How do you keep clusters running....
James.P.Lux at jpl.nasa.gov
Wed Apr 3 17:15:58 PST 2002
You know, fans shouldn't fail...... There are fans available with 50,000
hour MTBFs.. sure, they cost a bit more than $5, but, given the cost of the
time to replace them (especially if you cook something), it might be a good
You might cannibalize one of your failed fans to look for the number and
kind of bearings. I have heard that some "ball bearing" fans actually have
sleeve bearings, a sure recipe for short life. It's not unheard of to have
some fans that are mislabelled. Bear in mind that most fans have two
bearings (one on each end of the shaft) and it is entirely possible to
build a fan with one sleeve and one ball bearing.
At 03:04 PM 4/3/2002 -0600, Cris Rhea wrote:
>What are folks doing about keeping hardware running on large clusters?
>Right now, I'm running 10 Racksaver RS-1200's (for a total of 20 nodes)...
>Sure seems like every week or two, I notice dead fans (each RS-1200
>has 6 case fans in addition to the 2 CPU fans and 2 power supply fans).
Spacecraft Telecommunications Equipment Section
Jet Propulsion Laboratory
4800 Oak Grove Road, Mail Stop 161-213
Pasadena CA 91109
818/354-2075, fax 818/393-6875
More information about the Beowulf