An automatic program translator for distributed memory MIMD machines
نویسنده
چکیده
. . . . . . . . . . . . . . . . . . . . . . . . viii Acknowledgements . . . . . . . . . . . . . . . . . . . . . x Chapter 1 Introduction . . . . . . . . . . . . . . . . . . . 1 1.1 The need for parallelism . . . . . . . . . . . . . . . . . 1 1.2 Types of parallel architecture . . . . . . . . . . . . . . . 2 1.3 An alternative to message-passing languages . . . . . . . . . . 7 1.4 Automatic data distribution . . . . . . . . . . . . . . . . 10 1.5 Research goal . . . . . . . . . . . . . . . . . . . . 12 1.6 An outline of the thesis . . . . . . . . . . . . . . . . . 14 Chapter 2 An Overview of the Fort Translation System . . . . . . . . 15 2.
منابع مشابه
A Branch-and-bound Algorithm For Array Distributions
An important problem facing parallelizing compilers for distributed memory mimd machines is that of distributing data across processors. This distribution aaects the amout of data movements among processors that is required to execute the computations of the input program. This work proposes a branch{and{bound based method to automatically compute a distribution for the arrays of the input prog...
متن کاملA Comparative Study of the Programmability of a Signal Processing Application in an MIMD and an SIMD Multiprocessor
In this report, we address the issues of compilation and execution of a functional program, SISAL (Streams and Iterations in a Single Assignment Language), on the MP-1TM SIMD (Single Instructionstream Multiple Data-stream) parallel machine. SISAL has been successful on many shared memory multiprocessors (SMM) as well as sequential machines. However, the compiler has not been available for distr...
متن کاملCompiler Optimizations for Fortran D on Mimd Distributed-memory Machines Compiler Optimizations for Fortran D on Mimd Distributed-memory Machines
Massively parallel MIMD distributed-memory machines can provide enormous computation power. However, the diiculty of developing parallel programs for these machines has limited their accessibility. This paper presents compiler algorithms to automatically derive eecient message-passing programs based on data decompositions. Optimizations are presented to minimize load imbalance and communication...
متن کاملAndes: a Performance Analyzer for Parallel Programs
ANDES is a performance monitor designed for MIMD distributed memory machines that inserts additional code in the program to be analyzed. ANDES determines the following metrics: speedup, eeciency, experimentally determined serial fraction, percentage of idle time per processor, load and communication balancing, synchronization time and percentage of cpu-communication overlapping. Within ANDES, a...
متن کاملState of the Art in Compiling Hpf Tr Emp Cri A-286
Proposing to the user a nice programming model based on the data-parallel paradigm is one thing. Running the resulting applications very fast is the next issue for a language aiming at high performance on massively parallel machines. This paper discusses the issues involved in HPF compilation and presents optimization techniques, targeting the message-passing SPMD programming model of distribut...
متن کامل