Sun guru foresees data center disaster in '08
December 07, 2007
One of the top engineers at Sun Microsystems this week predicted that a large data center would suffer a "massive failure" during the next twelve months, causing "major national effects," including possible "national security issues," and underscoring "the importance of data centers as national assets.” In a talk with reporters, Sun vice president and distinguished engineer Subodh Bapat said the disaster would be on the scale of that caused when Robert Morris unleashed his infamous Morris Worm on November 2, 1988.
It's unclear what Bapat believes will cause the meltdown, but it appears to be related to the vast electricity requirements of today's utility-scale server farms. Bapat pointed to a new data center currently being built for a "national lab" that will suck up 50 megawatts of power, more "than a small city would consume." Bapat said that "utilities are going to become a real problem" for such megacenters.
There have, of course, been many dire predictions about network disasters in the past, and they have, thankfully, gone unfulfilled. But whether or not Bapat's forecast pans out, he does point to a real and growing vulnerability in the new era of web-based computing. The internet was constructed as a relatively simple communications medium, the reliability of which was tied directly to its radical decentralization. No node on the net was particularly important, so the failure of any one node had little effect on the operation of the system as a whole. But now that the net is becoming a computing grid, delivering complex software services and massive amounts of data, it is also becoming much more centralized. While there's still no single point of failure, the rise of massive data centers serving millions of users means that the failure of a major center could have wide ramifications. And given the net's importance as an infrastructure for commerce, such a failure could indeed damage national security (as I discuss in some detail in Chapter 9 of The Big Switch).
Bapat is also right to point to electricity as a weak link in the system. The centralization of computing can bring dramatic increases in overall energy efficiency, compared to the fragmented, subscale private data centers we have today. The megacenters tend to operate at much higher levels of capacity utilization than the private centers can achieve, and their operators have the skill and wherewithal to install cutting-edge power-management and cooling technologies. At the same time, however, the centralization of computing assets concentrates energy demand, putting new strains on the aging electric grid. Efficiency increases, but vulnerability does, too.
Whether it's a failure on the scale that Bapat anticipates or a series of smaller problems, it seems inevitable that politicians will soon be forced to wake up to what Bopat accurately calls "the importance of data centers as national assets.”
Electricity is much more vital for people not for computers. Big cities will die in Apocalypse without it. And it really Doesn’t Matter that Internet or global banking may die – people will fight for water and food.
IT specialists often overrate importnace of IT and forget real human passions.
Posted by: Mikhail Elashkin at December 8, 2007 03:01 AM
I think it would be far better for major data center owners like Google and MS to diversify more, and build future farms next to hydro-electric plants, and then move the data to and from with multiple secure fiber pipes. Here in Canada, BC, Manitoba, James Bay and the North Shore of the St. Lawrence are all ideal sites. Oh, and the cooling costs will be much lower as well.
I am not a Sun engineer or executive, but my one specialty, mobile applications for the service industry, leads me to the following.
It's a matter of time. We are creating more meshed services where the seemingly unitary application is really a composite of linked calls across the internet via REST and SOAP. If one service fails, the whole app is useless.
I even recently came across a mobile application that was used for automated allocation of service resources and repair inventory. The system had recently replaced, or rather, augmented, the inbound phone system. Yet, the success of the J2ME mobile app was such that the client reduced their inbound phone and staffing services to such an extent that the manned service consoles could never shadow the mobile app in times of a crisis due to reduction in infrastructure and manpower.
Do you realize that this business relies 100% on this system and has no contingency for a failure.?
Im not even certain such a composite services mix can be made 'fault tolerant'. We are in for some interesting times.
Posted by: abm at December 9, 2007 04:48 PM
Sun management is looking a little desperate so they are re-hashing the Y2K fears: if you don't buy more stuff, there will be a global melt-down. Sun was a lot more fun when Bill Joy was there: quoting Unabomber Theodore Kaczynski and predicting the end of humanity from nanotechnology. Somehow, the prospect of some spook kicking out the plug on a rack of servers doesn't interest me like the prospect of being eaten alive by nanobots!
About the Morris Worm, perhaps one of the RT old times can correct me, but the security vulnerability the Robert used was a buffer over run that allowed stub code passed on the program stack to run as root. As I recall, Robert learned about it from his dad: he didn't discover it himself. It was a rather obscure programming error in a few utilities; contrast that with M$ Windows – seems like there is at least one massive worm attack against Windows each year. UNIX, particularly BSD, has redeemed it self to be the most secure and reliable OS know to man while Windows is a sitting duck for attack. Ok, Windows is secure – just as long as you don’t open your mail!
Posted by: Linuxguru1968 at December 9, 2007 09:48 PM
I don't see Sun implying that we need to buy more of their wares. Its a security threat and no more.
It's a good provocative story and one that should make us realise how important the net is to furthering out understanding of any situation. If the ability to understand is removed I'm guessing mainstream media is the last to report objectively.
Posted by: Charles Frith at December 13, 2007 07:32 PM
>> I don't see Sun implying that we need to buy
>> more of their wares.
Ha! If you read the article, I think you will find the contrary:
"On a more cheery note, Bapat and other Sun executives said that the IT industry is also on the verge of a construction boom that, if it happens, will lead to big orders for equipment for makers of servers, storage systems, and other data center equipment."
This is a common sales tactic: make some dire prediction then pitch your product. The subliminal implication is that buying your product will prevent the catastrophe. IMHO, Joy and Bechtolsheim were always optimists; its kind of depressing to see the current management has to resort to scare tactics.
Posted by: Linuxguru1968 at December 14, 2007 04:44 PM
Post a comment
Thanks for signing in, . Now you can comment. (sign out)(If you haven't left a comment here before, you may need to be approved by the site owner before your comment will appear. Until then, it won't appear on the entry. Thanks for waiting.)
"Riveting" -San Francisco Chronicle
"Rewarding" -Financial Times
"Ominously prescient" -Kirkus Reviews
"Riveting stuff" -New York Post