检索结果-内蒙古大学图书馆

Design, Automation and Test in Europe Conference and Exhibition

作者： Guibin Wang National Laboratory of Parallel and Distributed Processing School of Computer National University of Defense Technology Changsha Hunan China

ISBN: (纸本)9781612842080

Although general purpose GPUs have relatively high computing capacity, they also introduce high power consumption compared with general purpose CPUs. Therefore low-power techniques targeted for GPUs will be one of the most hot topics in the future. On the other hand, in several application domains, users are unwilling to sacrifice performance to save power. In this paper, we propose an effective kernel fusion method to reduce the power consumption for GPUs without performance loss. Different from executing multiple kernels serially, the proposed method fuses several kernels into one larger kernel. Owing to the fact that most consecutive kernels in an application have data dependency and could not be fused directly, we split large kernel into multiple slices with strip-mining method, then fuse independent sliced kernels into one kernel. Based on the CUDA programming model, we propose three different kernel fusion implementations, with each one targeting for a special case. Based on the different strip-ming methods, we also propose two fusion mechanisms, which are called invariant-slice fusion and variant-slice fusion. The latter one could be better adapted to the requirements of the kernels to be fused. The experimental results validate that the proposed kernel fusion method could effectively reduce the power consumption for GPU.

关键词： Kernel Energy consumption Graphics processing unit Power demand Instruction sets Programming Optimization

来源：评论

学校读者我要写书评

暂无评论

Fault recovery based on parallel recomputing in transactional memory system

引用

Lecture Notes in Electrical Engineering 2011年 98卷 995-1002页

作者： Song, Wei Jia, Jia National Laboratory for Parallel and Distributed Processing School of Computer National University of Defense Technology Changsha410073 China

ISBN: (纸本)9783642217647

This paper addresses the issue of fault recovery in transactional memory, and proposes a method of fault recovery based on parallel recomputing in transactional memory system. This method utilizes the dataversioning mechanism of transactional memory system to avoid the extra cost of state saving, rolls back a single transaction to avoid wasting the computing time of the fault-free transactions, and adopts the parallel recomputing method to reduce the cost of fault recovery. This paper applies this method to OpenTM programs, and proposes the implementation method of parallel recomputing in OpenTM. At last, this paper tests the performance of this method through a test program. The experimental results show that, compared with the fault recovery method of rolling back a single transaction, the parallel recomputing method in transactional memory system can execute the fault recovery quickly and accurately and the method has a well scalability. © Springer-Verlag Berlin Heidelberg 2011.

关键词： Fault tolerance

来源：评论

学校读者我要写书评

暂无评论

Context-Aware Scheduling in Wireless Networks with Successive Interference Cancellation

Context-Aware Scheduling in Wireless Networks with Successiv...

引用

IEEE International Conference on Communications Workshops

作者： Shaohe Lv Weihua Zhuang Xiaodong Wang Xingming Zhou National Laboratory of Parallel and Distributed Processing National University of Defense Technology Department of Electrical and Computer Engineering University of Waterloo

ISBN: (纸本)9781612842325

We consider the greedy scheduling based on the physical model in wireless networks with successive interference cancellation (SIC). There are two major stages in a scheduling scheme, link selection (to decide which link is scheduled next) and time slot selection (to deciding which slot is allocated to a given link). Most available schemes take a first-fit policy in the latter and strive to achieve good performance by careful selection of link ordering with respect to interference. Due to the accumulation effect and sequential detection nature of SIC, however, it is difficult to evaluate the interference of a link. As a result, many existing scheduling schemes become less efficient. In this paper, we take a new look on the problem and focus to the time slot selection stage. We define tolerance margin to measure the saturation of a link set and present two heuristic policies: one is to schedule a link to a slot such that the resulting set of links has a maximum tolerance margin;the other is to choose a slot such that the increase of tolerance margin is minimum. Simulation results show that the performance of the proposed schemes is better than the first-fit policy and is close to the optimal solution.

关键词： Link scheduling Successive interference cancellation Physical interference model

来源：评论

学校读者我要写书评

暂无评论

An efficient two-level bitmap index for cloud data management

An efficient two-level bitmap index for cloud data managemen...

