检索结果-内蒙古大学图书馆

A uniform approach for programming distributed heterogeneous computing systems

JOURNAL OF PARALLEL AND DISTRIBUTED COMPUTING 2014年第12期74卷 3228-3239页

作者： Grasso, Ivan Pellegrini, Simone Cosenza, Biagio Fahringer, Thomas Univ Innsbruck Inst Comp Sci A-6020 Innsbruck Austria Barcelona Supercomp Ctr Barcelona Spain

Large-scale compute clusters of heterogeneous nodes equipped with multi-core CPUs and GPUs are getting increasingly popular in the scientific community. However, such systems require a combination of different programming paradigms making application development very challenging. In this article we introduce libWater, a library-based extension of the OpenCL programming model that simplifies the development of heterogeneous distributed applications. libWater consists of a simple interface, which is a transparent abstraction of the underlying distributed architecture, offering advanced features such as inter-context and inter-node device synchronization. It provides a runtime system which tracks dependency information enforced by event synchronization to dynamically build a DAG of commands, on which we automatically apply two optimizations: collective communication pattern detection and device-host-device copy removal. We assess libWater's performance in three compute clusters available from the Vienna Scientific Cluster, the Barcelona Supercomputing Center and the University of Innsbruck, demonstrating improved performance and scaling with different test applications and configurations. (C) 2014 The Authors. Published by Elsevier Inc.

关键词： OpenCL MPI Distributed computing Heterogeneous computing programming model Runtime system

来源：评论

学校读者我要写书评

暂无评论

Task scheduling of cloud computing based on Improved CHC algorithm 4

Task scheduling of cloud computing based on Improved CHC alg...

引用

International Conference on Audio, Language and Image Processing

作者： Zhang, Liping Tong, Weiqin Lu, Shengpeng Shanghai Univ Sch Comp Engn & Sci Shanghai Peoples R China

ISBN: (纸本)9781479939039

In recent years, research on task scheduling in cloud computing has become a hot topic. Task scheduling is a NP hard problem. As there is a great number of users in cloud computing, the amount of task and data processing is also very huge. With regard to cloud computing programming model, we proposed an Improved CHC algorithm, which inherits the advantages of standard genetic algorithm (SGA) and CHC algorithm. The Improved CHC algorithm takes comprehensive consideration of the allocation of resources to tasks and the completion time of tasks and it also enhances the convergence speed of the algorithm. The experimental results show that Improved CHC algorithm has good efficiency and convergence, the average and completion time of the task scheduling is relatively shorter.

关键词： cloud computing task scheduling CHC algorithm programming model

来源：评论

学校读者我要写书评

暂无评论

τC: C with Process Network Extensions for Embedded Manycores

τC: C with Process Network Extensions for Embedded Manycore...

引用

14th Annual International Conference on Computational Science

作者： Goubier, Thierry Courousse, Damien Azaiez, Selma CEA LIST DACLE Lastre F-91191 Gif Sur Yvette France CEA LIST DACLE LIALP Grenoble France

Current and future embedded manycore systems bring complex and heterogeneous architectures with a large number of processing cores, making both parallel programming at this scale and understanding the architecture itself a daunting task. Process Networks and other dataflow based models of Computation (MoC) are a good base to present a universal model of the underlying manycore architectures to the programmer. If a language exposes a simple to grasp MoC in a consistent way across architectures, the programmer can concentrate the efforts on optimizing the expression of parallelism in the application instead of porting code to a given system. In this paper, we present a process network extension to C called tau C and its mapping to both a POSIX target and the P2012/STHORM platform, and show how the language offers an architecture independent solution of this problem.

关键词： manycore programming model process networks compiler toolchain

来源：评论

学校读者我要写书评

暂无评论

Approaching Coordination in Distributed Embedded Applications with the Peer model DSL 40

Approaching Coordination in Distributed Embedded Application...

引用

40th Euromicro Conference on Software Engineering and Advanced Applications (SEAA)

作者： Kuehn, Eva Crass, Stefan Hamboeck, Thomas Vienna Univ Technol Inst Comp Languages Vienna Austria

ISBN: (纸本)9781479957958

Coordination in distributed embedded systems requires complex synchronization of many concurrent activities. This task becomes especially difficult when network and system failures have to be assumed. The Peer model is a novel programming model for the design of coordination strategies among multiple nodes, aiming to bridge design and implementation. A major advantage is that designs based on the Peer model are very flexible regarding changing requirements and policies. The motivating use case is an application in the railway domain where embedded nodes detect approaching trains and route this information over several forwarder nodes to the level crossing. In this paper, we present a Domain Specific Language for the Peer model which allows to automatically generate a graphical documentation and source code for different embedded platforms. It lays the foundations for an embedded system software development toolchain. We prove the feasibility by implementing an event notification strategy for the level crossing use case.

