Re: User concurrency thresholding: where do I look?

From: "Jignesh K(dot) Shah" <J(dot)K(dot)Shah(at)Sun(dot)COM>
To: "Jignesh K(dot) Shah" <J(dot)K(dot)Shah(at)Sun(dot)COM>
Cc: Simon Riggs <simon(at)2ndquadrant(dot)com>, Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us>, Josh Berkus <josh(at)agliodbs(dot)com>, pgsql-performance(at)postgresql(dot)org, Alvaro Herrera <alvherre(at)commandprompt(dot)com>
Subject: Re: User concurrency thresholding: where do I look?
Date: 2007-07-27 20:04:57
Message-ID: 46AA4FE9.2030302@sun.com
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-performance

Using CLOG Buffers 32 and the commit sibling check patch I still see a
drop at 1200-1300 users..

bash-3.00# ./4_lwlock_waits.d 18250

Lock Id Mode Count
XidGenLock Shared 1
CLogControlLock Shared 2
ProcArrayLock Shared 2
XidGenLock Exclusive 4
CLogControlLock Exclusive 15
WALInsertLock Exclusive 18
WALWriteLock Exclusive 38
ProcArrayLock Exclusive 77

Lock Id Mode Combined Time (ns)
XidGenLock Shared 88700
WALInsertLock Exclusive 69556000
ProcArrayLock Shared 95656800
XidGenLock Exclusive 139634100
CLogControlLock Exclusive 148822200
CLogControlLock Shared 161630000
WALWriteLock Exclusive 332781800
ProcArrayLock Exclusive 5688265500

bash-3.00# ./4_lwlock_waits.d 18599

Lock Id Mode Count
ProcArrayLock Shared 2
XidGenLock Exclusive 3
XidGenLock Shared 4
CLogControlLock Shared 5
WALInsertLock Exclusive 10
CLogControlLock Exclusive 21
WALWriteLock Exclusive 28
ProcArrayLock Exclusive 54

Lock Id Mode Combined Time (ns)
XidGenLock Exclusive 5688800
WALInsertLock Exclusive 11424700
CLogControlLock Shared 55589100
ProcArrayLock Shared 135220400
WALWriteLock Exclusive 177906900
XidGenLock Shared 524146500
CLogControlLock Exclusive 524563900
ProcArrayLock Exclusive 5828744500

bash-3.00#
bash-3.00# ./6_lwlock_stack.d 4 18599

Lock Id Mode Count
ProcArrayLock Shared 1
ProcArrayLock Exclusive 52

Lock Id Mode Combined Time (ns)
ProcArrayLock Shared 41428300
ProcArrayLock Exclusive 3858386500

Lock Id Combined Time (ns)

postgres`LWLockAcquire+0x1f0
postgres`GetSnapshotData+0x120
postgres`GetTransactionSnapshot+0x80
postgres`PortalStart+0x198
postgres`exec_bind_message+0x84c
postgres`PostgresMain+0x17f8
postgres`BackendRun+0x2f8
postgres`ServerLoop+0x680
postgres`PostmasterMain+0xda8
postgres`main+0x3d0
postgres`_start+0x17c
Shared 41428300

postgres`LWLockAcquire+0x1f0
postgres`CommitTransaction+0x104
postgres`CommitTransactionCommand+0xbc
postgres`finish_xact_command+0x78
postgres`exec_execute_message+0x42c
postgres`PostgresMain+0x1838
postgres`BackendRun+0x2f8
postgres`ServerLoop+0x680
postgres`PostmasterMain+0xda8
postgres`main+0x3d0
postgres`_start+0x17c
Exclusive 3858386500

-Jignesh

In response to

Responses

Browse pgsql-performance by date

  From Date Subject
Next Message Steven Flatt 2007-07-27 21:32:11 Vacuum looping?
Previous Message Lew 2007-07-27 20:03:11 Re: multicolumn index column order