Strategic Info Management: Wide Area File Services

Lewis Carroll would've had a field day satirizing the re-emergence of WAFS (Wide Area File Services), a storage industry acronym with as many meanings as there are vendors offering products. Chase this particular white rabbit down its hole, and Alice the IT manager could embark on a journey at least as bizarre as her namesake's trip to Wonderland.

WAFS defines techniques for facilitating the sharing of files across a wide area network--little more than remote file caching. Say Pattie in Poughkeepsie wants to edit a file that's in use by Dan in Des Moines. Pattie should receive a locked version of the file until Dan completes his work.

From this perspective, WAFS sounds straightforward enough. It establishes a set of coherent rules for file access and sharing, even in cases where there are different OSs and file systems in play. Protocols such as Network File System (NFS), originally from Sun Microsystems; Common Internet File System (CIFS), now quizzically called "SMB" by Microsoft; and HTTP were all created years ago to facilitate long-distance file sharing over networks.The problem for WAFS vendors is features like file locking and version synchronization have been adapted to these de facto standards over time. Why aren't these file-management workhorses enough? Why haven't we built out extensions to NFS version 4, say, to supply lock-and-sync functions, rather than building proprietary algorithms that do the same thing while locking us into a particular vendor's wares?

The contemporary cadre of WAFS vendors, including Availl, Brocade Communications Systems, Cisco Systems, Packeteer (with its acquisition of Tacit Networks) and Riverbed Technology, might reply that NFS and its cousins don't go far enough. With files spread all over the contemporary distributed enterprise, WAFS addresses the additional issue of how to ensure that important documents are adequately protected from catastrophic loss or risk of disclosure. Conventional backups don't always work. WAFS vendors argue that better protection is required for the company's most irreplaceable asset: its data.

We'd buy that, except it's rare that WAFS is deployed with data protection front-of-mind. Rather, the technology seems to come to the fore only opportunistically, as a bolt-on to fix problems created by other architectural decisions.

The value-add of WAFS doesn't have much to do with enabling concurrent use or speeding up file access, either. For these arguments to make sense, the file must have been accessed at least once so that it's in a local cache. The first access is just as slow as it would be without a cache at all. Once cached, file access may be faster, but then you must keep the remote cached copy synchronized with the original, and vice versa. A more sensible strategy would be to leave the data in place and establish appropriate file locks.

Vendors regularly tout WAFS for disaster recovery because it's the only use case that would seem to make any sense: Centralize data so it can be backed up in a disciplined way, then echo cache copies so users don't get irritated by the World Wide Wait every time they need access to a file. Nice, but again, not crucial.Bottom line: In a non-Wonderlandesque world, WAFS may be the best technology you never need to deploy.

Jon William Toigo is a CEO of storage consultancy Toigo Partners International, founder and chairman of the Data Management Institute, and author of 13 books. Write to him at [email protected].