The lower-upper symmetric Gauss-Seidel method is modified for the simu
lation of viscous flows on massively parallel computers. The resulting
diagonal data-parallel lower-upper relaxation (DP-LUR) method is show
n to have good convergence properties on many problems. However, the c
onvergence rate decreases on the high cell aspect ratio grids required
to simulate high Reynolds number hows. Therefore, the diagonal approx
imation is relaxed, and a full matrix version of the DP-LUR method is
derived. The full matrix method retains the data-parallel properties o
f the original and reduces the sensitivity of the convergence rate to
the aspect ratio of the computational grid. Both methods are implement
ed on the Thinking Machines CM-5, and a large fraction of the peak the
oretical performance of the machine is obtained, The low memory use an
d high parallel efficiency of the methods make them attractive for lar
ge-scale simulation of viscous flows.