Re: SSI patch version 12

From: "Kevin Grittner" <Kevin(dot)Grittner(at)wicourts(dot)gov>
To: "Heikki Linnakangas" <heikki(dot)linnakangas(at)enterprisedb(dot)com>
Cc: "Jeff Davis" <pgsql(at)j-davis(dot)com>,<pgsql-hackers(at)postgresql(dot)org>
Subject: Re: SSI patch version 12
Date: 2011-01-18 20:38:44
Message-ID: 4D35A5F40200002500039762@gw.wicourts.gov
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-hackers

"Kevin Grittner" <Kevin(dot)Grittner(at)wicourts(dot)gov> wrote:

> If my back-of-the-envelope math is right, a carefully constructed
> pessimal load could need up to (max_connections / 2)^2 -- so 100
> connections could conceivably require 2500 structures, although
> such a scenario would be hard to create. Current "picked from
> thin air" numbers would have space for 500.

Er, actually, we would have space for 5000, because it's five times
the number of SERIALIZABLEXACT structures which is ten times
max_connections. I guess that would explain why I've never seen a
report of a problem.

Still, someone who creates a very large number of connections and
pounds them heavily with SERIALIZABLE transactions might conceivably
run into it. Since that's something the docs explicitly warn you
*not* to do with serializable transactions, I'm not sure we need to
do more than make sure the error message and hint are good.
Thoughts?

-Kevin

In response to

Browse pgsql-hackers by date

  From Date Subject
Next Message Bruce Momjian 2011-01-18 20:55:12 test_fsync label adjustments
Previous Message Robert Haas 2011-01-18 20:35:01 Re: SSI patch version 12