Parallel Implementation of FEM Solver for Shared Memory Using OpenMP

Table 15

Finite element solver timing, speedup, and efficiency for system of 2099201 unknowns.

Threads	1	2	3	4	5	6	7	8
Time	1671	1266	1031	956	743	706	686	649
Speedup	1.00	1.32	1.62	1.75	2.25	2.37	2.435	2.58
Efficiency	1.00	0.66	0.54	0.44	0.45	0.40	0.348	0.32
Threads	9	10	11	12	13	14	15	16
Time	534	532	490	460	408	377	374	321
Speedup	3.13	3.14	3.41	3.64	4.09	4.44	4.468	5.21
Efficiency	0.35	0.31	0.31	0.30	0.32	0.32	0.298	0.33