Move SpeedUpAmerica from using EC2 Instances to Fargate (2SP)

[Kourchenko]

Backstory
The current ECS Instances require a large (8GB) memory allocation in order to process the nightly import data. We recently had to upgrade from 4GB to 8GB in order to successfully process the nightly process for SpeedUpAmerica Gamma and Alpha sites. There are 3 ECS Instances for Gamma at 8GB and 3 ECS Instances for Alpha at 8GB, which doubles the cost for each instance each month, for each of the 6 Instances.

Task
Please investigate and implement moving SpeedUpAmerica ECS Instances from EC2 to Fargate to allow AWS to allocate memory usage as needed.

Expected behavior
Please verify with Matt Sayre after implementing the solution to verify if there is a reduced cost in monthly AWS costs.
Please verify in CloudWatch Logs to verify that the nightly processes are processing successfully.

Desktop (please complete the following information):
None.

Smartphone (please complete the following information):
None.

[mattsayre]

Added 2SP to this ticket.