检索结果-内蒙古大学图书馆

international workshop on Policies for distributed systems and Networks

作者： Byung H. Park Thomas J. Naughton Pratul Agarwal David E. Bernholdt Al Geist Jennifer L. Tippens Computer Science and Mathematics Division Oak Ridge National Laboratory Oak Ridge USA Computer Science and Mathematics Division Oak Ridge National Laboratory USA Information Technology Services Division Oak Ridge National Laboratory Oak Ridge USA

Many modern scientific applications, which are designed to utilize high performance parallel computers, occupy hundreds of thousands of computational cores running for days or even weeks. Since many scientists compete for resources, most supercomputing centers practice strict scheduling policies and perform meticulous accounting on their usage. Thus computing resources and time assigned to a user is considered invaluable. However, most applications are not well prepared for unforeseeable faults, still relying on primitive fault tolerance techniques. Considering that ever-plunging mean time to interrupt (MTTI) is making scientific applications more vulnerable to faults, it is increasingly important to provide users not only an improved fault tolerant environment, but also a framework to support their own fault tolerance policies so that their allocation times can be best utilized. This paper addresses a user level fault tolerance policy management based on a holistic approach to digest and correlate fault related information. It introduces simple semantics with which users express their policies on faults, and illustrates how event correlation techniques can be applied to manage and determine the most preferable user policies. The paper also discusses an implementation of the framework using open source software, and demonstrates, as an example, how a molecular dynamics simulation application running on the institutional cluster at Oak Ridge National Laboratory benefits from it.

关键词： Monitoring Fault tolerant systems Fault tolerance Biomedical monitoring Software Registers Correlation

来源：评论

学校读者我要写书评

暂无评论

Rapid prototyping of OpenCV image processing applications using ASP

Rapid prototyping of OpenCV image processing applications us...

引用

international workshop on Rapid System Prototyping (RSP)

作者： Felix Mühlbauer Michael Großhans Christophe Bobda Chair of Computer Engineering University of Potsdam Germany CSCE University of Arkansas USA

Image processing is becoming more and more present in our everyday life. With the requirements of miniaturization, low-power, performance in order to provide some intelligent processing directly into the camera, embedded camera will dominate the image processing landscape in the future. While the common approach of developing such embedded systems is to use sequentially operating processors, image processing algorithms are inherently parallel, thus hardware devices like FP-GAs provide a perfect match to develop highly efficient systems. Unfortunately hardware development is more difficult and there are less experts available compared to software. Automatizing the design process will leverage the existing infrastructure, thus providing faster time to market and quick investigation of new algorithms. We exploit ASP (answer set programming) for system synthesis with the goal of genarating an optimal hardware software partitioning, a viable communication structure and the corresponding scheduling, from an image processing application.

关键词： Hardware Computer architecture Program processors Image processing Delay resource management

来源：评论

学校读者我要写书评

暂无评论

Exploring Fine-Grained Task-Based Execution on Multi-GPU systems

Exploring Fine-Grained Task-Based Execution on Multi-GPU Sys...

引用

IEEE international Conference on Cluster Computing

作者： Long Chen Oreste Villa Guang R. Gao Qualcomm Incorporated San Diego CA USA Pacific Northwest National Laboratory Richland WA USA University of Delaware Newark DE USA

Using multi-GPU systems, including GPU clusters, is gaining popularity in scientific computing. However, when using multiple GPUs concurrently, the conventional data parallel GPU programming paradigms, e.g., CUDA, cannot satisfactorily address certain issues, such as load balancing, GPU resource utilization, overlapping fine grained computation with communication, etc. In this paper, we present a fine-grained task-based execution framework for multi-GPU systems. By scheduling finer-grained tasks than what is supported in the conventional CUDA programming method among multiple GPUs, and allowing concurrent task execution on a single GPU, our framework provides means for solving the above issues and efficiently utilizing multi-GPU systems. Experiments with a molecular dynamics application show that, for nonuniform distributed workload, the solutions based on our framework achieve good load balance, and considerable performance improvement over other solutions based on the standard CUDA programming methodologies.

关键词： Graphics processing unit Kernel Load management Containers Programming Performance evaluation Force

来源：评论

学校读者我要写书评

暂无评论

A distributed server architecture supporting dynamic resource provisioning for BPM-oriented workflow management systems

引用

JOURNAL OF systems AND SOFTWARE 2010年第8期83卷 1538-1552页

作者： Tsai, Ching-Hong Huang, Kuo-Chan Wang, Feng-Jian Chen, Chun-Hao Natl Taichung Univ Dept Comp & Informat Sci Taichung Taiwan Natl Chiao Tung Univ Dept Comp Sci Hsinchu Taiwan

Workflow management systems have been widely used in many business process management (BPM) applications. There are also a lot of companies offering commercial software solutions for BPM. However, most of them adopt a simple client/server architecture with one single centralized workflow-management server only. As the number of incoming workflow requests increases, the single workflow-management server might become the performance bottleneck, leading to unacceptable response time. Development of parallel servers might be a possible solution. However, a parallel server architecture with a fixed-number of servers cannot efficiently utilize computing resources under time-varying system workloads. This paper presents a distributed workflow-management server architecture which adopts dynamic resource provisioning mechanisms to deal with the probable performance bottleneck. We implemented a prototype system of the proposed architecture based on a commercial workflow management system, Agentflow. A series of experiments were conducted on the prototype system for performance evaluation. The experimental results indicate that the proposed architecture can deliver scalable performance and effectively maintain stable request response time under a wide range of incoming workflow request workloads. (C) 2010 Elsevier Inc. All rights reserved.

