Search code examples
hadoopelastic-map-reduce

Easiest way to get started with Hadoop

I was looking for the simplest way(s) to submit a MapReduce job.

I am looking for a platform similar in complexity (or simplicity) such a Heroku (is to Ruby) or picloud.com is to map.

The idea is where a beginner can submit a MapReduce job without having to deal with the complexities of setting up a Hadoop Cluster. Elastic Map Reduce comes close but the start up times to set up the instances is a big turn off.

So basically I was looking for someone who had done the picloud equivalent of MapReduce.

I also read the following answers but didn't find what I was looking for: this one and this one

Thanks.

Solution

  • Two good starter resources:

    Using Amazon Elastic Mapreduce: http://aws.amazon.com/aws-training/elasticmapreduce/

    Using the Cloudera CDH solution (more involved and relies on Hive instead of actual MapReduce jobs): http://www.infoq.com/articles/BigDataBlueprint

    EDIT: If you aren't fond of Elastic MapReduce, you might want to give Sears MetaScale a try (I haven't tried it): http://www.metascale.com/services/big-data