Please enable JavaScript.
Coggle requires JavaScript to display documents.
Elastic Map Reduce(Hadoop and Apache Spark) H= hugh and Spark with ligh…
Elastic Map Reduce(Hadoop and Apache Spark) H= hugh and Spark with ligh
Map Reduce
function
to paralllerize work load
to process hugh dataset
Provide framework for big data processing
such as
Hadoop
Apache Spark
2 components
Map
Split the workload to into independent chunk & process it
Reduce
Combine the multiple chunk into single dataset back
Elastic Map Reduce
Cluster of Ec2 which consists
1 Master node and Task node (can use spot instance)
funciton
can be used to move data to redshift and S3 via AWS backbone
For business intelligent and analytic
can be launched by cloud formation and Lamda
can choose which bigdata framework to install: Apache Spark or Hadoop