引用

International Conference on Communication Software and Networks, ICCSN

作者： Huang Bin Peng Yu-Xing School of Computer Science Wuhan University of China Wuhan China National Laboratory of Parallel and Distributed Processing School of Computer Science National University of Defense Technology Changsha China

A Cloud may be seen as a type of flexible computing infrastructure consisting of many compute nodes, where resizable computing capacities can be provided to different customers. To fully harness the power of the Cloud, efficient data management is needed to handle huge volumes of data and support a large number of concurrent end users. To achieve that, a scalable and high-throughput indexing scheme is generally required. Such an indexing scheme must support parallel search to improve scalability. In this paper, we present a bitmap based indexing scheme for efficient data processing in the Cloud. Our approach can be summarized as follows. First, we build a local bitmap index for each compute node which only indexes data residing on the node. Second, we organize the compute nodes as a structured overlay and each node maintains a portion of the global index for the whole different data. The global index is also bitmap index to indicate the node each data resides in. Third, all bitmaps are compressed by adopting run-length coding for reducing storage requirement. We conduct extensive experiments on a LAN, and the results demonstrate that our indexing scheme is dynamic, efficient and scalable.

关键词： TV

来源：评论

学校读者我要写书评

暂无评论

An Outpost Vector placement evaluation of an incremental learning algorithm for Support Vector Machine

An Outpost Vector placement evaluation of an incremental lea...

引用

International Joint Conference on Neural Networks (IJCNN)

作者： Piyabute Fuangkhon Thitipong Tanprasert Distributed and Parallel Computing Research Laboratory Faculty of Science and Technology Assumption University Bangkok Thailand

Outpost Vector model synthesizes new vectors at the boundary of two classes of data in order to increase the level of accuracy of classification. This paper presents a performance evaluation of four different placements of outpost vectors in an incremental learning algorithm for Support Vector Machine (SVM) on a non-complex problem. The algorithm generates outpost vectors from selected new samples, selected prior samples, both samples, or generates no outpost vector at all. After that, they are included in the final training set, as well as new samples and prior samples, based on the specified parameters. The experiments are conducted with a 2-dimension partition problem. The distribution of training and test samples is created in a limited location of a 2-dimension donut ring. There are two classes of data which are represented as 0 and 1. The context of the problem is assumed to shift 45 degrees in counterclockwise direction. Every consecutive partition is set to have different class of data. The experimental results show that the placement of outpost vectors generated from only selected new samples yields the highest level of accuracy of classification for both new data and old data. As a result, using samples from different part of the algorithm to generate outpost vectors affects the level of accuracy of classification.

关键词： Training Support vector machines Accuracy Shape Data models Context Vectors

来源：评论

学校读者我要写书评

暂无评论

An Economical Power Management for High Performance Computers

An Economical Power Management for High Performance Computer...

引用

International Conference on Computational and Information sciences (ICCIS)

作者： Huiquan Wang Xinhai Xu Jing Zhou Hao Zhou Guibin Wang National Laboratory for Parallel and Distributed Processing National University of Defense Technology Changsha Hunan P. R. China

The large amount of energy consumption of high performance computers represent significant financial cost. If we consider the specific situation of electricity price, a measure which makes the energy consumption the lowest is usually not the best on economy. This paper proposes an economical approach for designing a power management plan that reduces the assessment of energy consumption of high performance computers. By taking account of the electricity price in different periods during the day, we assign the workload to each period properly to make the electricity fee as low as possible. A simulation of the cases of a single task and multi-task demonstrates how our approach can be used. The results of the simulation show that our approach can save the electricity fee produced by high performance computers effectively. In the single task case, the proposed approach saves the electricity fee of 10.08% on average, and in the multi-task case the proportion of saved fee is 12.67%.

关键词： Electricity Supercomputers Energy consumption Mathematical model Power demand Equations

来源：评论

学校读者我要写书评

暂无评论

A Unified Co-Processor Architecture for Matrix Decomposition

引用

Journal of Computer science & technology 2010年第4期25卷 874-885页

作者：窦勇周杰邬贵明姜晶菲雷元武倪时策 National Laboratory for Parallel & Distributed Processing National University of Defense Technology National Laboratory for Parallel & Distributed Processing.National University of Defense Technology

QR and LU decompositions are the most important matrix decomposition algorithms. Many studies work on accelerating these algorithms by FPGA or ASIC in a case by case style. In this paper, we propose a unified framework for the matrix decomposition algorithms, combining three QR decomposition algorithms and LU algorithm with pivoting into a unified linear array structure. The QR and LU decomposition algorithms exhibit the same two-level loop structure and the same data dependency. Utilizing the similarities in loop structure and data dependency of matrix decomposition, we unify a fine-grained algorithm for all four matrix decomposition algorithms. Furthermore, we present a unified co-processor structure with a scalable linear array of processing elements （PEs）, in which four types of PEs are same in the structure of memory channels and PE connections, but the only difference exists in the internal structure of data path. Our unified co-processor, which is IEEE 32-bit floating-point precision, is implemented and mapped onto a Xilinx Virtex5 FPGA chip. Experimental results show that our co-processors can achieve speedup of 2.3 to 14.9 factors compared to a Pentium Dual CPU with double SSE threads.

