[SATLUG] Next month's presentation
jeremy at bioc09.v19.uthscsa.edu
Wed Aug 28 20:22:01 CDT 2002
> Mosix is a good way to get the multiple copies of the client program
> running on the different nodes. Just start them on the master, and let
> them distribute themselves out. If a node fails, the data is sent to a
> known node, and the problem node will receive another set, if it
> requests it. After so many failures, the node is marked as bad,
> requests from it are ignored, and the administrator is notified.
I choose openMosix simply because it was something I never used before. I
had heard about it, but the cluster at work was already in place and
running smoothly. So I decided to try experiementing with it at home on my
server and my laptop. Yep, just 2 nodes at home ;) After experimenting and
tweaking this and that, I have it always running.
If you missed my last presentation, I do a lot of video processing. After
installing openMosix, I was simply surprised at how all the tools I use
work awesome across an openMosix cluster.
I just took a screenshot of my 'cluster' processing some video material
with MJPEGTools. Notice the Load Balance Efficiency. All 3 processors (2
in the server and 1 in the laptop) are using somewhat the same CPU usage.
This is so cool ;) The Gkrellm readout isn't accurate since the nature of
openMosix is that is shifts the load according to actual usage of the
computer. This was taken on the server (pos450) so while gimp was loading
it shifted the load the laptop (jeremy). Now (after the screenshot) the
CPUs on pos450 are 80 and 60 respectively.
> Just the results of my experimentation...
> I think I'll show up for the next meeting also. Could be fun!
Cool! We need more clusterphobes at SATLUG :) hehe...Cya there!!
jeremy at biochem.uthscsa.edu
University of Texas Health Science Center
Bioinformatics Core Facility
Phone: (210) 567-2672
More information about the Satlug