[Beowulf] Definition of HPC

Ellis H. Wilson III ellis at cse.psu.edu
Thu Apr 18 10:05:36 PDT 2013


On 04/18/2013 12:57 PM, Nicholas M Glykos wrote:
>
>
>
>>> In the same way you wouldn't allow a general
>>> user to override the safety interlocks of an X-ray generator, you
>>> shouldn't allow root access to the general users of a shared computing
>>> facility.
>>
>> Please describe what the grad student who pioneers new tech for X-ray
>> generators should do, particularly if they are trying to develop new safety
>> interlocks.
>
> They work on their supervisor's own X-ray generator at the basement of
> their department under strict supervision. And not at a synchrotron site
> with 40 people doing their own experiments.
>
>> There needs to be, at the very least, a testbed in universities where students
>> can receive and be able to use root on a limited set of machines.
>
> Fully agree. Students should be able to play and break things, but with
> their group's toy cluster, and not the central cluster of their
> university.

While I would love that, and in an ideal world there would be a totally 
separate, at-scale cluster for me to research on as I alluded to, good 
luck convincing universities to buy such things (individually owned 
X-ray generators or clusters to break).  I've not yet been able to, at 
least.

My entire point is that there is a marginalized sector of research in 
the EDU space because of the current manner in which things are guarded. 
  If you're a theory person, you're golden.  Just run your shitty Matlab 
script on the cluster and moan when the cluster is down.  Feel free to 
forget about reality, or how that cluster works, or where and how your 
script files and data is stored.

Hence why real hardcore computer engineering (where things can and do 
break, and people learn the MOST from those breakages) is done in the 
private sector.

It's really a sad state of affairs.

ellis


More information about the Beowulf mailing list