Article

HP makes the case for mainframe migration

Matt Stansberry
John Pickett, Hewlett-Packard's worldwide mainframe alternative program manager, discusses who should "augment" their mainframes, who should migrate applications to another platform, and who should replace it outright.

What is the main reason to leave the mainframe?

Pickett: Primarily there are three reasons customers are approaching HP to shift a mainframe workload: hardware and software costs; the need to respond quickly to business changes; and disappearing skill sets. People are retiring or going away.

Requires Free Membership to View

More on mainframe migration:
Korean insurer retires 7,000 MIPS mainframes

Fast guide to mainframe migration

Who is the ideal HP candidate for mainframe re-hosting?

Pickett: If you look from just a hardware perspective, we're probably looking at someone with a workload of less than 1,500 MIPS. We've replaced more than that, but 1,500 MIPS covers 80% of the mainframes out there. The thing we look for is customers in a dynamic industry, looking to save and control costs. One of the biggest line items in the data center is the mainframe. Customers ask: Do we need it? Is it providing mission critical apps? Can we have it for less on an alternative platform? That, I'd say, is driving a lot of business to HP.

What are the HP options for mainframe replacement?

Pickett: There are three pillars to our overall strategy. Augment or surround -- if you have a mainframe and want to hang onto it, we do storage attach and software support. The next approach is modernizing the infrastructure -- taking a mainframe application to a Superdome or NonStop system.

The third area is modernizing applications. We assess the environment and determine if your COBOL applications should be re-hosted on another platform. Should you buy new off-the-shelf applications? Re-write the code? Retain the code? Part of our strategy is to take into account that some apps are running on the mainframe efficiently -- we aren't going to re-host if it doesn't make sense.

Do you see other vendors like Sun competing for the mainframe re-hosting deals?

Pickett: Generally not. I haven't seen Sun in a mainframe replacement deal in a while. As a matter of fact, the [mainframe re-hosting] software Unkix was picked up by Clerity Solutions Inc. back in June [2006]. The Unikix code has been ported over to HP-UX and Linux.

Most people I've talked to are pretty attached to their mainframes. How do you deal with that aspect?

Pickett: To be truthful, if I'm in there position I'd be the same way. We're not here to dis[respect] the mainframe. It's the birthplace of a lot of technologies. But the thing I find, people that have been working with the mainframe aren't aware that these technologies are available on open systems. Because they are diehards, they know the mainframe inside and out, but they aren't paying attention to the outside world.

Are former mainframe customers turning to HP-UX or Windows?

Pickett: I think it goes back to the application decision 70% of the time. The Windows platform obviously has a lot of breadth. If customers are predisposed to .Net or J2EE, that will determine where the customer is going. The Samsung Life Insurance re-hosting project is running HP-UX. [Mainframe migration customer] the São Paulo Stock Exchange, BOVESPA is running Windows.

That's the other thing that HP brings to the table -- choice. Integrity systems can run HP-UX, Windows, OpenVMS, Linux at the same time, all in different partitions. Montecito has helped both on performance and availability features and the ISV excitement we see is over 10,000 applications sitting on top of Integrity.

Where are the savings on switching from the mainframe to HP?

Pickett: Moving licensing costs from a mainframe using COBOL on top of DB2, it's significantly cheaper on an HP-UX or Windows platform running on a Windows server. Up front hardware cost, maintenance cost and software are also big differentiators.

Also, the Superdome has significantly better power and cooling savings. IBM highlights the mainframe's advantage in power and cooling costs compared to running one-unit rack servers with a single application. What they don't do is compare mainframe power and cooling costs to a more comparable, high-end system like a Superdome. But if I was in their position I'd be doing the same thing.

Has IBM been making your job harder lately with the new features on the z9 and the specialty engines?

Pickett: I wouldn't say they've made it harder, but let's give credit where it's due. They're a good company and they've been doing a good job. But the fact remains that if I've got Linux, am I going to be running it on a mainframe? An IFL [Integrated Facility for Linux] on a mainframe is $95k now? The question that needs to be asked is how many Linux servers does it take to justify that? It's certainly made it a little more interesting. I like my job because you're always learning something -- and staying ahead of where IBM wants to take it makes it fun.

Let us know what you think about the article; e-mail: Matt Stansberry, Site Editor


There are Comments. Add yours.

 
TIP: Want to include a code block in your comment? Use <pre> or <code> tags around the desired text. Ex: <code>insert code</code>

REGISTER or login:

Forgot Password?
By submitting you agree to receive email from TechTarget and its partners. If you reside outside of the United States, you consent to having your personal data transferred to and processed in the United States. Privacy
Sort by: OldestNewest

Forgot Password?

No problem! Submit your e-mail address below. We'll send you an email containing your password.

Your password has been sent to: