Skip site navigation (1) Skip section navigation (2)

Re: Why are we waiting?

From: Gregory Stark <stark(at)enterprisedb(dot)com>
To: "Staale Smedseng" <Staale(dot)Smedseng(at)Sun(dot)COM>
Cc: "Simon Riggs" <simon(at)2ndquadrant(dot)com>, <pgsql-hackers(at)postgresql(dot)org>
Subject: Re: Why are we waiting?
Date: 2008-02-06 14:56:16
Message-ID: 87ir12uoa7.fsf@oxford.xeocode.com (view raw or flat)
Thread:
Lists: pgsql-hackers
"Staale Smedseng" <Staale(dot)Smedseng(at)Sun(dot)COM> writes:

> The stack trace shows that the only time the lock is acquired
> exclusively is from the call to ProcArrayEndTransaction() in
> CommitTransaction().

I'm not sure but I think that's only true in 8.3. As I understood it in 8.2
transaction start also needed the exclusive lock.

> Also, an interesting observation is that the hot locks seem to have
> changed from v8.2 to v8.3, making the ProcArrayLock more contended. See
> the following outputs:
>
> PostgreSQL 8.2 (32-bit):
>...
> PostgreSQL 8.3 (64-bit):
>...

I'm not sure 32-bit and 64-bit cases are going to be directly comparable. We
could have a problem with cache line aliasing on only one or the other for
example.

But that is a pretty striking difference. Does the 8.3 run complete more
transactions in that time?

-- 
  Gregory Stark
  EnterpriseDB          http://www.enterprisedb.com
  Ask me about EnterpriseDB's 24x7 Postgres support!

In response to

Responses

pgsql-hackers by date

Next:From: Simon RiggsDate: 2008-02-06 15:13:24
Subject: pg_dump additional options for performance
Previous:From: Dave PageDate: 2008-02-06 14:49:24
Subject: Re: PostgreSQL 8.4 development plan

Privacy Policy | About PostgreSQL
Copyright © 1996-2014 The PostgreSQL Global Development Group