Hi, I'm interested in looking at workload with highly variable mapper/reducer execution time. Which mahout workload do you recommend to run? I see clustering, classification, recommendation, etc. Also, could you recommend what dataset to use?
Thanks. -Brian
