Ads by TechWords

See your link here
Receive the latest technology news and information.
Hardware
Computerworld Daily News (First Look and Wrap-Up)
Computerworld Blogs Newsletter
The Weekly Top 10
Cloud Computing
View all newsletters




Privacy Policy
 

'Get Back' to IBM

February 25, 2002 12:00 PM ET

Computerworld -





Join the online discussion about this column.




A couple of weeks ago, I suggested that IBM has an ace in the hole called "hardware devolution." The best way to describe hardware devolution is to examine the so-called evolution of the PC. If the word Beatles makes you think of screaming teens fainting over the Fab Four, you're probably old enough to remember how the first PCs secretly invaded our workplace. "In My Life," I recall that the
"Revolution" began because whenever anyone wanted something from computer services, the answer was "Don't Bother Me," "You Can't Do That" or, worst of all, "No Reply." The PC gave users the ability to circumvent computer services. Naturally, the company policy was often that, " 'If You've Got Troubles,' it doesn't matter how much you 'Twist and Shout' for 'Help.' 'Think for Yourself,' because we're not going to 'Carry That Weight.' " But we kept on using the PCs "In Spite of All the Danger," because when they worked, "A Hard Day's Night" at the PC usually gave us the answers we wanted.
Then we got on "The Long and Winding Road" of networking so that our PCs could "Come Together" to share resources, such as printers and hard drives. Computer services morphed into IT, which better described the broader responsibilities of managing all the networked desktop computers and everything stored on them. The computer room started filling up with servers. Thanks largely to "The Fool on the Hill" in the Northwest, everyone wore beepers and worked "Eight Days a Week," "Fixing a Hole" "Here, There and Everywhere." As the problems increased, so did the budgets to address them.
The demands of enterprise-scale applications, followed by the need to provide high-availability services on the Internet, meant that IT had to purchase more PC servers to handle those demands. More servers meant more components and more software installations, and IT departments had to look after "Every Little Thing."
Vendors figured, "We Can Work It Out" and make some "Money (That's What I Want\)." So some vendors consolidated servers into racks, and others created KVM switches to manage several servers with a single monitor, keyboard and mouse.
So there they were. Gobs of computing power in dozens of servers all in one rack, and no easy way to distribute and redistribute the various tasks and loads across them. There's clustering, but clustering is easy "For No One." If you can solve that problem in a hardware box, "Baby, You're a Rich Man."
Enter server blades, which are low-power server cards that you can combine on a backplane in less space than a rack. Blades are easy for just about anyone. At the low end, for $600 a pop, OmniCluster has a SlotServer you can plug into the Peripheral Component Interconnect slots in your existing server machines. One server suddenly becomes two, three, four or more. And OmniCluster runs the fastest-growing server operating system, Linux, in addition to Windows.
At the high end, you've got solutions like Egenera BladeFrame, a 96-processor box that also runs Linux. This puppy provides fail-over support for any blade, and its special control blades let you redistribute computing power in real time.
BladeFrame is pricey, starting at more than $200,000 (as of November, anyway), but considering what you get, it's probably worth every penny.
"I Don't Want to Spoil the Party" for the blade vendors, but "I've Got a Feeling" that as terrific as they are, server blades will have a limited shelf life. What is the BladeFrame? One box, many servers, reliable, real-time management of processing power. That also describes the new IBM Linux-based mainframe, which starts at approximately $400,000. The difference is that it doesn't need multiple servers. But if you do, it offers virtual servers.
I recall that Bill Gates and "Some Other Guy" named Stewart Alsop predicted a day when we'd unplug the last mainframe and replace it with a PC. "Do You Want to Know a Secret?" "I Should Have Known Better," but I used to agree. Now I'd have to say that the PC will never replace the mainframe. What PC-based servers like server blades may do, however, is devolve into mainframes. If that is indeed our future, IBM's Linux-based mainframe isn't a blast from the past, but a glimpse of the future.
"I'll Be on My Way." "You Know My Name (Look Up the Number)" or send me e-mail at nicholas@petreley.com.



Jump to comments

Hardware

Additional Resources

WHITE PAPER
Approximately 60 percent of data migration projects overrun time or budget, while some fail completely. Download this white paper, "Enhancing Your Chance for Successful Data Migration," to learn the critical steps you need to take to execute a data migration project with minimum cost and risk to your business.
WHITE PAPER
Read the Gartner research note to learn why the TCO of a server-based computing deployment used to deliver all applications to users is around 50% lower than that of an unmanaged desktop deployment.
WHITE PAPER
Economic downturns have a tendency to accelerate emerging technologies, boost the adoption of effective solutions, and punish solutions that are not cost competitive or that are out of synch with industry trends. This IDC White Paper presents the results of an IDC survey of 330 companies in Western Europe, Asia/Pacific and the Americas that measures the receptiveness to Linux and takes into consideration changing views driven by the disruptive economic environment that businesses face today.