Where to run Elastic Map Reduce CLI from or use an alternative to EMR CLI?

Question

Where do I let EMR CLI run as a recommended case? From my local Linux workstation or from a AWS Virtual Server?

Ar there (better) alternatives to EMR CLI, in case I want to programmatively access my clusters and perform Map Reduce jobs?

What do you mean "Is it possible that I need to run elastic-mapreduce directly from S3"? Can you provide details how do you launch cluster? — mr0re1, Jun 21 '14 at 01:42
I am running the cluste(s)r from the AWS/S3 Web interfaces, the ruby client itself is running on my Ubuntu 14.04. I followed the official docs and the readme file. I will try tonight without proxy, I hope it was the root cause. If not I still will need some guidance. I need to get this going quickly. — Stephan Kristyn, Jun 21 '14 at 20:03
Now you question seems non-useful. Can explain what do you try to do and what kind of problems did you meet? BTW: EMR Ruby CLI provide arguments to use proxy. — mr0re1, Jun 21 '14 at 20:10
Hi, I did set the Proxy in the config files.I did alter the question now altogether to be more simple. — Stephan Kristyn, Jun 22 '14 at 21:06

score 0 · Answer 1 · answered Jun 23 '14 at 04:43

0

Usually there is no reason to run EMR Ruby CLI on AWS cluster instead of your local machine.

Currently (as I know) there is no CLI alternative for EMR Ruby CLI. The AWS CLI doesn't support full stack of EMR API.

If you want to automate EMR usage (some repeatedly use cases for example) you can try to use AWS SDK for Java, Python and other languages.

answered Jun 23 '14 at 04:43

mr0re1

Do you mean Python Boto? https://github.com/boto/boto – Stephan Kristyn Jun 23 '14 at 08:50
Yes, boto for example, but you shouldn't use it if your use case simple unrepeatable job. – mr0re1 Jun 23 '14 at 17:25
Use case is Terrabytes of data in my case. – Stephan Kristyn Jun 24 '14 at 21:29

score 0 · Accepted Answer · answered Jun 24 '14 at 21:26

0

answered Jun 24 '14 at 21:26

Stephan Kristyn

2 Answers2