lynch@batcomputer.tn.cornell.edu (Tim Lynch) (08/23/89)
I'm posting this for a colleague. Please repond to me via e-mail. We're especially interested in any benchmark results you might have regarding SGI machines: We are in the process of running molecular modelling benchmarks on a number of mini-supercomputers and supercomputers. We would be very interested in receiving benchmark data from individuals who have run molecular modelling (e.g., molecular dynamics, free energy perturbation, etc.) programs on several different machines. We will E-mail a copy of our benchmarks if anyone requests. We could post if interest warrants. We expect our runs to be finished by early October. So far, the computers we are benchmarking include the Convex C210, Silicon Graphics 4D/240 and 4D/280, FPS 500, IBM 3090-600E, and Apollo. E-mail to me if you would be willing to share your benchmark results. Comments on ease of code conversion, or advantages/disadvantages of features on different machines would also be appreciated. Thanks!