AWS hadoop runs never complete

[blueeyedJ]

Hi,
I'm very new to this and am having trouble with hadoop and aws.

Running 0.10.1.
Cmd: /pkb.py --cloud=AWS --benchmarks=hadoop_terasort --machine_type=t1.micro --terasort_num_rows=1000

Is there a minimum machine_type? I would think 1000 rows would go very quickly but it never completed, even after 3 hours of running. Ideas/Suggestions?
Thanks,

[cmccoy]

There should be a run-specific verbose log file in /tmp/perfkitbenchmarker/run_<run_uri>/pkb.log - could you paste the last few commands run?

I wouldn't be surprised if a t1.micro didn't have enough memory to run Hadoop, but I haven't tried.

[blueeyedJ]

yeah that was it. Might consider adding a note to the pkb help output. Something along the lines of "resource intensive workload, may not work well with micro tiers".

[cmccoy]

Glad to hear it. I opened #144 to track updating the help.