关键词： co-processor matrix decomposition fine-grained parallel FPGA

来源：评论

学校读者我要写书评

暂无评论

An energy-efficient scheduling algorithm for sporadic real-time tasks in multiprocessor systems

An energy-efficient scheduling algorithm for sporadic real-t...

引用

13th IEEE International Workshop on FTDCS 2011, the 8th International Conference on ATC 2011, the 8th International Conference on UIC 2011 and the 13th IEEE International Conference on HPCC 2011

作者： Zhang, Dong-Song Chen, Fang-Yuan Li, Hong-Hua Jin, Shi-Yao Guo, De-Ke National Laboratory for Parallel and Distributed Processing School of Computer National University of Defense Technology Changsha 410073 China School of Computing Science Simon Fraser University 8888 University Drive Burnaby BC V5A 1S6 Canada National Laboratory for Information System Engineering School of Information Systems and Management National University of Defense Technology Changsha 410073 China

ISBN: (纸本)9780769545387

As the energy consumption of embedded multiprocessor systems becomes increasingly prominent, the real-time energy-efficient scheduling in multiprocessor systems becomes an urgent problem to reduce the system energy consumption while meeting real-time constraints. For a multiprocessor with independent DVFS and DPM at each processor, this paper proposes an energy-efficient real-time scheduling algorithm named LRE-DVFS-EACH, based on LRE-TL which is an optimal real-time scheduling algorithm for sporadic tasks. LRE-DVFS-EACH utilizes the concept of TL plane and the idea of fluid scheduling to dynamically scale the voltage and frequency of processors at the initial time of each TL plane as well as the release time of a sporadic task in each TL plane. Consequently, LRE-DVFS-EACH can obtain a reasonable tradeoff between the real-time constraints and the energy saving. LRE-DVFS-EACH is also adaptive to the change of workload caused by the dynamic release of sporadic tasks, which can obtain more energy savings. The experimental results show that compared with existing algorithms, LRE-DVFS-EACH can not only guarantee the optimal feasibility of sporadic tasks, but also achieve more energy savings in all cases, especially in the case of high workloads. © 2011 IEEE.

关键词： Real time systems

来源：评论

学校读者我要写书评

暂无评论

An ordering-based approach to the evaluation of trust systems

An ordering-based approach to the evaluation of trust system...

引用

International Conference on Communication Software and Networks, ICCSN

作者： Wei Liu Yang-Bin Tang Huai-Min Wang Gang Lu School of Computer National University of Defense Technology Changsha Hunan China National Laboratory for Parallel and Distributed Processing Changsha Hunan China

Trust systems provide a promising way to build trust relationships among users in distributed and opening systems. However, it is difficult to make quantitatively comparative analysis on different trust systems because of the different application settings and the lack of effective measures. This paper constructs a framework of trust systems in terms of linear algebra, which helps us model and implement different systems in a uniform way. Besides, we propose an ordering-based approach to evaluating trust systems, then give two relevant ordering-base measures. The experiment results suggests that our method provides an effective way to analyze and evaluate trust systems.

关键词： Computational modeling Blogs

来源：评论

学校读者我要写书评

暂无评论

The dynamic allocation model for the resources of cloud services delivery networks

引用

Jisuanji Xuebao/Chinese Journal of Computers 2011年第12期34卷 2305-2318页

作者： Shi, Pei-Chang Wang, Huai-Min Yin, Gang Liu, Xue-Ning Yuan, Xiao-Qun Shi, Dian-Xi National Laboratory for Parallel and Distributed Processing School of Computer and Science National University of Defense Technology Changsha 410073 China Department of Computer Science and Technology Tsinghua University Beijing 100084 China Department of Electronics and Information Engineering Huazhong Univ. of Sci. and Technol. Wuhan 430074 China

Cloud Services Delivery Networks (CSDN) constructs a layer distributed server overlay over the Internet, which uses the way to the nearest and on-demand approach providing services to end users. Facing the scale and diversification of the resource demand characteristics of the Internet cloud services, CSDN forms different logical sub-server overlay for different kinds of cloud services. However, most servers and bandwidth resources of CSDN are used to deliver the streaming and downloading kind of cloud services, and the dynamic allocation of their delivery resource is the main research emphasis in this paper. This paper first models the problem to be a multi-dimensional facility location problem, according to the two characteristics: the memory resource and bandwidth resource of this kind of application are the bottleneck resource;the hot contents of this kind of application can be delivered using the Peer-to-Peer mechanisms. After the model analyzed and its NP-Complete proved, we then propose a heuristic algorithm. Finally, using the service delivery cost savings as the performance metrics, while the actual system's operation trace is as the input, the effectiveness of the algorithm are comprehensively assessed.

关键词： Heuristic algorithms

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：