The free sharing of ideas, successes, and failures around Linux clustering was, in part, responsible for the rapid growth of Linux clusters. But, where are all the HPC user groups now?
Back in the winter of 1998 a small group of people held a meeting in Santa Fe, New Mexico to discuss Linux Clustering. The meeting was sponsored by the Advanced Computing Laboratory at Los Alamos National Lab (LANL) and the group was intentionally kept small as the event was by invitation only. Not that there would have been large crowds in any case. At the time Linux or Beowulf clusters were often considered bit of a back room lark that would never amount to much. I mean, who in their right mind would build supercomputers from off-the-shelf hardware and some free operating system written by a graduate student in Finland?
We all know how the Linux cluster story turned out, but often the important sidebars go unmentioned. For example, the little meeting in Santa Fe mentioned above. Somehow I was invited and found myself sitting in a meeting with some scary looking guy named Jon ‘maddog’ Hall, a guy who seemed to know a lot about Ethernet named Don Becker, then there was this excitable guy named Bob Young from a small company called Red Hat.
There were others, like Dave Turek of IBM, Walt Ligon from Clemson, and of course the guy who organized it all, Pete Beckmann from LANL. To keep the discussion open and honest, there were no precedings or documents created at this meeting. There were some presentations, but mostly we talked. And, we talked some more. The discussions ranged from clusters to Linux, to TCP…
Please log in to view this content.
Not Yet a Member?
Register with LinuxMagazine.com and get free access to the entire archive, including:
