Back to the main page

Mailing List Logs for ShadowRN

From: Mike Elkins <MikeE@*********.COM>
Subject: Server Prices (was Re: Hacking Security Tallies)
Date: Tue, 26 May 1998 15:33:15 -0500
>Depends. You can get a kick butt multi-cpu intel based system for what
>$2000 or so. I think the Alpha 500 Mhz is probably in the $5000 range
>(somebody check me on that), and IBM's SP-2's are considerably more
>expensive (Into the 100,000's) depending on how much memory they
>have and when you bought them.
>As far as mainframes go, I have no idea.

Well, you get what you pay for. $2000 gets you a PC. If you try to run
server software on a PC it sort of works, but if you push the load on it
up towards the maximum, it starts getting very flakey.

Essentially there are 3 things that make a server a server, and not all
servers have all three, but that it what to aim for:
1) Reliability. You can expect a server to run 24 hours a day, 7 days a
week, for months or years at a time. Expect to pay at least $10,000 for
a PC with this kind of reliability.
2) Capacity: The ability to add lots and lots of disk space, RAM, CPUs
etc. You can add these to a PC, but expect to hit limits.
3) Speed: The least important of these, in some ways. Similar to
Capacity.

I haven't priced a mainframe lately, but my last company's mainframe
cost $3.5 Mil, and it was a small one.

Here would be my rule of thumb: Base cost: 1-3 users: $8000
4-10 users $15000, 10-50 users $40,000, 50-100 users: $100,000

Excellant Reliability: price x 5
Rock-Solid Reliability: price x 15-25

Security: Green System: price x 1.5
Orange System: price x 6
Red System: price x 15
Black System: GM's call

Add some more money for higher system ratings, too.

All of this is off the top of my head, and IMHO.

Sanity Check: An Orange system for a small research group (25-50
people), mission critical work: $40k*20*6 = $4.8 million. Yup, I can live
with that.

Double-Domed Mike

Disclaimer

These messages were posted a long time ago on a mailing list far, far away. The copyright to their contents probably lies with the original authors of the individual messages, but since they were published in an electronic forum that anyone could subscribe to, and the logs were available to subscribers and most likely non-subscribers as well, it's felt that re-publishing them here is a kind of public service.