A: I've seen backup servers maxed out by anywhere from five to 2,500 hosts. There's no single rule of thumb. Instead, look for one of the following things to happen:

1. The I/O requirements of the backup indices/catalogs get too great for your server to handle. If you have a Unix server, run iostat and top. On NT/2000, use the system performance monitor.
2. You're backing up the maximum amount your bandwidth will allow. Assuming a 12MB/s theoretical maximum (100Mb-/8=12MB/s), and an 8-hour backup window, the maximum a 100Mb connection could theoretically do is 345GB in one night (12MB x 60 seconds x 60 minutes x 8 hours = 345,600). If you're anywhere close to that - or even 70% to 80% of that - you've hit a wall.
3. The CPU or memory on your box is completely gone, and you haven't experienced No. 2. If you can't upgrade the CPU or memory, it's time for another box. (By the way, No. 1 can cause No. 3.) Also, different backup software packages have different backup requirements.

Additional Questions:
	Aug 2002:	Why doesn't my mail client software create a lock for e-mail files if it's reading it off a NAS device? Which non-EMC NAS devices provide a gateway to connect to an EMC Symmetrix SAN?
	July 2002:	What does storage pooling mean? What do you think of it?
	May 2002:	What is the difference between a NAS gateway and a SAN appliance? Is tape still the best for archiving media files? Does it offer a lower TCO than disk?
	Apr 2002:	In terms of feed speed, what does that measurement refer to? Is it from the primary storage to the secondary storage? Does it stop at the write speed on the tape drive?
	Mar 2002:	How can you use solid-state disk (SSD) to accelerate SAN performance?