A On-the-Fly Pipeline Parallelism
نویسندگان
چکیده
Pipeline parallelism organizes a parallel program as a linear sequence of stages. Each stage processes elements of a data stream, passing each processed data element to the next stage, and then taking on a new element before the subsequent stages have necessarily completed their processing. Pipeline parallelism is used especially in streaming applications that perform video, audio, and digital signal processing. Three out of 13 benchmarks in PARSEC, a popular software benchmark suite designed for shared-memory multiprocessors, can be expressed as pipeline parallelism. Whereas most concurrency platforms that support pipeline parallelism use a “construct-and-run” approach, this paper investigates “on-the-fly” pipeline parallelism, where the structure of the pipeline emerges as the program executes rather than being specified a priori. On-the-fly pipeline parallelism allows the number of stages to vary from iteration to iteration and dependencies to be data dependent. We propose simple linguistics for specifying on-the-fly pipeline parallelism and describe a provably efficient scheduling algorithm, the PIPER algorithm, which integrates pipeline parallelism into a workstealing scheduler, allowing pipeline and fork-join parallelism to be arbitrarily nested. The PIPER algorithm automatically throttles the parallelism, precluding “runaway” pipelines. Given a pipeline computation with T1 work and T∞ span (criticalpath length), PIPER executes the computation on P processors in TP ≤ T1/P+O(T∞ + lgP) expected time. PIPER also limits stack space, ensuring that it does not grow unboundedly with running time. We have incorporated on-the-fly pipeline parallelism into a Cilk-based work-stealing runtime system. Our prototype Cilk-P implementation exploits optimizations such as lazy enabling and dependency folding. We have ported the three PARSEC benchmarks that exhibit pipeline parallelism to run on Cilk-P. One of these, x264, cannot readily be executed by systems that support only construct-and-run pipeline parallelism. Benchmark results indicate that Cilk-P has low serial overhead and good scalability. On x264, for example, Cilk-P exhibits a speedup of 13.87 over its respective serial counterpart when running on 16 processors.
منابع مشابه
On-the-fly pipeline parallelism Citation
Pipeline parallelism organizes a parallel program as a linear sequence of s stages. Each stage processes elements of a data stream, passing each processed data element to the next stage, and then taking on a new element before the subsequent stages have necessarily completed their processing. Pipeline parallelism is used especially in streaming applications that perform video, audio, and digita...
متن کاملAnalysis of the causes affecting the occurrence of occupational accidents in fly-in / fly-out workers of a gas pipeline dispatching project
Introduction: Prevention of occupational accidents is of a great importance. Analysis of accidents and their causes could prevent the accidents recurrence. The study objective was to investigate the causes of occupational accidents in a dispatching project. Material and method: This research is descriptive-analytic. Data were extracted from the reports of occupational near-misses and accident...
متن کاملA Study of the Role of Repetition of Music in the Nimayee Poems of Akhavan Sales
In this article, the aim of the authors is to study the role of repetition in creating music and parallelism in Nimayee poems in the three collections of poems by Mehdi Akhavan Sales: Zamestan, Akhere Shahname and Az Avesta. Accordingly, the researchers have investigated the various manners of repetition in creating parallelism in the poems at three levels: phonological parallelism, lexical par...
متن کاملRun-time self-reconfigurable 2D convolver for adaptive image processing
Two-dimensional (2D) convolution is a basic operation in digital signal processing, especially in image and video applications. Although its computation is conceptually simple, a sum of products of constants by variables, its implementation is highly demanding in terms of computational power, especially when addressed to real-time embedded systems. This work brings an innovative approach orient...
متن کاملExtending Synchronization Constructs in OpenMP to Exploit Pipeline Parallelism on Heterogeneous Multi-core
The ability of expressing multiple-levels of parallelism is one of the significant features in OpenMP parallel programming model. However, pipeline parallelism is not well supported in OpenMP. This paper proposes extensions to OpenMP directives, aiming at expressing pipeline parallelism effectively. The extended directives are divided into two groups. One can define the precedence at thread lev...
متن کامل