Databases are usually IO bound , vmstat results can confirm individual cases and setups. In case the server is IO bound the entry point should be setting up properly performing IO. RAID10 helps a great extent in improving IO bandwidth by parallelizing the IO operations, more spindles the better. Also write caches helps in great deal in caching the writes and making commits faster.
In my opinion system level tools (like vmstat) at peak load times can be an entry point in understanding the bottlenecks of a particular setup. if there is swapping u absolutely need to double the ram . ( excess ram can be used in disk block caching) if its cpu bound add more cores or high speed cpus if its io bound put better raid arrays & controller. regds mallah. On Thu, Mar 12, 2009 at 4:22 PM, Nagalingam, Karthikeyan <karthikeyan.nagalin...@netapp.com> wrote: > Hi, > Can you guide me, Where is the entry point to get the documentation for > Postgresql performance tuning, Optimization for Postgresql with Storage > controller. > > Your recommendation and suggestion are welcome. > > Regards > Karthikeyan.N > > -- Sent via pgsql-performance mailing list (pgsql-performance@postgresql.org) To make changes to your subscription: http://www.postgresql.org/mailpref/pgsql-performance
