[Beowulf] Docker vs KVM paper by IBM

Ellis H. Wilson III ellis at cse.psu.edu
Wed Jan 28 12:47:38 PST 2015


On 01/28/2015 03:39 PM, Jason Riedy wrote:
> But the growing quantity of crap
> software from a system level that produces useful science results
> can be supported without the current levels of pain.

This cannot be overstated.  Even (or perhaps, ESPECIALLY) in CS there is 
an egregious amount of absolutely broken programs out there that only 
compile on such and such a set of compilers/libraries/etc.  The use of 
these becomes habit amongst research groups and all the sudden IT shops 
can't upgrade above CENTOS 5 because the advisor with the big grants 
doesn't want to break usability with some piece of garbage code-slop 
that some PhD many moons ago put together thinking he or she would be 
the only one ever using it.  Then other researchers who actually want to 
use toolchains newer than 10 years old are on the other side of that 
drawing and quartering of the IT staff.

Best,

ellis


More information about the Beowulf mailing list