|
http://www.postgresql.org/docs/current/static/wal-async-commit.html " the server waits for the transaction's WAL records to be flushed to permanent storage before returning a success indication to the client." I think with fynch=off, whether WAL gets written to disk or not is still controlled by synchronous_commit parameter. guessing here... > Date: Wed, 16 Jun 2010 12:19:20 -0700 > Subject: Re: PostgreSQL as a local in-memory cache > From: jgardner@xxxxxxxxxxxxxxxxxxx > To: josh@xxxxxxxxxxxx > CC: pgsql-performance@xxxxxxxxxxxxxx > > On Wed, Jun 16, 2010 at 12:00 PM, Josh Berkus <josh@xxxxxxxxxxxx> wrote: > > > >> * fsync=off => 5,100 > >> * fsync=off and synchronous_commit=off => 5,500 > > > > Now, this *is* interesting ... why should synch_commit make a difference > > if fsync is off? > > > > Anyone have any ideas? > > > > I may have stumbled upon this by my ignorance, but I thought I read > that synchronous_commit controlled whether it tries to line up commits > or has a more free-for-all that may cause some intermediate weirdness. > > -- > Jonathan Gardner > jgardner@xxxxxxxxxxxxxxxxxxx > > -- > Sent via pgsql-performance mailing list (pgsql-performance@xxxxxxxxxxxxxx) > To make changes to your subscription: > http://www.postgresql.org/mailpref/pgsql-performance The New Busy is not the old busy. Search, chat and e-mail from your inbox. Get started. |
