SUN vs SGI

Chad Price price at cse.unl.edu
Tue Apr 27 08:45:33 EST 1993


In article <1993Apr27.001246.17373 at kakwa.ucs.ualberta.ca>, cup at bones.biochem.ualberta.ca writes:

|> But other considerations....?  Cost for comparable machines (I'm looking
|> into it), ease of administration (I'm no progammer). Probably better
|> graphics packages for displaying protein structures on the SGI, this could
|> be important.

We did a short comparison of Sun and SGI for the purpose of buying one to run
GCG (we is a group at the Univ of NE Medical Center). The results were that Sun
provides more bang for the buck. We decided on a Sparc 10 Mod 41 for several
reasons, including price; but more importantly expandability for price. The
equivalent price SGI is the Indigo. The Indigo maxes out at 7GB disk space and
96MB RAM; and if that sounds like a lot, look at the requirements for GCG. As I
interpreted their information, they want you to have 64MB RAM to support 5
concurrent users comfortably. In addition, you will need 50-150MB swap space. Add
that to 150MB for the programs and 400+MB for the data, and you already need a
lot of disk space. They recommend a minimum configuration of 1.4GB. With the
Human Genome project (not to mention everything else that is going on), we expect
the size of the database to increase radically over the next few years (2-5),
perhaps even doubling. If anything else is to run on the machine, you need LOTS
of RAM and disk space.

I'm the "computer nerd" who will be administering the machine (NOT a biochemist),
and I don't much care whether the machine is a Sun or an SGI from that point of
view - both are either currently, or soon to be Sys V based, and so it should all
be fairly similar - both will have their good and bad points.

Note that Sun is generally the development platform for many commercial packages,
and packages are often out and supported on Suns before anything else.

This is not to say that SGI is a bad machine - it is generally a bit faster than
Suns and tends to have better graphics. Certainly my protein crystallographer
friends love their SGIs!

chad
price at cse.unl.edu
chad at windsurf.unmc.edu




More information about the Bio-soft mailing list