Wow. That's a pretty wide open question. Here are some thoughts:
1a. Make sure you know what your backup and recovery plan is; performance isn't worth too much if a single point of failure or a user error causes an extended period of downtime.
1b. Make sure you TEST your recovery plan. If you don't, it's not a plan (it's something between a hope and a dream).
2. Make sure your performance and monitoring effort is in keeping with the priorities of the business; you can spend a LOT of time and money in this area, but if it isn't the business' priority, then you're wasting it.
3. Read, read, read.
MS SQL Server 2000 DBA Survival Guide (SAMS, 2000)
Microsoft SQL Server 2000 Operations Guide (Microsoft, 2001)
Inside SQL Server 2000 (Microsoft Press, 2000)
Books On Line
Have you hugged your backup today?