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

MAINFRAME > Administrator > Backup and Recovery

Disaster Planning Pays Big Dividends

A D/R plan is essential in today's environment.

A D/R plan is essential in today's environment.

Most companies running mainframes can't afford to not have a disaster recovery (D/R) plan. Most financial institutions will go out of business if they are off the air for a few hours or, in some cases, minutes. Some organisations can afford to be down for a longer time, such as distribution companies, retailers, and the like.

But, nobody can afford to be down indefinitely. So, everybody needs one, but not everybody has one! D/R plans are akin to insurance policies: how much do you invest in time and money? That is, if you do decide to invest.

The biggest problems are:

  1. How often do you test it?
  2. What do you test?
  3. How do you test it?

Since there are very few stand-alone mainframe applications these days, with alternate platforms involved, I shall touch (briefly) on some of the issues involved outside of the mainframe. But, I will primarily focus on the mainframe, and z/OS.

How often do you test it?

Most shops test - if at all - once a year. And, it's a major project with many aspects involved. You need commitment from business users, applications, operations, technical support, call centres, and (if you use one) your out-sourcer. Of course, if you need an off-site location, the D/R provider also has to be involved.

Coordination of all the people involved also requires the discipline of project management. D/R is such a large scale test these days, most companies can't afford to do it more than once a year - some even less frequently.

Once you add in the alternate platforms, you increase the complexity and the management of the test by an order of magnitude, at minimum.

The alternative to testing is a much bigger problem. People used to say: "It'll never happen to me!" But, after Hurricane Katrina, I think there are fewer of those around.

What do you test?

Even the best laid plans have hitches. One company that comes to mind moved their people to an alternate location, well outside of the boundaries of Hurricane Katrina, but they couldn't communicate with the support staff. Everybody had a cell-phone, and the numbers were duly recorded. But, all cell calls go through the home central office of the area code for the phone, first. Unfortunately, those offices were (of course) down. How do you even think of something like that? I never would have given it a second thought, until Katrina.

Remember D/R is more than just testing the restore capabilities of the environment, and the fact that all systems IPL. You must test the critical business processes, including those that use little or no processing resources.

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



2019 Solutions Edition

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


Active/Active Sites Helps Distribute Workloads During Outages

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