Job_flowのaws-sdkを使った設定例 for Ruby

job_flow.rb

    emr = AWS::EMR.new({
      :access_key_id => AWS_ACCESS_KEY,
      :secret_access_key => AWS_SECRET_KEY,
      :region => "#{AWS_REGION}",
      :emr_endpoint => "#{ENDPOINT_EMR}"
    })
    emr_result = emr.job_flows.create(
      "#{BATCHNAME}", {
        :log_uri => "s3://#{S3BUCKET}/batch_logs",
        :ami_version => 'latest',
        :instances => {
          :ec2_key_name => "#{KEYFILE_ADDR}",
          :instance_count => "#{INSTANCE_COUNT}",
          :master_instance_type => "#{INSTANCE_TYPE}",
          :slave_instance_type => "#{INSTANCE_TYPE}",
          :termination_protected =>  false,
        },
        :steps =>
          [{
            :name => "#{BATCHNAME}",
            :action_on_failure => 'CONTINUE',
            :hadoop_jar_step => {
              :jar => '/home/hadoop/contrib/streaming/hadoop-streaming.jar',
              :args => [
                  "-mapper","ruby s3://#{S3BUCKET}/mapper/mapper.rb",
                  "-reducer","ruby s3://#{S3BUCKET}/reducer/reducer.rb",
                  "-input","s3://#{S3BUCKET}/#{BATCHNAME}/input",
                  "-output","s3://#{S3BUCKET}/#{BATCHNAME}/output",
          ]}}],
    })