Homework 5 -- Estimated Cloud Costs

You have 300 TB of data which can be processed in parallel (i.e., the data may be sliced into any sized chunk)
You have done some benchmarking where it 10 hours to process 2GB of data on a single core:
- It takes 8GB of RAM when those 2GB of data are loaded in memory and being processed
- Compute time scales linearly with the amount of input data (e.g., 20GB of data takes 10x longer than 2GB of data)
- Memory usage scales linearly with the amount of input data (e.g., 20GB of data take 10x more RAM than 2GB of data)
How much will it cost to process all the data in 1 hour? in 1 day?
The final output is 2TB. How much does cost to transfer the data to AWS and transfer the results back?
What is the difference between using different Amazon facilities? Compare US West Coast, Tokyo, and San Paulo.
Make sure to show your work and system configuration!
Note: This is an individual assignment (not a team project)
Due 2h before class (7:30am) on Sept. 21!