Re: Extended Prefetching using Asynchronous IO - proposal and patch - Mailing list pgsql-hackers

From John Lumby
Subject Re: Extended Prefetching using Asynchronous IO - proposal and patch
Date
Msg-id BAY175-W124C3149E15060B95EFA39A3D50@phx.gbl
Whole thread Raw
In response to Re: Extended Prefetching using Asynchronous IO - proposal and patch  (John Lumby <johnlumby@hotmail.com>)
Responses Re: Extended Prefetching using Asynchronous IO - proposal and patch
List pgsql-hackers
I am attaching a new version of the patch for consideration in the current commit fest.

Relative to the one I submitted on 25 June in BAY175-W412FF89303686022A9F16AA3190@phx.gbl
the method for handling aio completion using sigevent has been re-written to use
signals exclusively rather than a composite of signals and LWlocks,
and this has fixed the problem I mentioned before with the LWlock method.
More details are in postgresql-prefetching-asyncio.README
(search for   
     "Checking AIO Completion"
)

I also have worked my benchmark database and one application into a publishable state.
However,   the database is in the form of a compressed pg_dump and is around
218 MB in size.    I was hoping to come up with a generation program to load it
but have not had the time for that.     Is there some place on postgresql.org for
such a large file?   If not I will try to think of some place for it.

John Lumby 




Attachment

pgsql-hackers by date:

Previous
From: Andres Freund
Date:
Subject: Re: 9.3: more problems with "Could not open file "pg_multixact/members/xxxx"
Next
From: Alvaro Herrera
Date:
Subject: Re: [patch] pg_copy - a command for reliable WAL archiving