The economy didn't help much, but this was still an exciting year in the storage industry. Here are my picks for the Top Trends in 2003, in no particular order.
Disk-based backup. End users have been waiting for D2D to be an economically feasible alternative for years, and now it finally is, thanks to low-cost ATA and Serial ATA disk drives and some nifty new software. The only problem now is that there seem to be dozens of vendors trying to get a piece of this lucrative pie, and they all have different ways of implementing disk-to-disk backup. If you're confused, re-read industry analyst Dianne McAdam's article, "Evaluating options for disk-to-disk backup," October 2003, p. 34.
IP storage and iSCSI. Okay, before we goof on the snail's pace progress of iSCSI, remember that Fibre Channel took just as long (longer, actually) to go from specification to end-user adoption. In the case of iSCSI, it has been three years since IBM and Cisco hammered out the spec, but this year a number of watershed events occurred for iSCSI, starting with the ratification of the standard early in the year, followed by Microsoft's software drivers mid-year, and announcements of support from big names like EMC and Network Appliance. Not to mention Cisco, Intel, Adaptec, and a dozen or so start-ups. If you want to see how and why your peers are using iSCSI, re-read senior editor Lisa Coleman's article, "Low cost lures users to IP SANs," October 2003, p. 1.
Mergers and acquisitions. Storage acquisitions came rapid-fire this year, at times seemingly on a weekly basis. Let's review a few: Brocade-Rhapsody, Sun-Pirus, IBM-TrelliSoft, Veritas-NTP, Veritas-Precise, Veritas-Jareva, EMC-Legato, EMC-Documentum, McData-Nishan, McData-Sanera, AMCC-JNI, Emulex-Vixel, NetApp-Spinnaker (see related article in this issue). For sheer chutzpah, the World Domination award goes to EMC, but Veritas gets an honorable mention for also venturing beyond the storage playground.
Start-ups. Between 1999 and 2002, an unprecedented number of storage start-ups were funded—unprecedented for any IT segment. A lot of these start-ups bit the dust under the hard heels of the economic stomp down, but in 2003 some of the fledglings managed to get more funding, keep the VCs at bay, and actually ship product. Impressive, indeed, but that was the easy part. The hard part will be selling against the entrenched behemoths. But hats off to the start-ups because they at least woke up the market leaders and forced them to try to solve end users' problems instead of just pushing profits.
Serial ATA. "Good-enough" storage is good enough for many end users. Arguments over whether Serial ATA is "enterprise" capable (whatever that means) will continue, but for the majority of applications—new and old—Serial ATA-based disk arrays are good news for users with tight budgets.
Storage area networks. Okay, so SANs weren't a 2003 trend, but they're what kept the storage industry exciting, confounding, profitable, and challenging. SANs lined the pockets of a lot of vendors and made heroes out of IT managers who were actually able to implement them, manage them, and prove the resulting business benefits. Next year, we'll see how IP SANs stack up against their Fibre Channel forefathers.
Fabric-based intelligence. This will be an interesting trend next year, but since the big switch vendors made so much noise about it this year (and spent so much money trying to get a place at the table via acquisitions), it qualifies marginally as a 2003 trend. I also included it because our readers think it's cool (although that may be due to a delusion that running storage applications on switches will be cheaper than running them on hosts or disk arrays). Let's see: If you have a volume manager running on 24 hosts in a SAN, and you switch to two copies of the volume manager running on two switches in the SAN, will it be one-twelfth the cost? Not.
Information life-cycle management. Nah, let's save this one for 2004.