Exploiting Course Grain Parallelism from FORTRAN by Mapping it to IF1
نویسندگان
چکیده
FORTRAN, a classical imperative language is mapped into IF1, a machine-independent (SAS) dataaow graph description language with Single Assingment Semantics. Parafrase 2 (P2) is used as the front-end of our system. It parses the source code, generates an intermediate representation and performs several types of analysis. Our system extends the internal representation of P2 with two data structures: the Variable to Edge Translation Table and the Function to Graph Translation Table. It then proceeds to generate IF1 code. The generated IF1 is then processed by the back-end of the Optimizing SISAL Compiler that generates executables on multiple target platforms. We have tested the correctness and the performance of our system with several benchmark. The results show that even on a single processor there is no performance degradation from the translation to SAS. Furthermore, it is shown that on a range of processors, reasonable speedup can be achieved.
منابع مشابه
Dynamic Dependency Analysis of Ordinary Programs 1
A quantitative analysis of program execution is essential to the computer architecture design process. With the current trend in architecture of enhancing the performance of uniprocessors by exploiting ne-grain parallelism, rst-order metrics of program execution, such as operation frequencies, are not su cient; characterizing the exact nature of dependencies between operations is essential. Thi...
متن کاملCoarse-Grain Task Parallel Processing Using the OpenMP Backend of the OSCAR Multigrain Parallelizing Compiler
This paper describes automatic coarse grain parallel processing on a shared memory multiprocessor system using a newly developed OpenMP backend of OSCAR multigrain parallelizing compiler for from single chip multiprocessor to a high performance multiprocessor and a heterogeneous supercomputer cluster. OSCAR multigrain parallelizing compiler exploits coarse grain task parallelism and near ne gra...
متن کاملObject Based Concurrency for Data Parallel Applications: Programmability and Effectiveness
Increased programmability for concurrent applications in distributed systems requires automatic support for some of the concurrent computing aspects. These are: the decomposition of a program into parallel threads, the mapping of threads to processors, the communication between threads, and synchronization among threads. Thus, a highly usable programming environment for data parallel applicatio...
متن کاملRoxana Diaconescu Object Based Concurrency for Data Parallel Applications : Programmability and Effectiveness
Increased programmability for concurrent applications in distributed systems requires automatic support for some of the concurrent computing aspects. These are: the decomposition of a program into parallel threads, the mapping of threads to processors, the communication between threads, and synchronization among threads. Thus, a highly usable programming environment for data parallel applicatio...
متن کاملSelective inline expansion for improvement of multi grain parallelism
This paper proposes a selective procedure inlining scheme to improve a multi-grain parallelism, which hierarchically exploits the coarse grain task parallelism among loops, subroutines and basic blocks and near fine grain parallelism among statements inside a basic block in addition to the loop parallelism. Using the proposed scheme, the parallelism among different layers(nested levels) can be ...
متن کامل