Horton

Re: measuring lwlock-related latency spikes

Albe Laurenz <laurenz.albe@wien.gv.at>

View thread
From: "Albe Laurenz" <laurenz.albe@wien.gv.at>
To: "Robert Haas *EXTERN*" <robertmhaas@gmail.com>, "Greg Stark" <stark@mit.edu>
Cc: <pgsql-hackers@postgresql.org>
Date: 2012-04-02T07:11:00Z
Lists: pgsql-hackers

Commits

Same data as JSON: GET /api/v1/messages/:b64id/commits the thread's linked commits as JSON, with link sources. API reference →

  1. Add new files to NLS file lists

Robert Haas wrote:
> I suppose one interesting question is to figure out if there's a way I
> can optimize the disk configuration in this machine, or the Linux I/O
> scheduler, or something, so as to reduce the amount of time it spends
> waiting for the disk.

I'd be curious to know if using the deadline scheduler will improve
things.  I have experienced pretty bad performance with cfq under
load, where sequential table scans were starved to the point where
they took hours instead of less than a minute (on an idle system).
But I believe that also depends a lot on the storage system used.

Yours,
Laurenz Albe