[Beowulf] IEEE 1588 (PTP) - a better cluster clock?

[Jim Lux]

At 07:15 AM 7/24/2007, Ashley Pittman wrote:
>On Fri, 2007-07-20 at 16:41 -0700, Jim Lux wrote:
> >
> > There are tons of reasons why one might want more precision than NTP
> > can provide, leaving aside the obvious applications for things like
> > controlling test equipment.
>
>But none of the ones you list are more than vaguely related to HPC.

Highly parallelized real time signal processing?  Seems like a 
classic HPC application to me, especially if you want to change the 
processing over time, as opposed to dedicating hardware.  Or, if you 
want to leverage commodity computing widgets, trading more of cheaper 
hardware that has less performance per unit instead of fewer 
expensive customized boxes.  To that end, standardized ways of 
providing high quality synchronization is important, because it 
drives down the cost and provides a larger pool of people who know 
how to use it (actually, another cost factor)

>In short, I'd use it if it was packaged with the distro and just worked
>out the box (without using multicast preferably) but it's absence isn't
>enough to cause me problems.


And, of course, that IS the beauty of commodity OSes and hardware.. 
If you don't need it, you don't have to use it.


James Lux, P.E.
Spacecraft Radio Frequency Subsystems Group
Flight Communications Systems Section
Jet Propulsion Laboratory, Mail Stop 161-213
4800 Oak Grove Drive
Pasadena CA 91109
tel: (818)354-2075
fax: (818)393-6875