You have a high performance compute application and you need to minimize network latency between EC2
instances as much as possible. What can you do to achieve this?
Use Elastic Load Balancing to load balance traffic between availability zones
Create a CloudFront distribution and to cache objects from an S3 bucket at Edge Locations.
Create a placement group within an Availability Zone and place the EC2 instances within that placement
Deploy your EC2 instances within the same region, but in different subnets and different availability zones
so as to maximize redundancy.