Re: [PERFORM] Sun Donated a Sun Fire T2000 to the PostgreSQL - Mailing list pgsql-hackers

From Tom Lane
Subject Re: [PERFORM] Sun Donated a Sun Fire T2000 to the PostgreSQL
Date
Msg-id 21889.1153489365@sss.pgh.pa.us
Whole thread Raw
In response to Re: [PERFORM] Sun Donated a Sun Fire T2000 to the PostgreSQL  (Robert Lor <Robert.Lor@Sun.COM>)
Responses Re: [PERFORM] Sun Donated a Sun Fire T2000 to the PostgreSQL  (Sven Geisler <sgeisler@aeccom.com>)
Re: [PERFORM] Sun Donated a Sun Fire T2000 to the PostgreSQL  (Robert Lor <Robert.Lor@Sun.COM>)
Re: [PERFORM] Sun Donated a Sun Fire T2000 to the PostgreSQL  (Robert Lor <Robert.Lor@Sun.COM>)
List pgsql-hackers
Robert Lor <Robert.Lor@Sun.COM> writes:
> I ran pgbench and fired up a DTrace script using the lwlock probes we've
> added, and it looks like BufMappingLock is the most contended lock, but
> CheckpointStartLocks are held for longer duration!

Those numbers look a bit suspicious --- I'd expect to see some of the
LWLocks being taken in both shared and exclusive modes, but you don't
show any such cases.  You sure your script is counting correctly?
Also, it'd be interesting to count time spent holding shared lock
separately from time spent holding exclusive.

            regards, tom lane

pgsql-hackers by date:

Previous
From: "moises"
Date:
Subject: Transaction Speed and real time database
Next
From: "Jim C. Nasby"
Date:
Subject: Re: Freezing tuples on pages dirtied by vacuum