
192 High Performance Programming for Soft Computing
8.5.1 Analysis of the parallel implementation of ACO
Figure 8.12 shows a graph representing the execution time of a PCB design
for a 5V DC Power Supply, consisting of 25 holes; the optimized G-code for
drilling operations is generated by using the Parallel ACO.
We can observe that when the number of processors is increased, the
computing time can be reduced because each processor has fewer ants
assigned, but due to the overhead communication, which increases the
total execution time, the speedup of our algorithm cannot increase linearly
with the increasing of the number of processors exactly ...