You are currently on IBM Systems Media’s archival website. Click here to view our new website.

MAINFRAME > Administrator > Systems Management

The Case for Capacity Planning


Aside from my own agenda, having been a capacity planner for more than 20 years, there are valid reasons for planning and managing the resources on the raised floor (the mainframe and associated peripherals). The obvious one, in the mainframe arena, is software cost. I'll come to that in due course, once I address two other factors: performance/throughput and hardware cost.

Performance/Throughput

If you don't have any metrics regarding performance of your mainframe-based (or any platform-based) applications, how do you know if things are doing well? And, if you don't know what your performance profile is, how do you know if you have a tuning opportunity, an upgrade requirement or a configuration-change requirement of any type? You can't manage what you can't measure.

Any installation that's supporting a business has a true requirement to manage performance. Here are two examples where performance (or capacity or configuration) management is vital.

      1.  Consumers typically have more than one credit card. If a transaction takes too long or times out, the consumer might offer another card. The organization from the first card loses that business, but performance management can circumvent that loss by developing a solution. Perhaps an agreement with its subscriber institutions allows authorization to proceed automatically after a predetermined amount of time has elapsed in the approval process. Any crosschecking can take place behind the scenes between the provider and the financial institution to verify the transaction without slowing the purchase or driving the customer to a different provider.

      2.  Some enterprises with proprietary networks are linked through a common organization that allows them to communicate electronic financial interactions. If a transaction takes longer than a predetermined amount of time, the subscriber institution is charged a nominal penalty. While small, these fees add up quickly as transactions accumulate, resulting in potentially exponential costs. Capacity planning could help provide a means for avoiding these fees entirely.

Hardware Cost

I started working as a capacity analyst when a 3033U cost $5 million and had a capacity of five MIPS. My personal BlackBerry has more power than that and it didn't even cost $500 CDN

Yes, hardware is cheap, but even pennies can add up. Also, from a non-U.S. perspective, the exchange rate is a factor. Every mainframe I've been involved with was shipped from Poughkeepsie, N.Y., to Toronto. While the current 15 percent isn't bad, we had a long span at 65 percent. Add taxes, duties and one of the largest bottlenecks in the plan-to-implementation process, capital expenditure procedures. Those costs in time, money and frustration demand a capacity-planning process. This applies to processors, memory, DASD, tape, network and all other peripherals. As inexpensive as they've become, they're still not free.

Ted MacNeil is a capacity/performance analyst with more than 25 years in the IBM mainframe environment. Ted can be reached at tedmacneil@alumni.uwaterloo.ca.



Advertisement

Advertisement

2019 Solutions Edition

A Comprehensive Online Buyer's Guide to Solutions, Services and Education.

Active Energy Manager Pulls Resources Together

Integration with InfraStruXure Centeral provides single management interface.

Amplifying Your Energy Efficiency

IBM BladeCenter and virtualization team up for greener data centers

IBM Systems Magazine Subscribe Box Read Now Link Subscribe Now Link iPad App Google Play Store
Mainframe News Sign Up Today! Past News Letters