关键词： Workflow management system Business process management Dynamic resource provisioning

来源：评论

学校读者我要写书评

暂无评论

A Joint distributed resource management and scheduling Scheme for Future UWB-based High-Rate WPAN

A Joint Distributed Resource Management and Scheduling Schem...

引用

4th international Conference on Signal Processing and Communication systems (ICSPCS)/12th international Symposium on DSP and Communication systems (DSPCS)/9th workshop on the Internet, Telecommunications and Signal Processing (WTSP)

作者： Ali-Ahmad, Hassan Khalil, Ayman Crussiere, Matthieu Helard, Jean-Francois INSA IETR F-35708 Rennes France

ISBN: (纸本)9781424479078

In this paper, we investigate the resource management and scheduling issues for the future UWB-based high-rate WPAN systems. Based on a distributed management architecture, we propose a novel multiple access scheme in the context of multiuser channel and time allocation for the UWB applications. Using a self-learning and managing approach, the distributed resource allocation is based on the exchange of useful information among the existing users to achieve a priority-based scheduling. A service differentiation principle is thus followed to guarantee a high level of QoS support for high-priority applications that have strict requirements. However, with the evidence of the service urgency that gives some type of applications a kind of privileged and high-prioritized medium access, fairness among all the users has to be taken into consideration in order not to decrease significantly the performance of some low-priority applications for the sake of other ones. Simulation results demonstrate the efficiency of the simple scheduling scheme that ensures fairness among the competing users while guaranteeing the QoS support for high-priority users.

关键词： distributed management MB-OFDM QoS Fairness

来源：评论

学校读者我要写书评

暂无评论

Grid scheduling based on Collaborative Random Early Detection Strategies

Grid Scheduling based on Collaborative Random Early Detectio...

引用

18th Euromicro international Conference on parallel, distributed and Network-Based Processing (PDP)

作者： Brugnoli, Manuel Heymann, Elisa Senar, Miquel A. Univ Autonoma Barcelona E-08193 Barcelona Spain

ISBN: (纸本)9780769539393

A fundamental problem in large scale Grids is the need for efficient and scalable techniques for resource discovery and scheduling. In traditional resource scheduling systems a single scheduler handles information about all computing resources and schedules jobs. This centralized approach has a serious scalability problem, since it introduces a bottleneck, as well as a single point of failure. Some decentralized scheduling systems have been proposed to improve scalability. However, the main contributions in this area are generally carried out under the assumption of several coordinated schedulers. Nevertheless this approach leads to high communication costs. Such costs are mainly caused by the strong dependency on negotiation among scheduler-to-scheduler and scheduler-to-resource communication. Current approaches to decentralized resource management - in particularly approaches based on Random Early Detection (RED) - are non-coordinated since these schedulers make scheduling related decisions in an independent way. This paper introduces a collaborative model of decentralized scheduling that improves resource scheduling based on RED strategies via gossiping. With this approach, schedulers can receive information from other schedulers without creating a high communication overhead and continue scheduling jobs in an independent way. The simulation results shows that our proposal is scalable and it handles large resources efficiently on large scale Grids.

关键词： scheduling

来源：评论

学校读者我要写书评

暂无评论

Time and cost trade-off management for scheduling parallel applications on Utility Grids

引用

FUTURE GENERATION COMPUTER systems-THE international JOURNAL OF GRID COMPUTING-THEORY METHODS AND APPLICATIONS 2010年第8期26卷 1344-1355页

作者： Garg, Saurabh Kumar Buyya, Rajkumar Siegel, Howard Jay Univ Melbourne Grid Comp & Distributed Syst Lab Dept Comp Sci & Software Engn Melbourne Vic 3010 Australia Colorado State Univ Dept Elect & Comp Engn Ft Collins CO 80523 USA Colorado State Univ Dept Comp Sci Ft Collins CO 80523 USA

With the growth of Utility Grids and various Grid market infrastructures, the need for efficient and cost effective scheduling algorithms is also increasing rapidly, particularly in the area of meta-scheduling. In these environments, users not only may have conflicting requirements with other users, but also they have to manage the trade-off between time and cost such that their applications can be executed most economically in the minimum time. Thus, selection of the best Grid resources becomes a challenge in such a competitive environment. This paper presents three novel heuristics for scheduling parallel applications on Utility Grids that manage and optimize the trade-off between time and cost constraints. The performance of the heuristics is evaluated through extensive simulations of a real-world environment with real parallel workload models to demonstrate the practicality of our algorithms. We compare our scheduling algorithms against existing common meta-schedulers experimentally. The results show that our algorithms outperform existing algorithms by minimizing the time and cost of application execution on Utility Grids. (C) 2009 Elsevier B.V. All rights reserved.

