[OpenIndiana-discuss] hadoop+jcuda

[jason matthews]

Hello world,

has anyone tried running hadoop+jcuda for gpu acceleration on their 
hadoop cluster on any illumos derived OS? I am probably one of like 
eight people running such a cluster at scale, but it never hurts to ask.

I am interested to know if it works, meaning all the drive train linkage 
is there, before i start putting any time into the idea.

  * if does work for you, how would you characterize the stability?
  * how would characterize any performance boost for your work load?
  * how would you characterize your workload?
  * are there any use non-obvious use cases where the product fails to
    provide a performance increase?
  * what else can you tell me about your experience?


thanks in advance,
j.