April 07, 2011, 11:56 AM — At 20, the formerly underground, often dismissed, cult-worshipped, many-flavored, unusually stable, increasingly compatible open-source Linux operating system is getting ready for another big step into the enterprise with high availability features previously found only in commercial OSes.
The High Availability Working Group within the Linux Foundation is specc'ing a version of the OS stack designed for five-nines uptime when running on clusters of ordinary servers, rather than the highly redundant hardware HA operating systems have typically used in th past.
Clustering can give servers "a higher availability than any single one of them would achieve," according to Lars Marowsky-Bree, a Novell engineer who is the working group's architect.
The Foundation announced the new working group and new goal at its conference yesterday in San Francisco, where it also announced version 5 of its Carrier Grade Linux, which is designed with far higher reliability in its file systems, more data portability features, support for HA-enabling redundant hardware, better diagnostics, performance tuning and debugging.
CGL became available in 2002, three years after the development project was begun in 1999.
Linux has made a lot of progress since then in technical quality and capability. More importantly it's been accepted as a mainstream data-center OS rather than something to run on end-of-lifing hardware when the data-center crew needed another firewall or access or other single-function server but no budget to pay for it.
It's become accepted enough to let its public mouthpieces turn into smugly superior jerks – the antithesis of the modest, inclusive approach of Linux inventor Linus Torvalds. ("My life isn't glamorous," a recent blog entry begins.)
"I think we just don't care that much [about Microsoft] anymore," Linux Foundation executive director Jim Zemlin told Network World. "They used to be our big rival, but now it's kind of like kicking a puppy."