关键词： concurrency control distributed processing embedded systems railway engineering software engineering source code (software) system documentation approaching train detection complex synchronization concurrent activities distributed embedded systems domain specific language embedded nodes embedded platforms embedded system software devel- opment toolchain event notification strategy forwarder nodes graphical documentation level crossing peer model programming model railway domain source code system failures Containers Documentation Middleware Peer-to-peer computing Unified modeling language Wiring LEVEL CROSSING Embedded systems concurrent activities system failures program and system documentation Railway engineering Peer to peer computing Source code UML WIRING Concurrency control Distributed processing Middleware CONTAINERS Software engineering

来源：评论

学校读者我要写书评

暂无评论

ParaDIME: Parallel Distributed Infrastructure for Minimization of Energy 17

ParaDIME: Parallel Distributed Infrastructure for Minimizati...

引用

17th Euromicro Conference on Digital System Design (DSD)

作者： Rethinagiri, Santhosh Kumar Palomar, Oscar Sobe, Anita Knauth, Thomas Barczynski, Wojciech Yalcin, Gulay Hayduk, Yarco Cristal, Adrian Unsal, Osman Felber, Pascal Fetzer, Christof Ryckaert, Julien Alioto, Gina Barcelona Supercomp Ctr Ctr Nacl Supercomp BSC Barcelona Spain Univ Neuchatel Unine Neuchatel Switzerland Tech Univ Dresden Dresden Germany Cloud & Heat Technol GmbH Dresden Germany IMEC Heverlee Belgium Spanish Natl Res Council CSIC IIIA Madrid Spain

ISBN: (纸本)9781479957934

Dramatic environmental and economic impact of the ever increasing power and energy consumption of modern computing devices in data centers is now a critical challenge. On one hand, designers use technology scaling as one of the methods to face the phenomenon called dark silicon (only segments of a chip function concurrently due to power restrictions). On the other hand, designers use extreme-scale systems such as teradevices to meet the performance needs of their applications which in turn increases the power consumption of the platform. In order to overcome these challenges, we need novel computing paradigms that address energy efficiency. One of the promising solutions is to incorporate parallel distributed methodologies at different abstraction levels. The FP7 project ParaDIME focuses on this objective to provide different distributed methodologies (software-hardware techniques) at different abstraction levels to attack the power-wall problem. In particular, the ParaDIME framework will utilize: circuit and architecture operation below safe voltage limits for drastic energy savings, specialized energy-aware computing accelerators, heterogeneous computing, energy-aware runtime, approximate computing and power-aware message passing. The major outcome of the project will be a processor architecture for a heterogeneous distributed system that utilizes future device characteristics for drastic energy savings. Wherever possible, ParaDIME will adopt multidisciplinary techniques, such as hardware support for message passing, runtime energy optimization utilizing new hardware energy performance counters, use of accelerators for error recovery from sub-safe voltage operation, and approximate computing through annotated code. Furthermore, we will establish and investigate the theoretical limits of energy savings at the device, circuit, architecture, runtime and programming model levels of the computing stack, as well as quantify the actual energy savings achieved by the Para

关键词： Computer architecture Energy consumption Hardware Message passing Performance evaluation Runtime Servers Data center Energy minimization Heterogeneous devices Low-power programming model Runtime data center message passing energy saving Computer Architecture Computer hardware energy consumption Servers LOW POWER Performance Evaluation Energy sources Processor Architectures ENERGY MINIMISATION Accelerators

来源：评论

学校读者我要写书评

暂无评论

Message passing on InfiniBand RDMA for parallel run-time supports

Message passing on InfiniBand RDMA for parallel run-time sup...

引用

22nd Euromicro International Conference on Parallel, Distributed, and Network-Based Processing (PDP)

作者： Secco, Alessandro Uddin, Irfan Pezzi, Guilherme Peretti Torquati, Massimo Univ Torino Dept Comp Sci I-10149 Turin Italy Univ Pisa Dept Comp Sci I-56127 Pisa Italy

ISBN: (纸本)9781479927289

InfiniBand networks are commonly used in the high performance computing area. They offer RDMA-based operations that help to improve the performance of communication subsystems. In this paper, we propose a minimal message-passing communication layer providing the programmer with a point-topoint communication channel implemented by way of InfiniBand RDMA features. Differently from other libraries exploiting the InfiniBand features, such as the well-known Message Passing Interface (MPI), the proposed library is a communication layer only rather than a programming model, and can be easily used as building block for high-level parallel programming frameworks. Evaluated on micro-benchmarks, the proposed RDMA-based communication channel implementation achieves a comparable performance with highly optimised MPI/InfiniBand implementations. Eventually, the flexibility of the communication layer is evaluated by integrating it within the FastFlow parallel framework, currently supporting TCP/IP networks (via the ZeroMQ communication library).

