> 18% in s_lock is definitely bad :-(. Were you able to determine which
> LWLock(s) are accounting for the contention?
Gavin Sherry and Tom Daly (Sun) are currently working on identifying the
problem lock using DLWLOCK_STATS. Any luck, Gavin?
PostgreSQL @ Sun
In response to
pgsql-performance by date
|Next:||From: Jim Nasby||Date: 2006-06-17 19:25:14|
|Subject: Re: Precomputed constants?|
|Previous:||From: Tom Lane||Date: 2006-06-17 17:43:06|
|Subject: Re: Test request for Stats collector performance improvement |
pgsql-hackers by date
|Next:||From: Tom Lane||Date: 2006-06-17 19:21:09|
|Subject: Re: MultiXacts & WAL |
|Previous:||From: Heikki Linnakangas||Date: 2006-06-17 19:09:01|
|Subject: Re: MultiXacts & WAL|