In this paper, we develop a new software layer named Automatic parallel Detection Layer (APDL) for automatic transformation from sequential code to parallel one. The main interest, in this research, is the parallelism...
详细信息
ISBN:
(纸本)0780370872
In this paper, we develop a new software layer named Automatic parallel Detection Layer (APDL) for automatic transformation from sequential code to parallel one. The main interest, in this research, is the parallelism at loop level, because significant parallelism in programs almost invariably occurs in loops. The proposed APDL has five processes for code transformation: the sequential source code parser, data dependence analysis of this code, partitioning, scheduling both task and data, and generating parallel source code. Many cases have been studied to evaluate the performance of the developed layer. The performance is evaluated depended on the execution time of: sequential code, the parallel programmer code, and the code output from APDL for the same case study. Performance results show that APDL greatly improves the execution time with respect to sequential execution time, and save the high cost of parallel programmer.
The performance of parallel architectures are limited at least as much by data transfer ability as by computing power. The main limit concerns the transfers on the interconnection network. But it becomes apparent that...
详细信息
The performance of parallel architectures are limited at least as much by data transfer ability as by computing power. The main limit concerns the transfers on the interconnection network. But it becomes apparent that a majority of these communications can be known at compile time. The static model intends to exploit this a priori knowledge in order to drastically reduce the overhead of message passing, the ultimate goal being to confine the oommunication delays to the hardware propagation delays. In the paper, we present an abstract machine which is the target of a static-oriented compilation. We show how to recognize and sequence the static communication patterns, and we discuss the application scope of the model.
Communication set generation significantly influences the performance of parallel programs. However, seldom works gives attention to the communication generation problem for irregular applications. In this paper, we w...
详细信息
Communication set generation significantly influences the performance of parallel programs. However, seldom works gives attention to the communication generation problem for irregular applications. In this paper, we will explain how support to generate communication set for irregular array references in loops. We propose a compile-time algorithm by introducing some symbolic analysis techniques. In our symbolic analysis system, a set of symbolic solutions of a symbolic expression system is solved by limiting some restrictions. For this proposal, we introduce some symbolic analysis algorithms to fix solutions in a system of equalities and inequalities. Finally, we show experimental results on a parallel computer CM-5 that validate our approach.
暂无评论