| A. Reeves. The Paragon programming paradigm and distributed memory compilers. Technical Report EE-CEG-90-7, Cornell University Computer Engineering Group, Ithaca, NY, June 1990. |
....and Banerjee [GB90] propose a constraint based approach to automatically calculate suitable data decompositions. They support BLOCK and CYCLIC distributions, but do not specify alignment. Instead of standard BLOCK distributions, Superb [ZBG88, Ger89, Ger90] Suspense [RW88] and Paragon [CR89, Ree90] support arbitrary user specified contiguous rectangular distributions. Superb also originated the overlap concept as a means to both specify and store nonlocal data accesses. Wolfe [Wol89, Wol90] describes transformations such as loop rotation for programs with BLOCK distributions. Callahan ....
A. Reeves. The Paragon programming paradigm and distributed memory compilers. Technical Report EE-CEG-90-7, Cornell University Computer Engineering Group, Ithaca, NY, June 1990.
....boundaries including inheriting the distribution of the actual arguments. Distribution inquiry functions facilitate the writing of library functions which are optimal for multiple incoming distributions. High Performance Fortran effort has been based on the above and other related projects [8, 27, 38, 48, 58, 59, 60]. In the next few sub sections we provide, short introduction to HPF concentrating on the features which are critical to parallel performance. 4.1 HPF Overview High Performance Fortran z is a set of extensions for Fortran 90 designed to allow specification of data parallel algorithms. The ....
....initial implementations in the near future. However, several research projects have built prototype compilers for HPF like languages. This includes the Kali compiler [33] the SUPERB project [67] on which the Vienna Fortran compiler is based, the Fortran D compiler [29] and several other efforts [8, 24, 27, 32, 38, 48, 58, 59, 60] that have contributed to the overall goal of compiling global name space programs for distributed memory SIMD and MIMD machines. pC is just one example of a number of efforts to add parallelism to C . While pC has been ported to a wide variety of machines including the TMC CM 5, Intel ....
A. P. Reeves and C. M. Chase. The Paragon programming paradigm and distributed memory multicomputers. In Compilers and Runtime Software for Scalable Multiprocessors, J. Saltz and P. Mehrotra Editors, Amsterdam, The Netherlands, To appear 1991. Elsevier.
....back on the programmer (explicit imperative programming) and if a mistake is made, the resulting program may be a nondeterminate nightmare. Despite their difficulties, there have been numerous attempts at automatically parallelizing imperative languages [FHK 90, GB90, HA90, IFKF90, ZBG86, Ree90, RA90] To the extent that these compilers have been able to exploit a limited amount of parallelism in certain types of numerical applications, they have been successful. The difficulty in getting imperative programming languages to execute efficiently on parallel architectures without ....
A. Reeves. The Paragon programming paradigm and distributed memory compilers. Technical Report EE-CEG-90-7, Cornell University Computer Engineering Group, Ithaca, NY, June 1990.
....for reductions. Dino programs are deterministic unless special asynchronous distributed arrays are used. As with CM Fortran, Dino programs generate multiple processes per physical processor when large numbers of virtual processors are declared in the environment. 6.3. 4 Paragon Paragon [CR89, Ree90] is a programming environment targeted at supporting SIMD programs on MIMD distributed memory machines. It provides both language extensions and run time support for task management and load balancing. Data distribution in Paragon may either be performed by the user or the system. Parallel ....
A. Reeves. The Paragon programming paradigm and distributed memory compilers. Technical Report EECEG -90-7, Cornell University Computer Engineering Group, Ithaca, NY, June 1990.
....Exec. Time Bock Time, k lock step synch. min barrier synch. binary synch. Figure 11: Performance Comparison 5 Related Work A large number of data parallel languages have been designed including C [RS87] CM Fortran [Lab89a] DINO [RSW91] High Performance Fortran [For92] SUPERB [ZBG86] Paragon [Ree90] 5 The lower bound of 2 Cn represents the implementation where each processor sends a ready signal to a control processor which, in turn, informs all processors of arrival at the barrier. The upper bound of 6 Cn assumes p = 64 processors and a log p Cn overhead. 6 Percentage improvement ....
A. Reeves. The Paragon programming paradigm and distributed memory compilers. Report EE-CEG-90-7, Cornell University Computer Engineering Group, Ithaca NY, June 1990.
....descriptions, but to our knowledge no performance results have been reported from any implementation. There are no obvious implementation difficulties, however. A slightly different idea was used in the Paragon compiler, which allowed partitioning of 2 D arrays by arbitrary sized rectangles [78]. Both the one dimensional and multidimensional versions have many of the advantages of simple BLOCK distribution for nearest neighbor communication and are useful for solving load balancing problems. In addition, the addressing formulas are fairly simple; the most complex operation is looking up ....
A. Reeves and C. Chase. The Paragon programming paradigm and distributed-memory multicomputers. In J. Saltz and P. Mehrotra, editors, Languages, Compilers, and Run-Time Environments for Distributed Memory Machines. North-Holland, Amsterdam, The Netherlands, 1992.
No context found.
A. P. Reeves and C. M. Chase. The Paragon programming paradigm and distributed memory multicomputers. In Compilers and Runtime Software for Scalable Multiprocessors, J. Saltz and P. Mehrotra Editors, Amsterdam, The Netherlands, Elsevier, 1991.
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