检索结果-内蒙古大学图书馆

21st international workshop on languages and compilers for parallel computing

作者： Burtscher, Martin Kulkarni, Milind Prountzos, Dimitrios Pingali, Keshav Univ Texas Austin Ctr Grid & Distributed Comp Inst Computat Engn & Sci Austin TX 78712 USA

ISBN: (纸本)9783540897392

Irregular applications, i.e., programs that manipulate pointer-based data structures such as graphs and trees, constitute a challenging target for parallelization because the amount of parallelism is input dependent and changes dynamically. Traditional dependence analysis techniques are too conservative to expose this parallelism. Even manual parallelization is difficult, time consuming, and error prone. The Galois system parallelizes such applications using an optimistic approach that exploits higher-level semantics of abstract data types. In this paper, we study the performance and scalability of a Galoised, that is, automatically parallelized, version of Delaunay mesh refinement (DR) on a shared-memory system with 128 CPUs. DR is an important irregular application that is used, e.g., in graphics and finite-element codes. The parallelized program scales to 64 threads, where it reaches a speedup of 25.8. For large numbers of threads, the performance is hampered by the load imbalance and the nonuniform memory latency, both of which grow as the number of threads increases. While these two issues will have to be addressed in future work, we believe our results already show the Galois approach to be very promising.

关键词： parallel programming multicore processors sparse graph algorithm amorphous data-parallelism optimistic execution mesh refinement

来源：评论

学校读者我要写书评

暂无评论

Modular Array-Based GPU computing in a Dynamically-Typed Language 4

Modular Array-Based GPU Computing in a Dynamically-Typed Lan...

引用

4th ACM SIGPLAN international workshop on Libraries, languages, and compilers for Array Programming (ARRAY) / 1st ACM SIGPLAN international workshop on Machine Learning and Programming languages (MAPL)

作者： Springer, Matthias Wauligmann, Peter Masuhara, Hidehiko Tokyo Inst Technol Dept Math & Comp Sci Tokyo Japan

ISBN: (纸本)9781450350693

Nowadays, GPU accelerators are widely used in areas with large data-parallel computations such as scientific computations or neural networks. Programmers can either write code in low-level CUDA/OpenCL code or use a GPU extension for a high-level programming language for better productivity. Most extensions focus on statically-typed languages, but many programmers prefer dynamically-typed languages due to their simplicity and flexibility. This paper shows how programmers can write high-level modular code in Ikra, a Ruby extension for array-based GPU computing. Programmers can compose GPU programs of multiple reusable parallel sections, which are subsequently fused into a small number of GPU kernels. We propose a seamless syntax for separating code regions that extensively use dynamic language features from those that are compiled for efficient execution. Moreover, we propose symbolic execution and a program analysis for kernel fusion to achieve performance that is close to hand-written CUDA code.

关键词： GPGPU CUDA Ruby kernel fusion

来源：评论

学校读者我要写书评

暂无评论

A Fully parallel LISP2 Compactor with Preservation of the Sliding Properties

A Fully Parallel LISP2 Compactor with Preservation of the Sl...

引用

21st international workshop on languages and compilers for parallel computing

作者： Li, Xiao-Feng Wang, Ligang Yang, Chen Intel China Res Ctr Managed Runtime Optimizat Beijing Peoples R China

ISBN: (纸本)9783540897392

Compacting garbage collector (GC) is widely used due to its good properties of in-place collection and heap de-fragmentation. In addition, it supports fast bump-pointer allocation and provides good access locality. Most known commercial JVM or CLR implementations use compaction algorithm in certain garbage collection scenarios, such as in full heap or mature object space collections. LISP2 compactor is one of the best-known GC algorithms. As multi-core architecture prevails, several efficient parallel compactors have been proposed. Nevertheless, there is no parallel LISP2 compactor available that can preserve all the sliding properties of its sequential counterpart. That is, to compact live data in-place into a single contiguous region in one end of the heap while maintaining the original object order. In this paper, we propose a fully parallel LISP2 compactor that keeps all the sliding properties. We also prove the correctness of the design. This parallel LISP2 compactor is fully parallel because all of its four phases are parallelized and the workloads are well balanced among the collector threads. The compactor supports fall-back compaction and adjustable boundaries that help deliver the best performance. We have implemented the parallel LISP2 compactor in Apache Harmony, it product-quality open source Java SE implementation. We evaluate and discuss the design on an Intel 8-core platform with representative benchmark.

