S. I. Feldman, D. M. Gay, M. W. Maimone, and N. L. Schryer. A Fortran-to-C Converter. Technical Report No. 149, Bell Laboratories, NJ, 1995.  Home/Search   Document Not in Database   Summary   Related Articles   Check

.... [1] The loops, listed in Table 6, are all chosen because they represent a large fraction of the serial execution time, and they are not statically analyzable [16] by state of the art parallel compiler frameworks such as Polaris [2] All benchmarks have been translated from Fortran to C with f2c [4], and been compiled with gcc 2.95.2 with optimization level O2. Benchmark Subroutine: Loop Name of Serial Ocean FTRVMT: 109 45 Adm RUN: 20, 30, 40, 50, 60, 100 35 Track NLFILT: 300 52 Table 6: Loops chosen for evaluation. The serial execution times of the loops are measured on a Sun ....

S. I. Feldman, D. M. Gay, M. W. Maimone, and N. L. Schryer, \A Fortran-to-C Converter", Technical Report No. 149, AT&T Bell Laboratories, Murray Hill NJ 07974, May 1990.

....to evaluate supercomputers. These benchmarks are written in Fortran 77 and embedded with MPI calls for communication. Since MPI SIM currently supports localization for only C programs, it was necessary to convert the benchmarks to C. We were able to convert four out of the ve benchmarks using f2c[FGMN90] a Fortran to C converter. The speci c con gurations of the benchmarks that were used in the performance study were constrained primarily by their memory and CPU requirements. Four di erent asynchronous algorithms and a synchronous quantum protocol were used to synchronize the parallel program ....

S.I. Feldman, D. M. Gay, Mark W. Maimone, and N.L.Schryer. A fortran-to-c converter. Technical report no. 149, AT&T Bell Laboratories, Murray Hill, NJ, May 1990.

....evaluate supercomputers. These benchmarks are written in Fortran 77 and embedded with MPI calls for communication. Since MPI SIM currently supports localization for only C programs, it was necessary to convert the benchmarks to C. We were able to convert four out of the five benchmarks using f2c[FGMN90] a Fortran to C converter. The specific configurations of the benchmarks that were used in the performance study were constrained primarily by their memory and CPU requirements. Four different asynchronous algorithms and a synchronous quantum protocol were used to synchronize the parallel ....

S.I. Feldman, D. M. Gay, Mark W. Maimone, and N.L.Schryer. A fortran-to-c converter. Technical report no. 149, AT&T Bell Laboratories, Murray Hill, NJ, May 1990.

....hardware latency of 500ns. The NPB benchmarks are written in Fortran 77 with embedded MPI calls for communication. Since MPI SIM currently supports privatization only for C programs, it was necessary to convert the benchmarks to C. We were able to convert four out of the five benchmarks using f2c (Feldman et al. 1990), a Fortran to C converter. The specific configurations of the benchmarks that were used in the performance study were constrained primarily by their memory and CPU requirements. Table 1 summarizes the relevant configuration information for the benchmarks. Each benchmark was executed for three ....

Feldman, S. I., D. M. Gay, Mark W. Maimone, And N. L. Schryer. A Fortran-To-C Converter. Technical Report No. 149, AT&T Bell Laboratories, Murray Hill, NJ, May 1990.

No context found.

S. I. Feldman, D. M. Gay, M. W. Maimone, and N. L. Schryer. A Fortran-to-C Converter. Technical Report No. 149, Bell Laboratories, NJ, 1995.

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