关键词： message passing telecommunication channels transport protocols FastFlow parallel framework MPI-infiniband implementations RDMA-based communication channel RDMA-based operations TCP-IP networks ZeroMQ communication library communication subsystems high performance computing area infiniband RDMA infiniband networks message passing interface message-passing communication layer parallel run-time supports point-to-point communication channel programming model IP networks Libraries Message systems Peer-to-peer computing Ports (Computers) Protocols Receivers MPI distributed computing infiniband networks multi-core

来源：评论

学校读者我要写书评

暂无评论

programming model for a Heterogeneous x86 Platform 09

Programming Model for a Heterogeneous x86 Platform

引用

ACM SIGPLAN Conference on programming Language Design and Implementation

作者： Saha, Bratin Zhou, Xiaocheng Chen, Hu Gao, Ying Yan, Shoumeng Rajagopalan, Mohan Fang, Jesse Zhang, Peinan Ronen, Ronny Mendelson, Avi Intel Corp Santa Clara CA 95051 USA

ISBN: (纸本)9781605583921

The client computing platform is moving towards a heterogeneous architecture consisting of a combination of cores focused on scalar performance, and a set of throughput-oriented cores. The throughput oriented cores (e.g. a GPU) may be connected over both coherent and non-coherent interconnects, and have different ISAs. This paper describes a programming model for such heterogeneous platforms. We discuss the language constructs, runtime implementation, and the memory model for such a programming environment. We implemented this programming environment in a x86 heterogeneous platform simulator. We ported a number of workloads to our programming environment, and present the performance of our programming environment on these workloads.

关键词： heterogeneous platforms programming model

来源：评论

学校读者我要写书评

暂无评论

A Middleware for Managing Non-Functional Requirements in Cloud PaaS

A Middleware for Managing Non-Functional Requirements in Clo...

引用

International Conference on Cloud and Autonomic Computing (ICCAC)

作者： Braubach, Lars Jander, Kai Pokahr, Alexander Univ Hamburg Dept Comp Sci Distributed Syst Grp Hamburg Germany

ISBN: (纸本)9781479958412

A key aspect of distributed applications in platformas-a-service clouds revolves around the definition of requirements for quality of service and the most efficient use of the available resources to achieve them. Matching both the requirements and resources represents a unique challenge: While requirements are generally defined in relatively abstract form and defined as system-wide parameters, the resources are concrete, low-level entities that, by themselves, are only of marginal influence on the requirements. In order to establish a relationship between these two requires both the agglomeration of properties into more abstract ones as well as refining requirements in more detailed, concrete ones, both of which can often only be done to a certain degree. This paper presents a vision how both aspects can be brought to the same level and introduces an approach enhancing component-based cloud applications with non-functional properties in the context of an integrated programming model. This allows developers of such applications to formally declare properties that may influence the application during runtime and document the expectations regarding non-functional aspects of the overall system. This can be used to ensure the health of the application and, in case requirements are violated, actions like the deployment of additional resources can be performed in order to alleviate any degradation of its non-functional state, either by evaluating, choosing and performing pre-defined actions automatically or by manual administrative intervention.

关键词： cloud computing middleware object-oriented programming administrative intervention cloud PaaS component-based cloud applications integrated programming model nonfunctional properties nonfunctional requirements platform-as-a-service clouds Concrete Current measurement Middleware Monitoring programming Quality of service Runtime cloud applications components platform-as-a-service programming model Middleware Cloud Computing Runtime Quality of service Electric current measurement Current measuring object oriented programming programming Concrete linear programming models Integer programming models

来源：评论

学校读者我要写书评

暂无评论

programming model for a heterogeneous x86 platform

Programming model for a heterogeneous x86 platform

引用

作者： Saha, Bratin Zhou, Xiaocheng Chen, Hu Gao, Ying Yan, Shoumeng Rajagopalan, Mohan Fang, Jesse Zhang, Peinan Ronen, Ronny Mendelson, Avi Intel Corporation

关键词： Heterogeneous platforms programming model

来源：评论

学校读者我要写书评

暂无评论

τ C: C with Process Network Extensions for Embedded Manycores

引用

Procedia Computer Science 2014年 29卷 1100-1112页

作者： Thierry Goubier Damien Couroussé Selma Azaiez CEA LIST DACLE/Lastre 91191 Gif-sur-Yvette CEDEX France CEA LIST DACLE/LIALP Grenoble France

Current and future embedded manycore systems bring complex and heterogeneous architectures with a large number of processing cores, making both parallel programming at this scale and understanding the architecture itself a daunting task. Process Networks and other dataflow based models of Computation (MoC) are a good base to present a universal model of the underlying manycore architectures to the programmer. If a language exposes a simple to grasp MoC in a consistent way across architectures, the programmer can concentrate the efforts on optimizing the expression of parallelism in the application instead of porting code to a given system. In this paper, we present a process network extension to C called τ C and its mapping to both a POSIX target and the P2012/STHORM platform, and show how the language offers an architecture independent solution of this problem.

关键词： manycore programming model process networks compiler toolchain

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：