关键词： Garbage collector compactor parallelization

来源：评论

学校读者我要写书评

暂无评论

languages and compilers for parallel computing 2008

引用

丛书名： Lecture Notes in Computer Science

2008年

作者： José Nelson Amaral

来源：评论

学校读者我要写书评

暂无评论

Compiler Optimizations for parallel Programs 1

引用

31st international workshop on languages and compilers for parallel computing (LCPC)

作者： Doerfert, Johannes Finkel, Hal Argonne Natl Lab Argonne Leadership Comp Facil 9700 S Cass Ave Argonne IL 60439 USA

ISBN: (数字)9783030346270

ISBN: (纸本)9783030346270;9783030346263

This paper outlines a research and development program to enhance modern compiler technology, and the LLVM compiler infrastructure specifically, to directly optimize parallel-programming-model constructs. The goal is to produce higher-quality code, and moreover, to remove abstraction penalties generally associated with such constructs. We believe that such abstraction penalties are increasing in importance due to C++ parallel-algorithms libraries and other performance-portability-motivated programming methods. In addition, we will discuss when, and more importantly when not, explicit parallelism-awareness is necessary within the compiler in order to enable the desired optimization capabilities.

关键词： parallel programming LLVM OpenMP Compiler optimizations Intermediate representation Programming models

来源：评论

学校读者我要写书评

暂无评论

Principles of speculative run-time parallelization 11th

引用

11th international workshop on languages and compilers for parallel computing

作者： Patel, D Rauchwerger, L Texas A&M Univ Dept Comp Sci College Stn TX 77843 USA

ISBN: (纸本)3540664262

Current parallelizing compilers cannot identify a significant fraction of parallelizable loops because they have complex or statically insufficiently defined access patterns. We advocate a novel framework for the identification of parallel loops. It speculatively executes a loop as a doall and applies a fully parallel data dependence test to check for any unsatisfied data dependencies;if the test fails, then the loop is re-executed serially. We will present the principles of the design and implementation of a compiler that employs both run-time and static techniques to parallelize dynamic applications. Run-time optimizations always represent a tradeoff between a speculated potential benefit and a certain (sure) overhead that must be paid. We will introduce techniques that take advantage of classic compiler methods to reduce the cost of run-time optimization thus tilting the outcome of speculation in favor of significant performance gains. Experimental results from the PERFECT, SPEC and NCSA Benchmark suites show that these techniques yield speedups not obtainable by any other known method.

关键词： Program compilers

来源：评论

学校读者我要写书评

暂无评论

languages and compilers for parallel computing 1

引用

丛书名： Lecture Notes in Computer Science

1000年

作者： José Nelson Amaral

ISBN: (数字)9783540897408

ISBN: (纸本)9783540897392

In 2008 the workshop on languages and compilers for parallel computing left the USA to celebrate its 21st anninversary in Edmonton, Alberta, Canada. Following its long-established tradition, the workshop focused on topics at the frontierofresearchanddevelopmentinlanguages,optimizingcompilers,appli- tions, and programming models for high-performance computing. While LCPC continues to focus on parallel computing, the 2008 edition included the pres- tation of papers on program analysis that are precursors of high performance in parallel environments. LCPC 2008 received 35 paper submissions. Eachpaper received at least three independent reviews, and then the papers and the referee comments were d- cussed during a Program Committee meeting. The PC decided to accept 18 papers as regular papers and 6 papers as short papers. The short papers appear at the end of this volume. The LCPC 2008 program was fortunate to include two keynote talks. Keshav Pingali’s talk titled “Amorphous Data parallelism in Irregular Programs” - gued that irregular programs have data parallelism in the iterative processing of worklists. Pingali described the Galois system developed at The University of Texas at Austin to exploit this kind of amorphous data parallelism. The second keynote talk, “Generic parallelAlgorithms in Threading Building Bocks (TBB),” presented by Arch Robison from Intel Corporation addressed very practical aspects of using TBB, a production C++ library, for generic p- allel programming and contrasted TBB with the standard Template Library (stL).

