| From: | Claudio Freire <klaussfreire(at)gmail(dot)com> |
|---|---|
| To: | Robert Haas <robertmhaas(at)gmail(dot)com> |
| Cc: | Tomas Vondra <tv(at)fuzzy(dot)cz>, "pgsql-performance(at)postgresql(dot)org" <pgsql-performance(at)postgresql(dot)org> |
| Subject: | Re: Performance |
| Date: | 2011-04-26 07:49:39 |
| Message-ID: | BANLkTikfAzvviWVmp-e81Ei2gNV4aaECQg@mail.gmail.com |
| Views: | Raw Message | Whole Thread | Download mbox | Resend email |
| Thread: | |
| Lists: | pgsql-performance |
On Tue, Apr 26, 2011 at 7:30 AM, Robert Haas <robertmhaas(at)gmail(dot)com> wrote:
> On Apr 14, 2011, at 2:49 AM, Claudio Freire <klaussfreire(at)gmail(dot)com> wrote:
>> This particular factor is not about an abstract and opaque "Workload"
>> the server can't know about. It's about cache hit rate, and the server
>> can indeed measure that.
>
> The server can and does measure hit rates for the PG buffer pool, but to my knowledge there is no clear-cut way for PG to know whether read() is satisfied from the OS cache or a drive cache or the platter.
Isn't latency an indicator?
If you plot latencies, you should see three markedly obvious clusters:
OS cache (microseconds), Drive cache (slightly slower), platter
(tail).
I think I had seen a study of sorts somewhere[0]...
Ok, that link is about sequential/random access, but I distinctively
remember one about caches and CAV...
| From | Date | Subject | |
|---|---|---|---|
| Next Message | Sok Ann Yap | 2011-04-26 09:49:05 | reducing random_page_cost from 4 to 2 to force index scan |
| Previous Message | Mark Kirkwood | 2011-04-26 06:30:51 | Re: Time to put theory to the test? |