Hi,
I'm a bit late to this thread but anyways...

One other valuable profiling tool is Ganglia which you can install on
your cluster using a boostrap action

elastic-mapreduce --create --alive --instance-type m1.xlarge --num-instances 5 \
--bootstrap-action s3://elasticmapreduce/bootstrap-actions/install-ganglia

This tracks all the system level stuff you'd expect (cpu/ram/network
etc) as well as a bunch of hadoop level metrics

For more details see here...
http://docs.amazonwebservices.com/ElasticMapReduce/latest/DeveloperGuide/index.html?Ganglia_reports.html

Cheers,
Mat
(SDE on EMR)


On 14 April 2011 03:18, Thomas Rewig <[email protected]> wrote:
>  Hello
> right now I'm testing Mahout (taste) Jobs on AWS EMR.
> I wonder if anyone does have any experience with the best cluster size and
> the best EC2 instances. Are there any best practices for mahout (taste)
> jobs?

Reply via email to