The Eeective I/o Bandwidth Benchmark (b Ee Io)

| The eeective I/O bandwidth benchmark (b ee io) covers two goals: (1) to achieve a characteristic average number for the I/O bandwidth achievable with parallel MPI-I/O applications, and (2) to get detailed information about several access patterns and buuer lengths. The benchmark examines \\rst write", \rewrite" and \read" access, strided (individual and shared pointers) and segmented collective patterns on one le per application and non-collective access to one le per process. The number of parallel access-ing processes is also varied and wellformed I/O is compared with non-wellformed. On systems, meeting the rule that the total memory can be written to disk in 10 minutes, the benchmark should not need more than 15 minutes for a rst pass of all patterns. The benchmark is designed analogously to the eeective bandwidth benchmark for message passing (b ee) that characterizes the message passing capabilities of a system in a few minutes. First results of the b ee io benchmark are given for IBM SP and Cray T3E systems and compared with existing benchmarks based on parallel Posix-I/O.