I'm not at all familiar with Linux clustering, just the usual
stand-alone desktop / server. Any ideas, pointers, etc. appreciated.
First part of any cluster is deciding what application you are going to
run on it. That will determine how you need to present the hardware to it.
If your friend just wants to consolidate the compute power and then do
virtualized machines on top, that is different than say running some heavy
analytically software like protein folding.
This is right - you can't really gang a bunch of machines together to
act like one Desktop Linux computer. What you can do, is either spread
the applications across machines, or break up the data to be processed
into chunks for processing. If you just wanna see how many MIPS you can
put out and watch the blinkenlights, that's fun too.
That's a big topic, and highly dependent upon what kind of stuff you're
doing.
For the blinkenlights, there are projects like Folding@Home and
distributed.net are fun - I've been contributing to distributed.net for
14 years:
http://stats.distributed.net/participant/psummary.php?project_id=27&id=141698
I work on Internet applications, and frequently I see the end-users'
clicks simply go into a queue, which is handled by a set of machines
which process items from the queue. With modern cloud computing and some
handy open-source projects, this is relatively easy to do.
Uh, I have a ton of wifi/geodata that needs processing if you've got
some idle computing.
--
Drew from Zhrodague
lolcat divinator
[email protected]
--
You received this message because you are subscribed to the Google Groups
"NLUG" group.
To post to this group, send email to [email protected]
To unsubscribe from this group, send email to
[email protected]
For more options, visit this group at
http://groups.google.com/group/nlug-talk?hl=en
