Re: ratios -- machines to personnel

[Andy Poling <andy@realbig.com>]

On Tue, 16 Jul 1996, John F. Blasko wrote:
> In the 'old days' there used to be information available that tried
> to identify the number of systems administrators required to 
> support a certain number of machines.  The varibles included
> the amount of machine storage, the application suite, and other
> descriptions of a machine primary function.  Has there been
> any recent work done in this area?  Where might I look for
> this type of information.

It's kind of problematic to produce such figures these days.  The
distinction between small and large machines has blurred tremendously, and
you can easily have a small machine with more storage and more users than a
large machine.  Or you can have a large machine with many users using a wide
variety of applications.  How do you equate these systems?

Further complicating things is the proliferation of semi-distributed
information services (an obvious example being WWW servers).  These systems
often don't have an associated user body as their primary function is to
disseminate information as widely as possible.

Mostly though, I think the wide variety of uses to which we put our
computers has made this task very difficult.  How many personal workstations
have only a word processor and a spreadsheet on them anymore?  For that
matter, how many shared systems have only a few applications on them
anymore?  The systems have become diversified to the point that they can
even sometimes defy classification as to function.

At the level of workstations, I think the ratio of admin to user might be
useful.

For what it's worth...

-Andy

(this sig left intentionally blank)