Re: Speed up Clog Access by increasing CLOG buffers
Robert Haas <robertmhaas@gmail.com>
On Thu, Mar 9, 2017 at 5:49 PM, Robert Haas <robertmhaas@gmail.com> wrote: > However, I just realized that in > both this case and in the case of group XID clearing, we weren't > advertising a wait event for the PGSemaphoreLock calls that are part > of the group locking machinery. I think we should fix that, because a > quick test shows that can happen fairly often -- not, I think, as > often as we would have seen LWLock waits without these patches, but > often enough that you'll want to know. Patch attached. I've pushed the portion of this that relates to ProcArrayLock. (I know this hasn't been discussed much, but there doesn't really seem to be any reason for anybody to object, and looking at just the LWLock/ProcArrayLock wait events gives a highly misleading answer.) -- Robert Haas EnterpriseDB: http://www.enterprisedb.com The Enterprise PostgreSQL Company
Commits
-
Use group updates when setting transaction status in clog.
- baaf272ac908 11.0 landed
- ccce90b39867 10.0 landed
-
Improve 64bit atomics support.
- e8fdbd58fe56 10.0 landed
-
Add ProcArrayGroupUpdate wait event.
- d4116a771925 10.0 landed
-
Make the different Unix-y semaphore implementations ABI-compatible.
- be7b2848c6d8 10.0 cited
-
Fix broken ALTER INDEX documentation
- 2143f5e12790 9.6.0 cited
-
Code and docs review for commit 3187d6de0e5a9e805b27c48437897e8c39071d45.
- d12e5bb79bb5 9.6.0 cited
-
Partition the freelist for shared dynahash tables.
- 44ca4022f3f9 9.6.0 cited
-
Correct StartupSUBTRANS for page wraparound
- 481725c0ba73 9.6.0 cited
-
Make idle backends exit if the postmaster dies.
- ac1d7945f866 9.6.0 cited
-
contrib/sslinfo: add ssl_extension_info SRF
- 49124613f134 9.6.0 cited
-
Reduce ProcArrayLock contention by removing backends in batches.
- 0e141c0fbb21 9.6.0 cited
-
Fix `make installcheck` for serializable transactions.
- 253de7e1eb9a 9.6.0 cited
-
Lockless StrategyGetBuffer clock sweep hot path.
- d72731a70450 9.5.0 cited
-
Reduce sinval synchronization overhead.
- b4fbe392f8ff 9.2.0 cited