Micro-Rebooting

Friday, Nov. 12 2004 | Author: Nicole

The new concept of micro-rebooting takes the presence of software bugs as a given and, instead, works toward shortening recovery time.

That’s the concept behind “recovery-oriented computing,” a 180-degree turn from traditional thinking. The idea is that since software can’t be created without crash-causing flaws, it should be built to reboot much faster, allowing users to get back to work almost instantly.

Micro-rebooting as developed by Stanford and UC Berkeley researchers is built in the vein of recovery-oriented computing. Micro-reboots are used to quickly terminate enough of the problematic program processes to restabilize the system. The researchers have developed algorithms that determine software baseline processes, monitor them, and do micro-reboots if necessary.

The first software most likely to have this technology built into it? J2EE application server software. (Don’t laugh.)

[ via ACM TechNews ]

Category: News

4 Responses for "Micro-Rebooting"

Norgus

November 13th, 2004 at 4:10 pm
1

Microsoft should love this idea. Don’t create good code, make rebooting faster!

Interesting idea, but couldnt it lead to an excuse for even more sloppy coding?
Nicole

November 14th, 2004 at 1:06 pm
2

That’s kind of how I feel Norgus. There’s already a “well, just reboot” attitude, when I think the goal should be to push for higher quality code. But, maybe I’m too much of an idealist.
JC

November 15th, 2004 at 6:22 am
3

well, it’s a more effective solution. Get thousands of developers to write better code (most of whom probably don’t knowhow to or they’d have done it in the first place) or put together a software package to speed up rebooting where old and new software can benefit. After all, you’re going to have to reboot it on occasion anyway, even if it is running linux or something. (some of the linux security patches and the like require reboots, right?)

My take though is… if you’re in a situation where you can’t handle the downtime it takes to reboot a server, you shouldn’t be running in a single server environment. Cluster/load balance a couple of boxes and when one goes down, the other takes over the complete load and no one notices anything happening.
lucas

December 2nd, 2004 at 9:49 pm
4

this site is lots of fun