Re: Adding more memory = hugh cpu load

From: "Kevin Grittner" <Kevin(dot)Grittner(at)wicourts(dot)gov>
To: "alexandre - aldeia digital" <adaldeia(at)gmail(dot)com>
Cc: <pgsql-performance(at)postgresql(dot)org>
Subject: Re: Adding more memory = hugh cpu load
Date: 2011-10-10 19:39:28
Message-ID: 4E9303A00200002500041CFE@gw.wicourts.gov
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-performance

alexandre - aldeia digital <adaldeia(at)gmail(dot)com> wrote:

> From the point of view of the client, the question is simple:
> until the last friday (with 16 GB of RAM), the load average of
> server rarely surpasses 4. Nothing change in normal database use.

Really? The application still performs as well or better, and it's
the load average they care about? How odd.

If they were happy with performance before the RAM was added, why
did they add it? If they weren't happy with performance, what led
them to believe that adding more RAM would help? If there's a
performance problem, there's generally one bottleneck which is the
limit, with one set of symptoms. When you remove that bottleneck
and things get faster, you may well have a new bottleneck with
different symptoms. (These symptoms might include high load average
or CPU usage, for example.) You then figure out what is causing
*that* bottleneck, and you can make things yet faster.

In this whole thread you have yet to give enough information to know
for sure whether there was or is any performance problem, or what
the actual bottleneck is. I think you'll find that people happy to
help identify the problem and suggest solutions if you provide that
information.

-Kevin

In response to

Responses

Browse pgsql-performance by date

  From Date Subject
Next Message Shaun Thomas 2011-10-10 20:25:54 Re: Adding more memory = hugh cpu load
Previous Message alexandre - aldeia digital 2011-10-10 18:54:03 Re: Adding more memory = hugh cpu load