| Subject: Re: Running SETI CLI on Linux based Cobalt web server |
| From: baskitcaise |
| Date: 05/01/2004, 09:30 |
Thomas wrote:
Hi,
I have been given a half dozen Cobalt RaQ4i's (AMD 450Mhz CPU/128Mb
memory/20Gb HDD) which for the time being, I am not going to be using for
their original task.
So as they are sitting around doing nothing, I thought about SETI and
setting up a small "farm".
They do not have direct web access, so I'd like to run the CLI version of
SETI and to then be able to transfer the completed files from each RaQ to a
separate PC that is connected via broadband to the web.
It would be nice to be able to cache some work units on each RaQ, so I can
get them crunching for a while, instead of having to keep transferring
completed WU's one at a time from the RaQ's to the PC that has a 'Net
connection.
Anyone got any clues/pointers about achieving this (and assuming it's
possible)?
regards,
Thomas
(PS Email address is spam-trapped - pls reply to ng).
Have a look in the Add Ons link on the seti front page, there are some
commandline buffering progs on there that should do the job.
I don`t use a c/line buffer but the other week when seti@home had the database
crash I could not connect with one of my boxes running the GUI Ksetispy, so
all I did was copy the contents of the cache dir to a floppy and transfer
over to a machine that could connect and use that to send the results off,
worked like a charm so should not be any probs doing it with the c/line.
HTH
--
Mark
Twixt hill and high water.
N.Wales, UK.
Email is spam trap try baskitcaise at gmx dot co dot uk