First impression on unpacking the Q702 test unit was the solid feel and clean, minimalist styling.
Even storage that isn't clustered wants to be
- — 08 May, 2008 09:36
Soon it will seem like every storage vendor with a name will have a clustered storage box. Why? In a word: Web 2.0. It would appear that your father's RAID array just doesn't cut it anymore with the purveyors of Facebooks, YouTubes, and SmugMugs. Your father's RAID can't scale, it's too slow, and it's decidedly not sexy anymore, even after a vial of Viagra and a trip to The Hair Club for Men.
Personally, I've encountered some who can look at a storage grid straight in the eye and pronounce it "clustered." Ditto with some new arrays that blend-in a heavy dose of parallel processing.
Clustered storage is decidedly in vogue. Even storage that isn't clustered wants to be. And that's the issue. You may think you know what it is, and then you'll see another "clustered storage" product that is decidedly different from what you thought it would be.
Before you look at another "clustered" storage box, I suggest taking a big time out. First, fix in your own mind what you think it is. I began to see the word "clustered" used in the context of NAS a few years ago when the industry was looking for ways to address the "you'll love your first one but you'll hate your tenth one" problem. Back then, clustered storage was contiguous to the global namespace. Now it's invaded every major storage category - file-based, block-based, object-based - you name it. The Storage Networking Industry Association's online dictionary is of no help here.
Next, once you think you know what it is, decide whether or not it is in fact what you really want. If for example you primarily want raw performance for your Web 2.0 or video surveillance application, clustered storage may or may not fit the bill. A heavy dose of parallelism might be more like what you're after.
I enjoy historical inflection points like the one we're living in right now. For me, the truth is that my father's RAID array needs to accept its rightful place in the history of computing. What follows will be a parade of new storage approaches and architectures, and that's exciting. But please, don't try to hail them all as clustered.
John Webster is the principal IT adviser for research firm Illuminata. He is also the author of numerous articles and white papers on a wide range of topics and is the co-author of the book Inescapable Data: Harnessing the Power of Convergence ( IBM Press, 2005). Webster can be reached at firstname.lastname@example.org.