Can HPC become a commodity?

HP's acquisition of PolyServe suggests commoditization is possible, but some assembly is still required

My first reaction when I heard the news of HP signing on the dotted line to acquire PolyServe was: "What took them so long?"

[ MarioApicella's column is now a blog! Get the latest storage news from the Storage Adviser blog. ]

The two companies have been partners for quite some time, and there was a "high degree of potential synergy" between their offerings. I don't particularly like that expression, but if there was ever a slam dunk in this business, HP buying PolyServe was one.

Here's why: PolyServe creates a powerful software bridge between servers and storage, essentially giving business applications and file servers fast, parallel data transfers and the potential to crank up performance and resilience on demand.

Why is this important for HP? Because HP has an insatiable appetite for new customer wins or repeat sales on both ends of that server-storage bridge. If you've been watching the same movie as I have, you'll agree that there has been little synergy between those two powerful groups (servers and storage) inside HP.

Take, for example, NAS systems, which are a typical confluence point of servers and storage devices; they also create opportunities for joint purchase orders. For several years, HP has been quite successful in combining a joint offering of storage systems and servers topped by the Microsoft Windows Storage Server.

Despite that success, the glue to put together those systems -- the Microsoft OS -- came from outside HP. PolyServe creates a similar type of bond between servers and storage devices, promising an appealing level of performance and resilience.

PolyServe's customer roster already contains a wide range of vertical markets. That's because PolyServe is basically an HPC (high-performance computing) solution that business customers can deploy and manage without putting too much stress on their IT structure or their staff. With PolyServe, customers get the extra performance required by, say, the deluge of multimedia files common to every business sector and can keep using their same servers and storage devices.

It's no surprise that PolyServe attracted other partners, including Dell, IBM, and Microsoft. Of course, I should also note that PolyServe is OS agnostic and can speed up, for example, Linux as well as Windows files systems. Moreover, PolyServe pumps up some clustering muscles in Windows servers that Redmond has not yet been able to build.

As is often the case when an acquisition is still in progress (it should close in 60 days, I am told), we are not going to get satisfactory answers to what we can expect to see from this acquisition in the future. To be fair, HP and PolyServe have put together an FAQ on the acquisition, but that short list tells us mostly what we already know or can easily guess.

When the acquisition dust settles, however, I would like to see an easier-to-install solution emerge. The last time I checked out PolyServe, my major concern was a rather rough sailing during deployment. If HP wants to boost its sales of storage and servers and make HPC available to a larger base, fixing those deployment issues should be its first priority.

Join me on The Storage Network with questions or comments.

Copyright © 2007 IDG Communications, Inc.

How to choose a low-code development platform