关键词： Programming languages, compilers, Interpreters Programming Techniques Computer Communication Networks Data structures Artificial Intelligence Models and Principles

来源：评论

学校读者我要写书评

暂无评论

Algebraic processing of programming languages

引用

THEORETICAL COMPUTER SCIENCE 1998年第1-2期199卷 105-143页

作者： Rus, T Univ Iowa Dept Comp Sci Iowa City IA 52242 USA

Current methodology for compiler construction evolved in small increments over a long period of time. Its heritage is machine-dependent and derived from sequential Von Neumann machines. There is a growing emphasis on increasingly abstract paradigms for new programming languages. At the same time today's high performance distributed/parallel computing facilities depart from Von Neumann machines and provide a much more intricate execution environment. Therefore current methodology is being stretched beyond its intrinsic capacity in order to accommodate these two accelerating trends. We develop an alternative compiler construction methodology whose fundamental principles are: (1) decomposition of programming languages into simpler components, (2) development of machine independent specification and implementation tools for each language component, and (3) mathematical integration of language component processing algorithms into an algebraic compiler. This allows the specification and implementation of provably correct (commercial) compilers. This paper is a tutorial dedicated to presenting the infrastructure of an algebraic compiler in a do-it-yourself manner. (C) 1998-Elsevier Science B.V. All rights reserved.

关键词： algebra compiler derived-operation homomorphism language

来源：评论

学校读者我要写书评

暂无评论

HDArray: parallel Array Interface for Distributed Heterogeneous Devices 1

引用

31st international workshop on languages and compilers for parallel computing (LCPC)

作者： Cho, Hyun Dok Kwon, Okwan Midkiff, Samuel P. NVIDIA Corp Santa Clara CA 95051 USA Apple Inc Cupertino CA 95014 USA Purdue Univ W Lafayette IN 47907 USA

ISBN: (数字)9783030346270

ISBN: (纸本)9783030346270;9783030346263

Heterogeneous clusters with nodes containing one or more accelerators, such as GPUs, have become common. While MPI provides inter-address space communication, and OpenCL provides a process with access to heterogeneous computational resources, programmers are forced to write hybrid programs that manage the interaction of both of these systems. This paper describes an array programming interface that provides users with automatic and manual distributions of data and work. Using work distribution, and kernel def and use information, communication among processes and devices in a process is performed automatically. By providing a unified programming model to the user, program development is simplified.

关键词： Program processors

来源：评论

学校读者我要写书评

暂无评论

Euro-Par 2015: parallel Processing 1

引用

丛书名： Lecture Notes in Computer Science

1000年

作者： Jesper Larsson Träff Sascha Hunold Francesco Versaci

ISBN: (数字)9783662480960

ISBN: (纸本)9783662480953

This book constitutes the refereed proceedings of the 21st international Conference on parallel and Distributed computing, Euro-Par 2015, held in Vienna, Austria, in August 2015. The 51 revised full papers presented together with 2 invited papers were carefully reviewed and selected from 190 submissions. The papers are organized in the following topical sections: support tools and environments; performance modeling, prediction and evaluation; scheduling and load balancing; architecture and compilers; parallel and distributed data management; grid, cluster and cloud computing; distributed systems and algorithms; parallel and distributed programming, interfaces and languages; multi- and many-core programming; theory and algorithms for parallel computation; numerical methods and applications; and accelerator computing.

关键词： Programming languages, compilers, Interpreters Operating Systems System Performance and Evaluation Computer Systems Organization and Communication Networks Algorithm Analysis and Problem Complexity Software Engineering/Programming and Operating Systems

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：