080/9387 )
Change subject: [consensus_peers] micro-optimization on controller status
......................................................................
Patch Set 2:
sure, seems fine, but for 0.03% of the CPU in an error-injected workload I'm not sure it's that much of a win :)
--
To view, visit http://gerrit.cloudera.org:8080/9387
To unsubscribe, visit http://gerrit.cloudera.org:8080/settings
Gerrit-Project: kudu
Gerrit-Branch: master
Gerrit-MessageType: comment
Gerrit-Change-Id: I25ac21db191a21f3bdfb3378e32fbe366d98297c
Gerrit-Change-Number: 9387
Gerrit-PatchSet: 2
Gerrit-Owner: Alexey Serbin <as...@cloudera.com>
Gerrit-Reviewer: Alexey Serbin <as...@cloudera.com>
Gerrit-Reviewer: Andrew Wong <aw...@cloudera.com>
Gerrit-Reviewer: Kudu Jenkins
Gerrit-Reviewer: Todd Lipcon <to...@apache.org>
Gerrit-Comment-Date: Thu, 22 Feb 2018 05:05:48 +0000
Gerrit-HasComments: No
[kudu-CR] [consensus peers] micro-optimization on controller status
Posted by "Alexey Serbin (Code Review)" <ge...@cloudera.org>.
Alexey Serbin has submitted this change and it was merged. ( http://gerrit.cloudera.org:8080/9387 )
Change subject: [consensus_peers] micro-optimization on controller status
......................................................................
[consensus_peers] micro-optimization on controller status
While running raft_consensus_stress-itest where a lot of RPC fail,
I found the following stats for operator new[](unsigned long)
in tserver with regard of Status::CopyState():
- 1.13% kudu-tserver libtcmalloc.so.4.5.1 [.] operator new[](unsigned long)
- operator new[](unsigned long)
...
+ 3.21% kudu::Status::CopyState(char const*)
kudu::Status::Status(kudu::Status const&)
kudu::rpc::OutboundCall::status() const
kudu::rpc::RpcController::status() const
kudu::consensus::Peer::ProcessResponse()
...
Of course, that's not in the hot path in case of success scenarios,
but I think it makes sense to add these micro-optimizations.
Change-Id: I25ac21db191a21f3bdfb3378e32fbe366d98297c
Reviewed-on: http://gerrit.cloudera.org:8080/9387
Tested-by: Kudu Jenkins
Reviewed-by: Andrew Wong <aw...@cloudera.com>
Reviewed-by: Todd Lipcon <to...@apache.org>
---
M src/kudu/consensus/consensus_peers.cc
1 file changed, 9 insertions(+), 7 deletions(-)
Approvals:
Kudu Jenkins: Verified
Andrew Wong: Looks good to me, but someone else must approve
Todd Lipcon: Looks good to me, approved
--
To view, visit http://gerrit.cloudera.org:8080/9387
To unsubscribe, visit http://gerrit.cloudera.org:8080/settings
Gerrit-Project: kudu
Gerrit-Branch: master
Gerrit-MessageType: merged
Gerrit-Change-Id: I25ac21db191a21f3bdfb3378e32fbe366d98297c
Gerrit-Change-Number: 9387
Gerrit-PatchSet: 3
Gerrit-Owner: Alexey Serbin <as...@cloudera.com>
Gerrit-Reviewer: Alexey Serbin <as...@cloudera.com>
Gerrit-Reviewer: Andrew Wong <aw...@cloudera.com>
Gerrit-Reviewer: Kudu Jenkins
Gerrit-Reviewer: Todd Lipcon <to...@apache.org>
[kudu-CR] [consensus peers] micro-optimization on controller status
Posted by "Andrew Wong (Code Review)" <ge...@cloudera.org>.
Andrew Wong has posted comments on this change. ( http://gerrit.cloudera.org:8080/9387 )
Change subject: [consensus_peers] micro-optimization on controller status
......................................................................
