| From: | "Merlin Moncure" <merlin(dot)moncure(at)rcsonline(dot)com> |
|---|---|
| To: | "Alex Turner" <armtuk(at)gmail(dot)com> |
| Cc: | <pgsql-performance(at)postgresql(dot)org> |
| Subject: | Re: Hardware/OS recommendations for large databases ( |
| Date: | 2005-11-17 20:22:50 |
| Message-ID: | 6EE64EF3AB31D5448D0007DD34EEB3417DD907@Herge.rcsinc.local |
| Views: | Raw Message | Whole Thread | Download mbox | Resend email |
| Thread: | |
| Lists: | pgsql-performance |
> Remember - large DB is going to be IO bound. Memory will get thrashed
> for file block buffers, even if you have large amounts, it's all gonna
> be cycled in and out again.
'fraid I have to disagree here. I manage ERP systems for manufacturing
companies of various sizes. My systems are all completely cpu
bound...even though the larger database are well into two digit gigabyte
sizes, the data turnover while huge is relatively constrained and well
served by the O/S cache. OTOH, query latency is a *huge* factor and we
do everything possible to lower it. Even if the cpu is not 100% loaded,
faster processors make the application 'feel' faster to the client.
Merlin
| From | Date | Subject | |
|---|---|---|---|
| Next Message | David Boreham | 2005-11-17 20:29:59 | Re: Hardware/OS recommendations for large databases ( |
| Previous Message | Alex Turner | 2005-11-17 19:54:54 | Re: Hardware/OS recommendations for large databases ( |