What is the growth from VMS <negative> Should HP keep investing in it? <no>
There is limited capital and mindshare executives can spend on any enterprise.
VMS and Tandem represent sunset technologies, and no new opportunites.
Time and money spent on spoon feeding and wranging these dinosaurs means grids, and linux, and IPsans don't get attention, VMS and Tandem do. Its lost opportunites.
There are other companies that are expert at wringing out every last $ from a fading technology: Computer Associates for one.
Adding two giant legacy, fading, installed base only product lines really slows HP's progress. Imaging going into a new global, web savvy enterprise and pitching VMS. Impossible. Its a waste of time and resources and it always will be.
No opportunities? These are existing customers. When they need to upgrade, adopt new technology, are they going to see HP as a valued partner who supported them all toe way or as a callous vendor who ignored them and the platform their mission critical applications ran on?
Dump the 'old' technology before you have a viable alternative with a clear and manageable migration path and you have lost a customer, a lot of customers actually.
VMS customers are extremely loyal. HP has already ported the OS to Itanium. HP needs to find out all the stumbling blocks to the existing customer base moving over to the new platform.
Non-Stop is a potential gold mine. There is not other OS like it and the customers who run on it really don't have any other choices. It would be stuping to alienate that customer base. HP needs to figure out how to get the benefits of combination of the Non-Stop hardware and OS with its other hardware platforms.
Alienating VMS and Non-Stop customers would lock HP out of every financial trading business in the world. Pretty stupid.
Grids - I hear lots of hoopla about grids from software vendors who want a bigger piece of the IT pie. Have you tried implementing your application on a grid? Really, have you? Not just thought about, but ported and did stress tests. Have you gotten your developers to port a non-trivial application from SMP to grid? Come back when you have.
Linux - linux customers know the cost of everything and the value of nothing. What was that old saw - you get what you pay for?
IPsans - I see a big demand for that, I really do. Not. Maybe when 10Gb switches are cheap, but not now.
Perf, large numbers of first tier, multi-billion dollar financial enterprises are adopting Linux.
The stumbling block to selling VMS on Itanium is that the VMS folks don't want anything new.
Hardware HA is passe. The telcos were the pioneers in this and they are abandoning it. They are going for service redundency, not hardware redundency. An order of magnitude cheaper. In its simplest case: the DNS server. Anyone you know run it on Non-Stop? No. They run multiple DNS servers and keep searching for an 'up' server.
Does Google run Non-Stop? Every have a 505 Error from them? (Server unavailable) No. Why? Service redundency.
That's the way everything is going to go.
Best to get top dollar for these two vacuum tube hardwired behemoths while the base is still breathing.