Solutions to partial differential equations are required in many engin
eering applications. The multigrid method is an iterative technique fo
r speeding up the solution of these equations. The authors describe a
parallel implementation of the multigrid method on the connection mach
ine CM-5 architecture. An analytic model is presented for estimating t
he computation and communication times of the multigrid algorithm. The
times predicted by the analytic model are within 5% of the results ob
tained from CM-5. Results demonstrate that the communication overhead
incurred by the parallel multigrid algorithm is relatively small compa
red to the computation time. Consequently, implementations of the mult
igrid algorithm on the CM-5 easily achieve processor efficiencies near
100%.