In this paper, we present performance results from several parallel benchma
rks and applications on two large Linux clusters at Sandia National Laborat
ories. We compare the results on the Linux clusters to performance obtained
on a traditional distributed-memory massively parallel processing machine,
the Intel TeraFLOPS. We discuss the characteristics of these machines that
influence the performance results and identify the key components of the s
ystem that are important to allow for scalability of commodity-based PC clu
sters to hundreds and possibly thousands of processors. (C) 2001 Academic P
ress.