Semantically Sequential, Parallel Execution of Programs on Multiprocessors
نویسنده
چکیده
A program’s execution consistent with the user-specified program order simplifies many aspects of computer design, e.g., programming and resource management. This has greatly benefited uniprocessor system adoption. Parallel systems do not enforce an ordered program execution, complicating system design. Hence, I ask, can parallel execution of programs on multiprocessors be made program-order consistent? I explore one approach to do so, and its impact on performance and system design. Preliminary work shows that the approach is promising.
منابع مشابه
Performance Optimization using Critical Path Analysis in Multithreaded Programs on Multiprocessors
Efficient performance tuning of parallel programs is often hard. Optimization is often done when the program is written as a last effort to increase the performance. With sequential programs each (executed) code segment will affect the total execution time of the program. Thus, any code segment that is optimized in a sequential program will decrease the execution time. In the case of a parallel...
متن کامل&-Prolog and its Performance: Exploiting Independent And-Parallelism
An Independent And-Parallel Prolog model and implementation, ¿¿-Prolog, are described. The description includes a summary of the system's architecture, some details of its execution model (based on the RAP-WAM model), and most importantly, its performance on sequential workstations and shared memory multiprocessors as compared with state-of-the-art Prolog systems. Speedup curves are provided fo...
متن کاملموازی سازی شبیه سازی پدیده ناپایداری دوجریانی به روش PIC
Two stream instability in plasma is simulated by PIC method. The execution time of the sequential and parallizable sections of the program is measured. The sequential program is parallelized with the help of the MPI functions. Then, the execution time of the sequential program versus the number of the grid points and the execution time of the parallel program on 3 and 5 processors versus the nu...
متن کاملA Preliminary Evaluation of Cache-miss-initiated Prefetching Techniques in Scalable Multiprocessors
Prefetching is an important technique for reducing the average latency of memory accesses in scalable cache-coherent multiprocessors. Aggressive prefetching can signiicantly reduce the number of cache misses, but may introduce bursty network and memory traac, and increase data sharing and cache pollution. Given that we anticipate enormous increases in both network bandwidth and latency, we exam...
متن کاملAnd-Or Parallelism on Shared-Memory Multiprocessors
D This paper presents an extended and-or tree and an extended WAM (Warren Abstract Machine) for efficiently supporting both and-parallel and or-parallel execution of logic programs on shared-memory multiprocessors. Our approach for exploiting both andand or-parallelism is based on the binding-arrays method for or-parallelism and the RAP (Restricted And-Parallelism) method for and-parallelism, t...
متن کامل