cient Compilation of High - Level Data Parallel
نویسندگان
چکیده
We present a high-level parallel calculus for nested sequences, NSC, ooered as a possible theoretical \core" of an entire class of collection-oriented parallel languages. NSC is based on while-loops as opposed to general recursion. A formal, machine independent deenition of the parallel time complexity and the work complexity of programs in NSC is given. Our main results are: (1) We give a translation method for a particular form of recursion, called map-recursion, into NSC, that preserves the time complexity and adds an arbitrarily small overhead to the work complexity, and (2) We give a compilation method for NSC into a very simple vector parallel machine, which preserves the time complexity and again adds an arbitrarily small overhead to the work complexity.
منابع مشابه
Compositional Oil Reservoir Simulation in Fortran D: a Feasibility Study On Intel iPsc/860
This paper describes a study of the use of data-parallel languages such as Fortran D or High Performance Fortran (HPF) and their compilation systems for existing large scienti c applications. The central question addressed in this study is: Do such languages allow the expression of the parallelism available in the applications so that the underlying compilation systems are able to generate e ci...
متن کاملThe Plan - Do Style Compilation Technique for Eager Data Transfer in Thread - Based Execution
Plan-do compilation technique is a new, advanced compilation framework for eager data transfer on distributed-memory parallel architectures. The technique is especially eective for a recent breed of low-latency architectures by realizing a high-throughput low-latency communication scheme, pipelined sends. The compilation of high-level, plan-do style code into low-level, eager data transfer code...
متن کاملThe Plan-Du Style Compilation Technique for Eager Data Transfer in Thread-Based Execution
Plan-do compilation technique is a new, advanced compilation framework for eager data transfer on distributed-memory parallel architectures. The technique is especially eective for a recent breed of low-latency architectures by realizing a high-throughput low-latency communication scheme, pipelined sends. The compilation of high-level, plan-do style code into low-level, eager data transfer code...
متن کاملContributions to High Performance Fortran Compilation
This study presents our contributions to High Performance Fortran HPF compilation HPF is a data parallel language based on Fortran Directives are used to specify parallelism and data mapping onto distributed memory parallel architectures We aim at translating a global addressing implicit communication HPF pro gram into a message passing parallel model with local addressing and explicit communic...
متن کاملTraversals of Object Structures : Speci cation andE cient Implementation
We present a new approach, called strategies, to the task of traversing object structures. In our approach traversals are deened using a high-level directed graph description, which is compiled into a dynamic road map to assist run-time traversals. The complexity of the compilation algorithm is polynomial in the size of the strategy graph and the class graph of the given application. The implem...
متن کامل