Placement Groups

• Amazon provides a way of choosing physical servers by providing a flexibility while ec2 machines are under creation, This is clled as Placement Group.
• There are three popular strategies with in AZ
  • Cluster:
    • Used for HPC (High Performance Compute)
    • Enables low latency network performance for node-node communications
  • Spread:
    • Used to be tolerant towards related hardware failures
    • For Smaller Deployments
  • Partition:
    • Creating Partition Groups and launching machines

Scenario:

• Consider a Medical Record System (MRS) which has the following architecture
• This MRS has to run in all the states in us in QT-Chain of hospitals
• How do deploy this application on AWS
• We have decided to deploy the application in two regions (oregon) and ohio
• Now how to deploy this application
  • Application:
    • Our deployment strategy is to create an ASG with Application AMI
    • This autoscaling group is linked with load balancer
    • Load balancer, Autoscaling group, AMI are region specific
  • Database:
    • Databases have to be in sync
• My Solution is as shown below
• If you want to realize this use mrs software from here