Results 1 - 10 of 27

Extending the SkelCL Skeleton Library for Stencil Computations on Multi-GPU Systems

by Stefan Breuer, Michel Steuwer, Sergei Gorlatch
"... The implementation of stencil computations on modern, mas-sively parallel systems with GPUs and other accelerators currently relies on manually-tuned coding using low-level approaches like OpenCL and CUDA, which makes it a com-plex, time-consuming, and error-prone task. We describe how stencil compu ..."
Abstract - Cited by 1 (1 self) - Add to MetaCart
computations can be programmed in our SkelCL approach that combines high level of programming abstrac-tion with competitive performance on multi-GPU systems. SkelCL extends the OpenCL standard by three high-level features: 1) pre-implemented parallel patterns (a.k.a. skele-tons); 2) container data types

High-level programming for medical imaging on multi-GPU systems using the SkelCL library

by Michel Steuwer, Sergei Gorlatch - In ICCS , 2013
"... Application development for modern high-performance systems with Graphics Processing Units (GPUs) relies on low-level programming approaches like CUDA and OpenCL, which leads to complex, lengthy and error-prone programs. In this paper, we present SkelCL – a high-level programming model for systems w ..."
Abstract - Cited by 1 (0 self) - Add to MetaCart
(re)distribution mechanism allows for scalability when using multi-GPU systems. We use a real-world example from the field of medical imaging to motivate the design of our programming model and we show how application development using SkelCL is simplified without sacrificing performance: we were able

Parallel Processing Letters c © World Scientific Publishing Company HIGH-LEVEL PROGRAMMING OF STENCIL COMPUTATIONS ON MULTI-GPU SYSTEMS USING THE SKELCL LIBRARY

by Michel Steuwer, Michael Haidl, Stefan Breuer, Sergei Gorlatch , 2014
"... Communicated by Guest Editors The implementation of stencil computations on modern, massively parallel systems with GPUs and other accelerators currently relies on manually-tuned coding using low-level approaches like OpenCL and CUDA. This makes development of stencil applications a complex, time-co ..."
Abstract - Add to MetaCart
-consuming, and error-prone task. We describe how stencil computations can be programmed in our SkelCL approach that combines high-level programming ab-stractions with competitive performance on multi-GPU systems. SkelCL extends the OpenCL standard by three high-level features: 1) pre-implemented parallel patterns (a

Multi-target C++ implementation of parallel skeletons

by Wilfried Kirschenmann, Laurent Plagne, Stephane Vialle - In POOSC ’09: Proceedings of the 8th workshop on Parallel/High-Performance Object-Oriented Scientific Computing , 2009
"... This paper presents the design of an efficient multi-target (CPU+GPU) implementation for the Parallel_for skeleton. Emerging massively parallel architectures promise very high performances for a low cost. However, these architectures change faster than ever. Thus, optimization of codes becomes a ver ..."
Abstract - Cited by 4 (1 self) - Add to MetaCart
This paper presents the design of an efficient multi-target (CPU+GPU) implementation for the Parallel_for skeleton. Emerging massively parallel architectures promise very high performances for a low cost. However, these architectures change faster than ever. Thus, optimization of codes becomes a

Coordinating heterogeneous parallel systems with skeletons and activity graphs

by Murray Cole - Journal of Systems Integration
"... Abstract. Large scale parallel programming projects may become heterogeneous in both language and architectural model. We propose that skeletal programming techniques can alleviate some of the costs involved in designing and porting such programs, illustrating our approach with a simple program whic ..."
Abstract - Cited by 5 (1 self) - Add to MetaCart
which combines shared memory and message passing code. We introduce Activity Graphs as a simple and practical means of capturing model independent aspects of the operational semantics of skeletal parallel programs. They are independent of low level details of parallel implementation and so can act

A Gimmick to Integrate Software Testing Throughout the Curriculum

by Michael H. Goldwasser , 2002
"... We discuss our experiences in which students of a programming course were asked to submit both an implementation as well as a test set. A portion of a student 's grade was then devoted both to the validity of a student's program on others' test sets, as well as how that student's ..."
Abstract - Cited by 23 (0 self) - Add to MetaCart
's test set performed in uncovering flaws in others' programs. The advantages are many, as this introduces implicit principles of software testing together with a bit of fun competition. The major complication is that such an all-pairs execution of tests grows quadratically with the number

Equalizer: A Scalable Parallel Rendering Framework

by Stefan Eilemann, Maxim Makhinya, Renato Pajarola, Multimedia Lab - IEEE Trans. Visualization and Computer Graphics , 2008
"... Abstract — Continuing improvements in CPU and GPU performances as well as increasing multi-core processor and cluster-based parallelism demand for flexible and scalable parallel rendering solutions that can exploit multipipe hardware accelerated graphics. In fact, to achieve interactive visualizatio ..."
Abstract - Cited by 30 (1 self) - Add to MetaCart
pro-vides an application programming interface (API) to develop scalable graphics applications for a wide range of systems ranging from large distributed visualization clusters and multi-processor multipipe graphics systems to single-processor single-pipe desktop ma-chines. We describe the system

Parallel data mining on graphics processors

by Wenbin Fang, Ka Keung Lau, Mian Lu, Xiangye Xiao, Chi Kit Lam, Philip Yang Yang, Bingsheng He, Qiong Luo, Pedro V. S, Ke Yang , 2008
"... We introduce GPUMiner, a novel parallel data mining system that utilizes new-generation graphics processing units (GPUs). Our system relies on the massively multi-threaded SIMD (Single Instruction, Multiple-Data) architecture provided by GPUs. As specialpurpose co-processors, these processors are hi ..."
Abstract - Cited by 18 (0 self) - Add to MetaCart
We introduce GPUMiner, a novel parallel data mining system that utilizes new-generation graphics processing units (GPUs). Our system relies on the massively multi-threaded SIMD (Single Instruction, Multiple-Data) architecture provided by GPUs. As specialpurpose co-processors, these processors

Adaptive Parallelization of Strategies in Agent Based Systems

by C. Geiger, B. Kalthoff
"... In this paper we present an approach for adaptive partitioning of strategies in agent oriented systems based on algorithmic skeletons [2]. This provides the user with parallel programming templates for modeling agent strategies and their efficient parallelization. Information present in these skelet ..."
Abstract - Add to MetaCart
In this paper we present an approach for adaptive partitioning of strategies in agent oriented systems based on algorithmic skeletons [2]. This provides the user with parallel programming templates for modeling agent strategies and their efficient parallelization. Information present

Proceedings of the 28th Annual Hawaii International Conference on System Sciences- 1995 Software Reuse and Portability of Parallel Programs*

by Helmar Burkhart, Stephan Gutzwiller
"... The state-of-the-art of programming parallel com-puters is far from being successful. The main challenge today is therefore the development of techniques and tools that improve the programmer’s situation. Soft-ware reuse and software portability are two research areas where further progress is essen ..."
Abstract - Add to MetaCart
is essential. We present an approach that is applicable for compute-intensive programs with regular process topologies and execution patterns. After a short introduction, we summarize the Base1 Algorithm Classification Scheme, which is the base of all our implementation parts, and present three sam
Results 1 - 10 of 27