The Method Of Parallel Recognition And Parallel Optimization Based On Data Dependence With Sparse Matrix

摘要

Abstract: for application programs in scientific and technological fields have grown increasingly large and complex, it is becoming more difficult to parallelize these programs by hand using message-passing libraries. To reduce this difficulty, we are researching the compilation technology for serial program automatic parallelization. In this paper, the author puts forward a kind of parallel recognition algorithm in parallelization compiler with sparse matrix to reduce memory consumption and time complexity. In the algorithm the author adopts the idea of the medium grain parallel.