HP 3000 Product Futures at Fresche Legacy
Protecting HP 3000s Using Linux

Unix futures take Odyssey to good enough

Customers who are being wooed toward Unix from the HP 3000 have some good right-now reasons to choose HP-UX. The power of virtualization and ability to exploit an OS+hardware solution make the HP Unix an enterprise-grade choice -- one of the reasons that HP and its partners work to sell AS/400 and IBM mainframe sites on this switch. If just as with the HP 3000, you have an integrated value in the IT center, software built for specialized hardware like Itanium makes sense today.

The future might not be great, but good enough. HP's Odyssey project wants to bring "hardened" features to Linux, an OS more 3000 sites are now choosing when they move. Europ Assistance is the latest 3000 site we've learned about that's adopting Linux. HP doesn't want to be left out of the Linux currents. While there's a clear five-year future of HP-UX, the years beyond that are less defined. Since companies like Europ Assistance are going to take multiple years to make a migration, few of them want a future shorter than a decade.

Even the friends of HP's enterprise strategies see HP's Unix as an early casualty of the Odyssey. Dr. Bill Highleyman edits the High Availability Journal and judged the prospects of Odyssey success.

If Project Odyssey is wildly successful, it may drive a huge competitive advantage for HP. However, if HP customers embrace the move to highly reliable standard operating systems, HP-UX may be the first to go, since migrating Unix applications to Linux is a reasonable task.

It's commonplace to find HP-UX administrators on the LinkedIn forums who see Linux as their natural evolution path. But those companies are already enjoying the value of Unix, instead of paying for the move. It takes unusual features in an OS to protect it from this kind of wild success -- and as HP 3000 customers know, even a tech solution that is great can be overrun by good enough.

Odyssey can't deliver as much as HP's proprietary environments, such as MPE. Highleyman noted in his article that the fault-tolerance -- 3000 customers would call it reliability and planned-only downtime -- in HP's operating systems won't make the Odyssey.

Achieving the fault tolerance provided by NonStop systems and OpenVMS Split-Site Clusters is probably not in the cards. Sadly, if the reliability provided by hardened Linux and Windows systems is good enough, the market may see a declining need for great, continuously available systems. Let’s hope that great triumphs over good enough!

In the same way, the Odyssey analysis at the High Availability Journal hopes that IBM's multi-OS mantra will mean success at HP.

IBM’s proprietary operating system zOS has survived living alongside a hardened Linux. Hopefully this is an indication that the HP proprietary operating systems will survive alongside HP’s hardened Linux and Windows.

But HP's Business Critical Systems GM Martin Fink points out the differences in the HP and IBM enterprise strategies for Linux, not their similarities. 

IBM’s strategy is not at all like Project Odyssey. IBM’s Linux is a proprietary Linux. Applications have to be recompiled to run on the mainframe. IBM’s strategy is to extend the reach of the mainframe, and its proprietary Linux has not been all that successful. Project Odyssey is radically different because we do everything with one open platform.

HP 3000 customers have heard this single-platform pledge before, when the Spectrum Project was supposed to span three operating systems with a single hardware architecture. By the time it was released, one of the OS's was on its heels (RTE) while the other two fought it out for dominance in HP's strategy. The more popular and standards-based OS won. We're still looking for a reason why Linux won't do the same to HP-UX, the least fault-tolerant of HP's environments.