What are some good Computing Application Architecture tutorials?

Question

Mathematics

Hello. Please advise good literature or articles on the organization of the architecture of an application that performs heavy mathematical calculations, in several stages, with parallelization. I know that, for example, pipelined calculations, etc. can be used. Questions arise about how to organize data flows in such an application, how to maximize processor load, reduce core idle time, and quickly stop calculations. Do you know where this issue is well covered?

Answer 1

I don't know where it's lit.
I used thread-safe queues to pass data between computational threads. The scheme is working. If you correctly separate the flows, then all synchronization can occur only with the help of queues.
For pluses, you can use Intel TBB, just such an approach is used there.

Answer 2

OpenMP
parallel computing

Answer 3

CUDA examples on MatLab.