Just want to provide a note to distinguish different kinds of scaling here: First case: assign each frame to a machine (processor). Hence problem constrained as we did in homework 1. second case: given a fixed amount of time, try to create higher fidelity representation. Therefore time constrained. third case and last case: usually in order to deal with a lot of data (HD simulation in physics related fields), you can't handle the data set in one processor. Therefore assign the same amount of data to each processor and parallelize the simulation. Therefore it's memory constrained.

This comment was marked helpful 0 times.