Re: [PERFORM] Windows performance again

2005-12-21 Thread Merlin Moncure
> > It's *got* to be the network configuration on the client machine. > > We've seen gripes of this sort before --- check the list archives for > possible fixes. I seem to recall something about a "QoS patch", as > well as suggestions to get rid of third-party packages that might be > interfering

Re: [PERFORM] Windows performance again

2005-12-21 Thread Tom Lane
Richard Huxton writes: > Josep Maria Pinyol Fontseca wrote: >> Network utilization: 0% > It's *got* to be the network configuration on the client machine. We've seen gripes of this sort before --- check the list archives for possible fixes. I seem to recall something about a "QoS patch", as wel

Re: [PERFORM] Windows performance again

2005-12-21 Thread Greg Stark
Richard Huxton writes: > Josep Maria Pinyol Fontseca wrote: > > Windows XP client box with psql version 8.1.1 versus Windows XP server: > > select * from ; -> 60 seconds aprox. to obtain a results > > Network utilization: 3% The 60 seconds sounds suspiciously like a DNS problem. --

Re: [PERFORM] Windows performance again

2005-12-21 Thread Richard Huxton
Josep Maria Pinyol Fontseca wrote: Linux box with psql version 7.2.1 versus Windows XP server: select * from ; -> 7 seconds aprox. to obtain a results. Network utilization: 100% Windows XP client box with psql version 8.1.1 versus Windows XP server: select * from ; -> 60 seconds aprox. to obt

[PERFORM] Windows performance again

2005-12-21 Thread Josep Maria Pinyol Fontseca
Hello, We have the next scenario: Linux box with postgresql 7.2.1-1 (client) Windows XP with postgresql 8.1.1 (server) Windows XP with postgresql 8.1.1 (client) All connected in 10Mb LAN In server box, we have a table with 65000 rows and usign "psql" we have these results: Linux box with ps