Benchmark Results

To test the results of these changes the bench_64 test case was used. The table below gives the runtimes on a range of processor counts on HECToR and the resulting speedup:

Table 3: Comparison of bench_64 runtime before and after rs2pw optimisation

Cores	16	32	64	128	256	512
Before(s)	952	541	318	268	217	264
After(s)	938	519	296	247	190	235
Speedup(%)	2	4	7	9	14	12