Hi,<br /><br /><br />A quick overview of our database server:<br />* Four databases.<br />* Each database has about 20
schemas.<br/><br /><br />The largest database contains:<br />* select count(*) from pg_class where relkind = 'v'<br />
101<br/>* select count(*) from pg_class where relkind = 'r'<br /> 11911 (about 500 tables in each schema, I know,
it'sa lot - but I'd bet it's not uncommon)<br />* About 10 sequences.<br />* About 150 functions.<br /><br /><br
/>selectcount(*) from pg_class<br />> 36444<br /><br /><br />All these tables are large ones and have some toasted
rows(you can see it in pg_type).<br /><br /><br />Those queries are rather fast, it's just that operating over a
(relatively)slow network exposes us to latencies of shipping that much traffic.<br />I was wondering if the need to
accessthat table can be delayed, so queries would join against it instead of prefetching it? Or perhaps cache it
locallyon disk and fetch only higher OID values (I'm guessing here, possibly incorrectly, that rows are not updated but
onlyadded) this would enable one full fetch and incremental updates since.<br /><br /><br />Thanks,<br /> Sefer.<br
/><br/><br /><br /><hr />Hotmail: Powerful Free email with security by Microsoft. <a
href="http://clk.atdmt.com/GBL/go/196390710/direct/01/"target="_new">Get it now.</a>