On Fri, Nov 2, 2012 at 09:59:08AM -0400, John Lumby wrote:
> Thanks for the mentioning this posting. Interesting.
> However, the OP describes an implementation based on libaio.
> Today what we have (for linux) is librt, which is quite different.
> It is arguable worse than libaio (well actually I am sure it is worse)
> since it is essentially just an encapsulation of using threads to do
> synchronous ios - you can look at it as making it easier to do what the
> application could do itself if it set up its own pthreads. The linux
> kernel does not know about it and so the CPU overhead of checking for
> completion is higher.
Well, good thing we didn't switch to using libaio, now that it is gone.
Bruce Momjian <bruce(at)momjian(dot)us> http://momjian.us
+ It's impossible for everything to be true. +
In response to
pgsql-hackers by date
|Next:||From: Tom Lane||Date: 2012-11-05 20:40:12|
|Subject: Re: Doc patch, distinguish sections with an empty row in error code table|
|Previous:||From: Tom Lane||Date: 2012-11-05 20:30:32|
|Subject: Re: Pg_upgrade speed for many tables|