关键词： Grid computing Meta-scheduling Market-oriented resource management

来源：评论

学校读者我要写书评

暂无评论

resource discovery and scheduling in unstructured peer-to-peer desktop grids

Resource discovery and scheduling in unstructured peer-to-pe...

引用

international workshop on scheduling and resource management for parallel and distributed systems

作者： Kwan, Shun Kit Muppala, Jogesh K. Department of Computer Science and Engineering Hong Kong University of Science and Technology Clear Water Bay Kowloon Hong Kong

ISBN: (纸本)9780769541570

In this paper, we explore resource discovery and scheduling issues that arise in unstructured peer-to-peer (P2P) desktop grids. We examine the use of a super-peer based approach to address these issues. The super-peers form a resource information tracking and exchange overlay to enable users to rapidly locate resources for remote execution of jobs. resource availability information is exchanged among the super-peers using a light-weight threshold-driven gossip protocol with the aim of minimizing the resource discovery overhead. We conduct detailed simulation experiments to illustrate the comparative results. Our results indicate that this approach offers a lightweight and scalable method for managing resources in a desktop grid. © 2010 IEEE.

关键词： Scalability

来源：评论

学校读者我要写书评

暂无评论

On better performance from scheduling threads according to resource demands in MMMP

On better performance from scheduling threads according to r...

引用

international workshop on scheduling and resource management for parallel and distributed systems

作者： Weng, Lichen Liu, Chen Department of Electrical and Computer Engineering Florida International University Miami FL 33174 United States

ISBN: (纸本)9780769541570

The Multi-core Multi-threading Microprocessor introduces not only resource sharing to threads in the same core, e.g., computation resources and private caches, but also isolates those resources within different cores. Moreover, when the Simultaneous Multithreading architecture is employed, the execution resources are fully shared among the concurrently executing threads in the same core, while the isolation is worsened as the number of cores increases. Even though fetch policies regarding how to assign priorities in fetch stage are well designed to manage the shared resources in a core, it is actually the scheduling policy that makes the distributed resources available for workloads, through deciding how to send their threads to cores. On the other hand, threads consume various resources in different phases and Cycles Per Instruction Spent on Memory (CPImem) is used to express their resource demands. Consequently, aiming at better performance via scheduling according to their resource demands, we propose the Mix-scheduling to evenly mix threads across cores, so that it achieves thread diversity, i.e., CPImem diversity in every core. As a result, it is observed in our experiment that 63% improvement in overall system throughput and 27% improvement in average thread performance, when comparing the Mix-scheduling policy with the reference policy Mono- scheduling, which keeps CPImem uniformity among threads in every core on chips. Furthermore, the Mix-scheduling also makes an essential step towards shortening load latency, because it succeeds in reducing the L2 Cache Miss Rate by 6% from Mono- scheduling. © 2010 IEEE.

关键词： scheduling

来源：评论

学校读者我要写书评

暂无评论

Hierarchical neural model for workflow scheduling in Utility management systems

Hierarchical neural model for workflow scheduling in Utility...

引用

4th international workshop on Soft Computing Applications, SOFA 2010

作者： Vukmirović, Srdan Erdeljan, Aleksandar Imre, Lendak Nedić, Nemanja Faculty of Technical Sciences University of Novi Sad Novi Sad Serbia

ISBN: (纸本)9781424479849

The emerging computational grid infrastructure consists of heterogeneous resources in widely distributed autonomous domains, which makes job scheduling very challenging. Although there is much work on static scheduling approaches for workflow applications in parallel environments, little work has been done on a real-world Grid environment for industrial systems. Utility management systems (UMS) are executing very large numbers of workflows with very high resource requirements. Unlike the grid approach for standard scientific workflows, UMS workflows have different set of computation requirements and thereby optimization of resource usage has to be made in a different way. This paper proposes architecture for a new scheduling mechanism that dynamically executes a scheduling algorithm using near real-time feedback about current status Grid nodes. Two Artificial Neural Networks (ANN) were created in order to solve scheduling problem. First ANN predicts future state of Grid based on current state and types of workflows that are currently executing. Second ANN output is optimal workflow type that should be executed. Inputs for second ANN are current state of the Grid and predicted future state (output of first ANN). Performance tests show that significant improvement of overall execution time can be achieved by this Hierarchical Artificial Neural Networks. © 2010 IEEE.

关键词： Neural networks

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：