We've been hearing whispers about honest-to-goodness server Macs for the better part of a year, now, and we've still got zilch to show for it; Apple's idea of a server is still just a regular desktop Power Mac with a hefty chunk of extra software on the hard drive. In other words, we're talking no redundant hot-swappable disks, no redundant hot-swappable power supplies, no rack-mount-friendly enclosures, and a whole host of features that are pretty close to useless in a box that's generally going to be running unattended in a closet somewhere. Really, who needs a RADEON with 32 MB of DDR RAM in a server that would ideally be run without a display most of the time anyway?

So, desktops with software bundles-- that's where we are now, and that's where we've been for ages, probably ever since Apple junked those mammoth Apple Network Servers that ran AIX, IBM's flavor of UNIX. We'd always had a feeling that Apple probably wouldn't get serious about servers again until 1) Mac OS X had hit its stride, and 2) the company was ready to tackle the enterprise market with a vengeance. After all, who else but big business IT geeks would get all hot and bothered about relatively drab rectangular slabs designed to be bolted into a metal rack?

Answer: why, hardcore bioscience nerds, of course! Faithful viewer gumby tipped us off to a WIRED article which describes how genetics researcher Patrick Gavin, like others in his field, is clustering dozens of what amount to individual desktop-class personal computers into a massively parallel (and, relatively speaking, dirt cheap) supercomputer for the sole purpose of crunching DNA data. And he'd love to use Macs, because "the PowerPC architecture is vastly superior to anything else out there in terms of power consumption versus processing power." The problem is that Power Macs are simply entirely the wrong shape; they aren't really stackable and they take up too much space if you're looking to cram eighty of them into a room-- let alone several hundred. So, no Mac cluster for Patrick.

Now, while Apple has been extremely reluctant to go paddling into the enterprise waters where Dell reigns supreme, it's no secret that Cupertino has been courting the science market pretty heavily for a while, particularly in the area of genetics research. Between Art Levinson of Genentech joining Apple's board of directors, a distinct pitch of the UNIXy power of Mac OS X in science circles, and the recent release of Apple's improved version of the bioinformatics tool BLAST, clearly this is a market that Apple wants to own. Since that market apparently wants "a scalable, high-density hardware solution," hey, maybe we'll see some specialized clusterable Mac servers sooner rather than later. And that'd be good for us here at AtAT, too-- because while we're about as interested in mapping the human genome as we are in ironing our socks, we wouldn't mind serving AtAT from a svelte little rack guy someday.