CES 2009: When (if ever) will home servers shine?

Though undeniably useful, NAS (network-attached-storage) products thus far have proven difficult for consumers to grasp and implement. Will the category ever catch fire in the mass market?

By Brian Dipert, Senior Technical Editor -- EDN, 1/7/2009

As is often the case, I've been a technology early-adopter with respect to home (and SOHO for that matter, since I have a home-based office) NAS (network-attached storage). Considering that I've been dabbling with network storage gear for a decade, though, you might think that by now consumer NAS would have either withered on the vine or blossomed into a broadly embraced product category. That neither spectrum-of-adoption endpoint has come to pass tells me two things:

1. The technology is compelling enough that it's cultivated a sufficiently large enthusiast base to keep it on life support, but
2. The concept isn't sufficiently intuitive to ensure must-have purchase by a larger population of consumers, and/or the implementations haven't been sufficiently intuitive to ensure long-term adoption (along with subsequent advocacy to the next level of friends, acquaintances, and family members).

This is the second day of the Storage Visions show preceding CES. Microsoft aggressively promoted its Windows Home Server NAS O/S at last year's show (a campaign that was largely neutered by the concurrent discovery of a nasty file-corruption bug that took many months to fix, likely because most of Microsoft's software engineers were otherwise focused on getting higher-leverage Windows Vista SP1, Windows XP SP3, and Windows Server 2008 out the door). Both factoids rationalize an in-depth exploration of the NAS topic at this time. Numerous NAS suppliers (coming from both hard-drive and networking heritages, along with plenty of previously unaffiliated companies) are in attendance here in Las Vegas, and the product category is of equivalent interest to many other CES-attending vendors hungry for tangible incremental business opportunities: HDD and CPU manufacturers (the latter both x86 and other), and developers of DRAM, Ethernet, and other silicon and software building blocks.

To my mind, home NAS applications cluster into a few high-level categories:

• General-purpose shared access to generic files for reading, updating, and other management tasks
• The specific ability to access multimedia files (music tracks, pictures, and video clips) via UPnP/DLNA, Bonjour, and other discovery protocols, over TCP, UDP and other transport protocols, and by various LAN-connected clients (general-purpose computers, dedicated media extenders, handheld devices, etc.).
• An expanded ability to share files with others around the world over a broadband WAN link to the NAS, and
• Full and incremental backup of computers' valuable stored information, to account for inevitable HDD demise.

At first glance, the market opportunity for NAS in the home might seem robust and ripe, judging from the booming sales of routers suggestive of an expanding home LAN ecosystem. My admittedly unscientific regular surveys of friends and family members, however, suggest otherwise. The overwhelmingly dominant (and solitary) use of a router in most homes, my research suggests, is to enable shared access to a common Internet connection. Much less frequently, consumers take their LANs to the next level, that of sharing a networked printer.

Unfortunately, though, the concept of enabling computers to talk to each other for file-sharing purposes seems to be beyond the intuitive grasp of most consumers. Shared access to a common storage "pool" is equally alien. This dark cloud does have a silver lining; each time I've pointed out to someone that computer-to-computer and computer-to-networked storage communication is possible, along with explaining (even if only briefly) the advantages of such a scheme, the light bulb immediately goes on and they enthusiastically embrace and implement (and subsequently regularly employ) the idea. This result suggests that a little advocacy of the concept of networked storage in the home by relevant vendors and industry groups will go a long way toward achieving the desired end result.

NAS's lingering IT heritage, translating to consumer-unfriendly setup schemes and other overly techie issues, doesn't help. To that point, Oxford Semiconductor (representatives of which I met with yesterday, and which, thanks in part to recent purchaser PLX Technology, is aggressively moving beyond its FireWire direct-attached storage roots toward both USB3-based DAS and Ethernet-based NAS, the latter exemplified by a just-introduced Iomega storage device implementation, pictured at left) is advocating a renaming of the acronym to "Network Accessible Storage." Similarly, I suspect that whereas the bulk of today's consumer NAS business is supplied by legacy storage firms, over the long term two alternative and more consumer-cognizant vendor categories will dominate; networking equipment companies such as Cisco/Linksys, D-Link, and Netgear (in no small part because I suspect that HDD-inclusive routers will become increasingly prevalent in the future), along with computer OS providers like Apple and Microsoft. More on that latter category in a bit.

With respect to NAS's intra-LAN media-sharing facility, the longstanding codec and protocol incompatibilities between source and destination devices have unfortunately been a recurring theme in my print and online hands-on evaluation reports over the years. Note Monday's writeup, for example, or a piece I crafted last summer in conjunction with a trip to Taiwan. Single-vendor network topologies are inherently less prone to such problems, and as such I was a little surprised that Apple didn't unveil a media server during yesterday's Macworld Expo keynote. Granted, Apple's routers do already support USB-tethered HDDs, and the company's also done a decent job of making intra-Mac media sharing simple. But I still expect Apple to eventually unveil either a NAS-centric device or a multimedia-optimized version of the current router-plus-HDD Time Capsule. Don't expect NAS-housed data access to extend beyond the LAN, though, as that would negate consumers' need for a lucrative-to-Apple $100/year MobileMe (previously .Mac) subscription.