Issue No. 12 - December (1994 vol. 5)
DOI Bookmark: http://doi.ieeecomputersociety.org/10.1109/71.334900
<p>With the advent of new routing methods, the distance that a message is sent isbecoming relatively less and less important. Thus, assuming no link contention,permutation seems to be an efficient collective communication primitive. In this paper, we present several algorithms for decomposing all-to-many personalized communication into a set of disjoint partial permutations. We discuss several algorithms and study theireffectiveness from the view of static scheduling as well as run-time scheduling. Anapproximate analysis shows that with n processors, and assuming that every processorsends and receives d messages to random destinations, our algorithm can perform thescheduling in O(dn In d) time, on average, and can use an expected number of d+log dpartial permutations to carry out the communication. We present experimental results ofour algorithms on the CM-5.</p>
Index Termsscheduling; multiprocessor interconnection networks; performance evaluation; run-time algorithms; static algorithms; all-to-many personalized communication; permutation networks; run-time scheduling; CM-5
S. Ranka, J. Wang and G. Fox, "Static and Run-Time Algorithms for All-to-Many Personalized Communication on Permutation Networks," in IEEE Transactions on Parallel & Distributed Systems, vol. 5, no. , pp. 1266-1274, 1994.