Patch Set 2: Code-Review+1
(1 comment)
http://gerrit.cloudera.org:8080/#/c/9387/2//COMMIT_MSG
Commit Message:
http://gerrit.cloudera.org:8080/#/c/9387/2//COMMIT_MSG@13
PS2, Line 13: 1.13% kudu-tserver libtcmalloc.so.4.5.1 [.] operator new[](unsigned long)
: - operator new[](unsigned long)
: ...
: + 3.21% kudu::Status::CopyState(char const*)
: kudu::Status::Status(kudu::Status const&)
: kudu::rpc::OutboundCall::status() const
: kudu::rpc::RpcController::status() const
: kudu::consensus::Peer::ProcessResponse()
: ...
Are these from `perf`?
--
To view, visit http://gerrit.cloudera.org:8080/9387
To unsubscribe, visit http://gerrit.cloudera.org:8080/settings
Gerrit-Project: kudu
Gerrit-Branch: master
Gerrit-MessageType: comment
Gerrit-Change-Id: I25ac21db191a21f3bdfb3378e32fbe366d98297c
Gerrit-Change-Number: 9387
Gerrit-PatchSet: 2
Gerrit-Owner: Alexey Serbin <as...@cloudera.com>
Gerrit-Reviewer: Alexey Serbin <as...@cloudera.com>
Gerrit-Reviewer: Andrew Wong <aw...@cloudera.com>
Gerrit-Reviewer: Kudu Jenkins
Gerrit-Reviewer: Todd Lipcon <to...@apache.org>
Gerrit-Comment-Date: Thu, 22 Feb 2018 04:33:45 +0000
Gerrit-HasComments: Yes
[kudu-CR] [consensus peers] micro-optimization on controller status
Posted by "Todd Lipcon (Code Review)" <ge...@cloudera.org>.
Todd Lipcon has posted comments on this change. ( http://gerrit.cloudera.org:8080/9387 )
Change subject: [consensus_peers] micro-optimization on controller status
......................................................................
Patch Set 2: Code-Review+2
--
To view, visit http://gerrit.cloudera.org:8080/9387
To unsubscribe, visit http://gerrit.cloudera.org:8080/settings
Gerrit-Project: kudu
Gerrit-Branch: master
Gerrit-MessageType: comment
Gerrit-Change-Id: I25ac21db191a21f3bdfb3378e32fbe366d98297c
Gerrit-Change-Number: 9387
Gerrit-PatchSet: 2
Gerrit-Owner: Alexey Serbin <as...@cloudera.com>
Gerrit-Reviewer: Alexey Serbin <as...@cloudera.com>
Gerrit-Reviewer: Andrew Wong <aw...@cloudera.com>
Gerrit-Reviewer: Kudu Jenkins
Gerrit-Reviewer: Todd Lipcon <to...@apache.org>
Gerrit-Comment-Date: Thu, 22 Feb 2018 05:05:06 +0000
Gerrit-HasComments: No
[kudu-CR] [consensus peers] micro-optimization on controller status
Posted by "Alexey Serbin (Code Review)" <ge...@cloudera.org>.
Hello Kudu Jenkins, Todd Lipcon,
I'd like you to reexamine a change. Please visit
http://gerrit.cloudera.org:8080/9387
to look at the new patch set (#2).
Change subject: [consensus_peers] micro-optimization on controller status
......................................................................
[consensus_peers] micro-optimization on controller status
While running raft_consensus_stress-itest where a lot of RPC fail,
I found the following stats for operator new[](unsigned long)
in tserver with regard of Status::CopyState():
- 1.13% kudu-tserver libtcmalloc.so.4.5.1 [.] operator new[](unsigned long)
- operator new[](unsigned long)
...
+ 3.21% kudu::Status::CopyState(char const*)
kudu::Status::Status(kudu::Status const&)
kudu::rpc::OutboundCall::status() const
kudu::rpc::RpcController::status() const
kudu::consensus::Peer::ProcessResponse()
...
Of course, that's not in the hot path in case of success scenarios,
but I think it makes sense to add these micro-optimizations.
