A new memory-saving technique to map system of affine recurrenceequations (SARE) onto distributed memory systems

摘要

In this work we present a procedure for automatic parallel codegeneration in the case of algorithms described through Set of AffineRecurrence Equations (SARE); starting from the original SARE descriptionin an N-dimensional iteration space, the algorithm is converted into aparallel code for an m-dimensional distributed memory parallel machine(m