Article View: bionet.general
Article #29101Re: Directions for scientific computing centers for the future
From: Colston Chandler PHYSICS
Date: Thu, 13 Dec 1990 09:02
Date: Thu, 13 Dec 1990 09:02
47 lines
3304 bytes
3304 bytes
In article <1990Dec13.011756.22888@phri.nyu.edu> roy@alanine.phri.nyu.edu (Roy Smith) writes: > > We're thinking about future directions of scientific computing here >at the PHRI... > >Is the whole concept of a "central computing facility" outdated? If such a >facility should exist, what form should it take? If not, should there still >be some sort of central service department that supports individual personal >machines, and provides a network to connect them to each other and the >outside world? Provide shared peripherals like printers? Centralized >backups? File servers? User consulting? > Well, I am always willing to give my unqualified opinion! My background is probably different from a public health group, so take this with that in mind (my background is in large scale scientific computing - - i.e., thousands of hours of Cray time). However, the thing that strikes me is that with the power of workstations and the low cost, it makes little sense to have central computing facilities unless there is some unusual justifying factor (need for enormous memory computers for example). At Sandia National Laboratory for example, where I have worked for years on and off as a consultant, they are now moving away from VAX's to workstations (mostly Sun's). Each workstation is more powerful than their VAX's and alot cheaper. Also, it is hard to beat the productivity one can achieve on a workstation, UNIX notwithstanding. However, what appears to be happening there is that they still need a group of support people for each department to keep the systems up. UNIX is a real pain if you aren't a guru and few users are. It sure helps to have a few around to help. Also, these systems fail occasionally and when they do it is nice to have someone knowledgeable. Finally, as to central printers, etc..., I think it probably depends on the circumstances. One can buy laser printers real cheap nowadays, so you probably can hang them on every workstation (or at enough to handle your needs on a network). On the other hand, if you need high speed laser printers or other specialized equipment, then economics dictates having them centralized (at the level of a group of 15 - 20 people). But you can drive these with workstations or servers just fine. In general however, it is sure hard to beat having things at your fingertips (this from someone who historically had to' use mainframes and other centralized resources). John Prentice john@unmfys.unm.edu
Technical Details
Message-ID:
<1990Dec13.090245.16186@ariel.unm.edu>Path:
arc1-ca.pugleaf.net!gmdzi!unido!mcsun!uunet!bionet!news.cs.indiana.edu!ariel.unm.edu!hydra.unm.edu!prenticeReferences:
<1990Dec13.011756.22888@phri.nyu.edu>Show Full Headers
Complete Article Headers
X-Google-Language: ENGLISH,ASCII-7-bit X-Google-Thread: fbe8c,b39b445043dc866b X-Google-Attributes: gidfbe8c,public X-Google-Thread: 1119a6,b39b445043dc866b X-Google-Attributes: gid1119a6,public X-Google-ArrivalTime: 1990-12-13 12:55:27 PST Path: arc1-ca.pugleaf.net!gmdzi!unido!mcsun!uunet!bionet!news.cs.indiana.edu!ariel.unm.edu!hydra.unm.edu!prentice From: prentice@hydra.unm.edu (Colston Chandler PHYSICS) Newsgroups: bionet.general,sci.research Subject: Re: Directions for scientific computing centers for the future Message-ID: <1990Dec13.090245.16186@ariel.unm.edu> Date: 13 Dec 90 09:02:45 GMT References: <1990Dec13.011756.22888@phri.nyu.edu> Sender: John K. Prentice Organization: University of New Mexico, Albuquerque Lines: 47 Posted: Thu Dec 13 10:02:45 1990