Notwithstanding the point raised about file level fragmentation, shrinking databases can cause all manner of issues internally and may only cause extra growth, and more fragmentation when you next rebuild...

If you plan to use PAE there is an important o/s fix you should apply.

http://support.microsoft.com/default.aspx?scid=kb;en-us;838765

It's a real pain isn't it?  The only way around this is to stagger the index rebuilds to fit in with the logs so you generate a series of logs...

sorry, yup using 2k adv server, but thinking about 2003 ent server.

Just trying to clear in my head if a server with 8 zeon 4's ( with HT ) would...

shrink file can be tricky .. I often find it works best if you place the database into simple recovery model and issue a checkpoint. Try scripting the command rather...

I always blame the network

Most SQL Server books cover disk hardware. Try the administrators companion from ms press or inside sql 2k.  If you're going to administer a sql server you need a good...

You could use sql litespeed which produces compressed backups or use a zip program on the logs

I don't have this problem

I seem to remember that other than straight conversion I scripted out the objects and schema and applied that to a new database first. That picked up any issues as...

I have always updated stats as a scheduled job on production databases every night, using sp_updatestats,  as part of general housekeeping. I was prompted into my initial post as I'd...

Most modern boxes will come with hyperthreading so you'll almost get a 4 way upgrade.

I did a similar upgrade, for different reasons, and queries which took over an hour ran...

Interesting, many thanks.

I did have a search before posting the question .. think I've probably answered my own question by further research across a number of my servers and databases, I might...

SAN's are another subject all together. I did think most SANs connected by gb connections not by 100Mb fibre cards to the server. To be honest a single 100Mb connection...