Click here for Iris Computer Planning Sponsor Message

Performance Generalizations During MPE Upgrades

By George Stachnik
Hewlett-Packard Company

Customers today are asking if upgrading to a 5.x release from 4.0 will require more RAM. I am asked this question every time HP brings out a new release of MPE/iX, and the answer is always the same -- "it depends."

MPE/iX is a little different from other operating systems with respect to performance. With most OS's, whenever you add to the workload of a system, performance gets a little bit worse. As you get close to the limit, each additional user begins to "hurt" more and more -- but if you were to graph response time against the number of users, you'd get a curve that slopes up as the system gets busier, becoming steeper and steeper as you add more users.

By contrast, MPE/iX's performance curve tends to be relatively flat. That is, whatever response time you get with a small number of users will be pretty much the same response time you get with more users -- at least up to a point. The curve tends to stay flat until you get close to the limit. Then it shoots up dramatically.

Think of it this way: On most OS's, moving up the performance curve is like driving up a hill that keeps getting steeper until, at the limit, it's too steep to drive on. On MPE/iX, moving up the performance curve is like driving on a flat road until you hit the limit (the dreaded "knee of the curve"). Then adding even one user can cause performance to shoot up like a skyrocket. It is literally like running into a brick wall.

Now combine this with the fact that upgrading your OS will make your system work harder. This is true of all operating systems, not just MPE/iX. In other words, when you upgrade from 4.0 to 5.0 or 5.5, it's like moving along the performance curve by a step or two. And what's that going to do to your performance? The answer is (all together now) "It depends." If you're on the "flat part" of the curve, (that is, you aren't anywhere near the limit), then the update will be a non-event. But if you're near the limit, then the update can be like running into a brick wall.

As an example, let's say you currently have 80 users on the system, and you believe it will handle 105. Assuming that you're correct, you should still be on the flat part of the curve, and have some room to manouver. Moving up to from 4.0 up to 5.x doesn't sound like it ought to be a problem. Before you do the update, you might want to ask yourself these questions:

One last point -- everything that I've written here is a generalization. And all generalizations are wrong (including this one). I'd like to use this article to open a debate among our resident gurus on HP 3000 performance. I haven't looked at 3000 performance in detail recently. Is everything that I've said about the performance curve still true? I know that there are exceptions, particularly on very large systems.

For example, I've seen environments with many hundreds of users where the need to do a lot of disk transactions begins to outstrip the transaction manager's ability to log them. Then you begin to see I/O bottlenecks, and the performance curve begins to look more like that associated with other OS's. In extreme cases, the issue ceases to be performance and begins to become avoiding system aborts -- but that, as they say, is a whole 'nother issue.

George Stachnik is the host of HP's Technology Close Up satellite broadcasts and video training for HP business and technical computer systems.


Copyright 1996, The 3000 NewsWire. All rights reserved.