Re: Warm-cache prefetching - Mailing list pgsql-hackers

From Simon Riggs
Subject Re: Warm-cache prefetching
Date
Msg-id 1134145211.2906.1271.camel@localhost.localdomain
Whole thread Raw
In response to Re: Warm-cache prefetching  (Bruce Momjian <pgman@candle.pha.pa.us>)
List pgsql-hackers
On Fri, 2005-12-09 at 09:43 -0500, Bruce Momjian wrote:
> Do these optimizations have any affect on database software?  I know
> call overhead shows up as a performance bottleneck, but do these others
> optimizations also measurably improve performance?

Many of them can, but nowhere near as much benefit as you'd get by
applying them to a graphics/media app.

We've already taken note of the effects of buffer alignment and padding,
which was pretty much straight out of this manual. 

Recent research suggests that the CPU<->memory bottleneck is growing
worse each year, so attention to this area should have longer term
payoffs. Recent CPU changes indicate number of CPU cores available is
going to increase dramatically in next few years, so scalability is
critical also.

The trick is to find the optimization sweet spot, and to be certain that
it is reasonably broadly applicable. But I think we will find some
benefit, somewhere, so I think its worth the look, at the same time
checking that the compiler guys haven't solved it for us first.

Best Regards, Simon Riggs




pgsql-hackers by date:

Previous
From: Martijn van Oosterhout
Date:
Subject: Re: Upcoming PG re-releases
Next
From: Bruce Momjian
Date:
Subject: Re: Backslashes in string literals