[Beowulf] Docker vs KVM paper by IBM

Olli-Pekka Lehto olli-pekka.lehto at csc.fi
Wed Jan 28 12:47:39 PST 2015

On 28 Jan 2015, at 22:05, Ellis H. Wilson III <ellis at cse.psu.edu> wrote:

> So, the obvious answer here is, provide your "standard operating environments" in the form of containerized/VM/whatever images quartiles 1 and 2 can use, and allow quartiles 3 and 4 to spin up their own. Multiple environments means quartile 2 can probably just try their program A on environments X, Y, and Z, and find one that "just works." This reduces their time futzing with compilers or fixing other researcher's crappy code that breaks on GCC > 4.x.  Quartile 3 can spin up their own absolutely crap environment and think their L33t and not screw over their fellow researchers.  Quartiles 1 and 4 are basically untouched, since they were fine before as now.
> Everybody wins, probably most of all the IT department.
> Best,
> ellis

I see using Docker/KVM etc. too eagerly as a universal “out” for problem situations a slippery slope: 

Instead of trying to figure out what the problem is and how to improve the environment/documentation/software stack, people are pointed to “roll their own”. In the long term this could result in the stagnation of the in-house software stack.

I envision, in the worst case, a dystopia where user groups have re-invented the wheel with a custom stack with varying degrees of efficiency, probably on average much poorer than the highly optimized and user friendly in-house stack that could’ve been. 

That said, they definitely have a place in the ecosystem but I think it should not take too much away from trying to understand and educate even the quartile 3 users and develop a user-friendly and efficient software stack that could also cater to them. 


More information about the Beowulf mailing list