OptimoMapReducer helps you to optimize the cost of your big data platforms such as Amazon EMR using Spot Instances. Instead of using On-Demand instances, you can reliably use Spot Instances for task nodes to reduce the cost of your EMR clusters.
# Optimization Approch
We provide 2 ways to optimize cost of your EMR clusters.
# 1. Optimize existing EMR clusters.
Here, we use the Green-Blue deployment model to optimize the existing EMR cluster. In this case, we launch a parallel set of task nodes with the same set of configurations. Once all task nodes from the parallel set come into running mode, we phaseout existing task nodes. It ensures the smooth transition of the EMR cluster from existing OnDemand task nodes to Spot Instance-based newly launched task nodes with zero downtime.
# 2. Launch and Optimize copy of existing EMR cluster.
Here, we clone configurations from a given EMR cluster and launch a new EMR cluster with Spot Instance-based task nodes.
# How to configure OptimoMapReducer?
It just takes 2 steps to configure your first OptimoMapReducer.
Step-1: Configure Basic Details.
Configure basic configuration parameters such as name, region, optimization approch etc.
Step-2: Configure Basic Details.
Configure EMR source details, network details, instance types, target capacity and autoscaling policies.
Done, you're all set to launch your first OptimoMapReducer!!!