NWME: Why do we need to scrap Big Iron?
It’s because every customer we talk to typically run into the same three things, over and over again. One is the high cost of mainframe because it’s a monopoly world. Almost 70% of what you run on these systems is not modernized. And because it’s a well maintained monopoly, it throttles innovation. Customers usually spend 80% of the budgets straight on maintenance, not innovation at all. The other issue is the massive skills shortage. Because these systems were built 20-30 years go, there is an increasing shortage of skills worldwide. As a result, it takes a lot of time to put anything into production. Users are struggling to put new functionalities into production because of the legacy aspect of the stuff that runs on these mainframes.
NWME: But Mainframes do have advantages in power and cooling?
Not as far as we are concerned. If you compare quality of service, scalability, availability, security or price-performance, open systems have improved significantly while mainframe improvement has been very flat.
NWME: IBM has been adding many new features to its mainframe lately
They do. But again, this whole thing is not around whether their product is better than ours. As I said, 70% of what runs on these systems are legacy technologies such as COBOL, CICS, etc. You can add lot of features to the mainframe but that doesn’t change anything in terms of monopoly. It doesn’t bring you choice and competition. And we provide systems that are as good as mainframes. Whether it’s management, performance, capability or throughput, there is nothing we cannot do with our Superdome environment that IBM can do with its mainframe systems.
NWME: Don’t you think mainframe is an attractive proposition for data centre consolidation?
Yes you can consolidate on mainframes, but mainframe is not the deployment platform of choice of any standard packages. For example, you can’t find Siebel on mainframes.
NWME: Which are the ideal candidates for migration? Most users are worried about the time and cost it takes to migrate.
We are open in terms of the methodology we adopt to migrate. What we look for is to find the lowest risk possible migration to move to open systems technologies. There are two ways to do it. One is to modernize applications, which has an impact on business logic, and other way is to modernize the technological infrastructure. The last we do is through our modernization factory in Spain, which is actually a centre of excellence created 10 years ago. In this mode, we take the mainframe code, put it through the factory and the outcome would be open systems codes, without changing the business logic. This is what we call rehosting and infrastructure modernization and the results are very impressive. What we promise is TCO savings of over 50%, payback within 24 months and we have never failed in a project. In EMEA to date, HP is engaged in over 250 mainframe related modernization projects. We offer the entire product and services to completely replace the mainframe environment.