Balanced Loop Partitioning Using GTS

Graph Traverse Scheduling is a loop partitioning method for shared memory multiprocessors that achieves minimum execution time of the parallel code generated assuming that a sufficient number of processors are available and synchronization cost is negligible. The method considers the set of statements in the loop body in the partitioning process.

[1]  Constantine D. Polychronopoulos,et al.  Parallel programming and compilers , 1988 .

[2]  Jordi Torres,et al.  GTS: parallelization and vectorization of tight recurrences , 1989, Proceedings of the 1989 ACM/IEEE Conference on Supercomputing (Supercomputing '89).

[3]  Jih-Kwon Peir,et al.  Minimum Distance: A Method for Partitioning Recurrences for Multiprocessors , 1989, IEEE Trans. Computers.

[4]  Weijia Shang,et al.  Independent Partitioning of Algorithms With Uniform Data Dependencies , 1988, International Conference on Parallel Processing.

[5]  Ron Cytron,et al.  Doacross: Beyond Vectorization for Multiprocessors , 1986, ICPP.

[6]  Weijia Shang,et al.  Independent Partitioning of Algorithms with Uniform Dependencies , 1992, IEEE Trans. Computers.