- From: Mike Meyer <mwm@contessa.phone.net>
- Date: Fri, 5 Dec 1997 15:23:07 PST
- To: www-talk@w3.org
- Cc: mxenakis@pscwa.psca.com
> Has anyone out there done any work w/ this, or have ideas on how to = > handle the problem? A more specific description being as such... I've just starting facing it. The only small-systems solution I've been able to come up with is a proxy of some sort. If you've got enough hardware to dedicate one to the job, that probably works fine for a while. But if you're trying to unload a single processor, it doesn't help much. The large-systems solution is to go to a scalable architecture (I believe that means RISC), so you can add more processors/busses/etc at the same IP address. I've run into people who tell me that many small systems are a better solution than one multi-CPU RISC box; possibly someone who believes that can provide an answer. Thanx, <mike
Received on Friday, 5 December 1997 18:33:09 UTC