This is a practical example of how the characteristics of a parallel computer determine program optimization. It is based on a real-world physics endeavor. The parallel computers used cost hundreds of thousands of dollars and even small optimizations of 10-20% save thousands of dollars!
This post discusses communication routing strategies for data on massively parallel systems and supercomputers. Optimizations like these one can save thousands of dollars in real-world projects.
If you could go back in time, what would you do different?
These are computer projects I did while or was in high school and while I was a student. And what I learned from them.
Germans complain that service is awful in Germany. I agree with them. However, this also means that there is an outstanding opportunity to distinguish yourself from your competition and make an impression on your customers and/or your clients.