L. J. Podrazik and H. E. Conn. Parallel recurrence solvers for vector and SIMD supercomputers. In Proceedings International Conference on Parallel Processing, pages III.88--95, Aug. 1992.  Home/Search   Document Not in Database   Summary   Related Articles   Check

....(CREW) parallel random access machine (PRAM) but uses a different formulation and organizes the computation differently. When mapped onto real machines, the Regular Schedule will perform better since it has better utilization of locality of reference. The algorithm for first order LR in [7] reduces to the regular schedule only for first order LR, but the blocked first order formulation for mth order LR (m 2) described in [7] differs from the regular schedule and has longer execution time. The Regular Schedule is not time optimal under the CREW PRAM model both its execution time ....

....onto real machines, the Regular Schedule will perform better since it has better utilization of locality of reference. The algorithm for first order LR in [7] reduces to the regular schedule only for first order LR, but the blocked first order formulation for mth order LR (m 2) described in [7] differs from the regular schedule and has longer execution time. The Regular Schedule is not time optimal under the CREW PRAM model both its execution time and program space efficiency are inferior to many other schedules derived using Harmonic Schedule [33] However, its regular organization ....

H. Conn and L. Podrazik, "Parallel recurrence solvers for vector and SIMD supercomputers", Proc. 1992 International Conf. on Parallel Processing, pp. 88-95, Vol. 3, Aug. 17-21, 1992.

....model, but uses a different formulation and does not organize then computation in periods. When mapped onto real machines, the Regular Schedule will perform better since its period organization has significantly better utilization of locality of reference[17] The algorithm for first order LR in [5] reduces to the regular schedule only for first order LR, but the blocked first order formulation for mth order LR (m 2) described in [5] differs from the regular schedule and has longer execution time. The Regular Schedule is not time optimal under the CREW PRAM model both its execution time ....

.... will perform better since its period organization has significantly better utilization of locality of reference[17] The algorithm for first order LR in [5] reduces to the regular schedule only for first order LR, but the blocked first order formulation for mth order LR (m 2) described in [5] differs from the regular schedule and has longer execution time. The Regular Schedule is not time optimal under the CREW PRAM model both its execution time and program space efficiency are inferior to many other schedules derived using Harmonic Schedule [16] However, its regular organization ....

H. Conn and L. Podrazik, "Parallel recurrence solvers for vector and SIMD supercomputers", Proc. 1992 International Conf. on Parallel Processing, pp. 88-95, Vol. 3, Aug. 17-21, 1992.

.... recurrences [17, 10] Researchers have been studying parallel and vector algorithms to solve linear recurrences since the 1960 s [16, 3, 17, 29, 5, 15, 27, 21, 10, 33, 11, 22] and considerable effort has gone into producing fast implementations of these algorithms on parallel and vector machines [23, 20, 31, 30, 12, 24, 28, 26]. Some supercomputer manufacturers have considered the solution of linear recurrences important enough to warrant the addition of special hardware to execute recurrences efficiently [32] Three major classes of algorithms have emerged out of the research on algorithms to solve linear recurrences: ....

L. J. Podrazik and H. E. Conn. Parallel recurrence solvers for vector and SIMD supercomputers. In Proceedings International Conference on Parallel Processing, pages III.88--95, Aug. 1992.

....parallel random access machine(PRAM) model, but uses a different formulation and organizes the computation differently. When mapped onto real machines, the Regular Schedule will perform better since it has better utilization of locality of reference. The algorithm for first order LR in [7] reduces to the regular schedule only for first order LR, but the blocked first order formulation for mth order LR(m 2) described in [7] differs from the regular schedule and has longer execution time. The Regular Schedule is not time optimal under a CREW PRAM model, and in fact both its ....

....onto real machines, the Regular Schedule will perform better since it has better utilization of locality of reference. The algorithm for first order LR in [7] reduces to the regular schedule only for first order LR, but the blocked first order formulation for mth order LR(m 2) described in [7] differs from the regular schedule and has longer execution time. The Regular Schedule is not time optimal under a CREW PRAM model, and in fact both its execution time and program space efficiency are inferior to many other schedules derived using Harmonic Scheduling[27] However, its regular ....

H. Conn and L. Podrazik, "Parallel recurrence solvers for vector and SIMD supercomputers", Proc. 1992 International Conf. on Parallel Processing, pp. 88-95, Vol. 3, Aug. 17-21, 1992.

No context found.

L. J. Podrazik and H. E. Conn. Parallel recurrence solvers for vector and SIMD supercomputers. In Proceedings International Conference on Parallel Processing, pages III.88--95, Aug. 1992.

Online articles have much greater impact   More about CiteSeer.IST   Add search form to your site   Submit documents   Feedback  

CiteSeer.IST - Copyright Penn State and NEC