A multiprocessor with 256 cpus is organised as a 16*16 grid what is the worst case delay in hops that a messa?

Question

ge might have to take.releated to distributed computing

Answer 1

Worst case: from #1 (top left) to #256 (bottom right).
Message goes "down" 15 times, and "right" 15 more, for a total of 30.
General case in NxM grid = (N-1)x(M-1)

The solution is to "wrap" the grid, ie. connect bottom row to top row, and left column to right column.

Answer 2

i'll't communicate for multiprocessors, yet doesn't that stick with an identical regulation (as TCP/IP) for Time To stay (TTL?) if so, examine the TTL threshhold. it really is the optimal type of hops, or on your case, the max delay.