Change-Id: I25ac21db191a21f3bdfb3378e32fbe366d98297c
---
M src/kudu/consensus/consensus_peers.cc
1 file changed, 9 insertions(+), 7 deletions(-)
git pull ssh://gerrit.cloudera.org:29418/kudu refs/changes/87/9387/2
--
To view, visit http://gerrit.cloudera.org:8080/9387
To unsubscribe, visit http://gerrit.cloudera.org:8080/settings
Gerrit-Project: kudu
Gerrit-Branch: master
Gerrit-MessageType: newpatchset
Gerrit-Change-Id: I25ac21db191a21f3bdfb3378e32fbe366d98297c
Gerrit-Change-Number: 9387
Gerrit-PatchSet: 2
Gerrit-Owner: Alexey Serbin <as...@cloudera.com>
Gerrit-Reviewer: Alexey Serbin <as...@cloudera.com>
Gerrit-Reviewer: Kudu Jenkins
Gerrit-Reviewer: Todd Lipcon <to...@apache.org>
[kudu-CR] [consensus peers] micro-optimization on controller status
Posted by "Alexey Serbin (Code Review)" <ge...@cloudera.org>.
Alexey Serbin has posted comments on this change. ( http://gerrit.cloudera.org:8080/9387 )
Change subject: [consensus_peers] micro-optimization on controller status
......................................................................
Patch Set 2:
> Patch Set 2:
>
> sure, seems fine, but for 0.03% of the CPU in an error-injected workload I'm not sure it's that much of a win :)
Thank you for the review.
Nope, it's not a win, just a tiny clean-up :)
--
To view, visit http://gerrit.cloudera.org:8080/9387
To unsubscribe, visit http://gerrit.cloudera.org:8080/settings
Gerrit-Project: kudu
Gerrit-Branch: master
Gerrit-MessageType: comment
Gerrit-Change-Id: I25ac21db191a21f3bdfb3378e32fbe366d98297c
Gerrit-Change-Number: 9387
Gerrit-PatchSet: 2
Gerrit-Owner: Alexey Serbin <as...@cloudera.com>
Gerrit-Reviewer: Alexey Serbin <as...@cloudera.com>
Gerrit-Reviewer: Andrew Wong <aw...@cloudera.com>
Gerrit-Reviewer: Kudu Jenkins
Gerrit-Reviewer: Todd Lipcon <to...@apache.org>
Gerrit-Comment-Date: Thu, 22 Feb 2018 20:56:22 +0000
Gerrit-HasComments: No
[kudu-CR] [consensus peers] micro-optimization on controller status
Posted by "Alexey Serbin (Code Review)" <ge...@cloudera.org>.
Alexey Serbin has posted comments on this change. ( http://gerrit.cloudera.org:8080/9387 )
Change subject: [consensus_peers] micro-optimization on controller status
......................................................................
Patch Set 2:
> Patch Set 2: Code-Review+1
>
> (1 comment)
Yes, they are. I was just curious what were tablet servers doing (they were using a lot of CPU but could not converge in raft_consensus_stress-itest)
--
To view, visit http://gerrit.cloudera.org:8080/9387
To unsubscribe, visit http://gerrit.cloudera.org:8080/settings
Gerrit-Project: kudu
Gerrit-Branch: master
Gerrit-MessageType: comment
Gerrit-Change-Id: I25ac21db191a21f3bdfb3378e32fbe366d98297c
Gerrit-Change-Number: 9387
Gerrit-PatchSet: 2
Gerrit-Owner: Alexey Serbin <as...@cloudera.com>
Gerrit-Reviewer: Alexey Serbin <as...@cloudera.com>
Gerrit-Reviewer: Andrew Wong <aw...@cloudera.com>
Gerrit-Reviewer: Kudu Jenkins
Gerrit-Reviewer: Todd Lipcon <to...@apache.org>
Gerrit-Comment-Date: Thu, 22 Feb 2018 20:55:35 +0000
Gerrit-HasComments: No