检索结果-内蒙古大学图书馆

作者： Rao, Xiang Yin, Gang Wang, Huaimin Shi, Dianxi Zhu, Yanxu National Laboratory for Parallel and Distributed Processing National University of Defense Technology Changsha China

ISBN: (纸本)9788988678367

Mining log pattern to analyze the faults in large scale distributed system is affected by the existence of redundant and ambiguous noisy error logs. While existing works try to compress logs in a coarse granularity from temporal and spatial view to remove the redundancy, they fail to reserve those ambiguous logs that might truly relate to a fault, which misleads the fault characterizing result. By modeling error logs as time series and examining the similarity between trash error log template and target error log, the ambiguous error logs are kept and the affected patterns can be effectively removed. Experiments in a practical complex service-based storage show that up to 92% of the affected patterns can be filtered. © 2011 AICIT.

关键词： Errors

来源：评论

学校读者我要写书评

暂无评论

A programming language approach to internet-based virtual environment computing

A programming language approach to internet-based virtual en...

引用

作者： Ji, Wang Shen, Rui Wang, Huai-Min National Laboratory for Parallel and Distributed Processing School of Computer National University of Defense Technology Changsha 410073 China

There is an increasing need to build scalable distributed systems over the Internet infrastructure. However, the development of distributed scalable applications suffers from lack of a wide accepted virtual computing environment. Users have to take great efforts on the management and sharing of the involved resources over Internet, whose characteristics are intrinsic growth, autonomy and diversity. To deal with this challenge, Internet-based Virtual Computing Environment (iVCE) is proposed and developed to serve as a platform for distributed scalable applications over the open infrastructure, whose kernel mechanisms are on-demand aggregation and autonomic collaboration of resources. In this paper, we present a programming language for iVCE named Owlet. Owlet conforms with the conceptual model of iVCE, and exposes the iVCE to application developers. As an interaction language based on peer-to-peer content-based publish/subscribe scheme, Owlet abstracts the Internet as an environment for the roles to interact, and uses roles to build a relatively stable view of resources for the on-demand resource aggregation. It provides language constructs to use 1) distributed event driven rules to describe interaction protocols among different roles, 2) conversations to correlate events and rules into a common context, and 3) resource pooling to do fault tolerance and load balancing among networked nodes. We have implemented an Owlet compiler and its runtime environment according to the architecture of iVCE, and built several Owlet applications, including a peer-to-peer file sharing application. Experimental results show that, with iVCE, the separation of resource aggregation logic and business logic significantly eases the process of building scalable distributed applications. © 2011 Springer science+Business Media, LLC & science Press, China.

关键词： Fault tolerance

来源：评论

学校读者我要写书评

暂无评论

A task-mapping algorithm for performance simulation based on simulated annealing

A task-mapping algorithm for performance simulation based on...

引用

2011 International Conference on Computational and Information sciences, ICCIS 2011

作者： Guo, Xiaowei Lin, Yufei Xu, Xinhai Zhang, Xin National Laboratory for Parallel and Distributed Processing School of Computer National University of Defense Technology Changsha Hunan China

ISBN: (纸本)9780769545011

Nowadays, the scale of parallel computer systems is increasing, and simulation technology has become an important tool for performance prediction in the system development process. Task mapping approach is an important aspect affecting the performance of simulation. In this paper, in order to solve the task mapping problem in performance simulation, a task mapping algorithm based on simulated annealing is proposed, and we verified the correctness and effectiveness of the algorithm by experiments. Experimental results show that the algorithm has high efficiency, and can solve the large-scale problem with lower time cost. © 2011 IEEE.

关键词： Simulated annealing

来源：评论

学校读者我要写书评

暂无评论

Error detection by redundant transaction in transactional memory system

Error detection by redundant transaction in transactional me...

引用

IEEE International Conference on Networking, Architecture and Storage

作者： Song, Wei Jia, Jia Peng, Yu-Xing National Laboratory for Parallel and Distributed Processing School of Computer National University of Defense Technology Changsha 410073 China

ISBN: (纸本)9780769545097

This paper addresses the issue of error detection in transactional memory, and proposes a new method of error detection based on redundant transaction (EDRT). This method creates a transaction copy for every transaction, and executes both original transactions and transaction copies on adequate processor cores, and achieves error detection by comparing the execution results. EDRT utilizes the data-versioning mechanism of transactional memory to achieve the acquisition of an approximate minimum error detection comparing data set, and the acquisition is transparent and online. At last, this paper validates the EDRT through 5 test programs, including 4 SPLASH-2 benchmarks. The experimental results show that, the average error detecting cost is about 3.68% relative to the whole program, and it's only about 12.07% relative to the transaction parts of the program. © 2011 IEEE.

关键词： Fault tolerance

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

2011 International Conference on Computational and Information sciences, ICCIS 2011

作者： Wang, Huiquan Xu, Xinhai Zhou, Jing Zhou, Hao Wang, Guibin National Laboratory for Parallel and Distributed Processing School of Computer National University of Defense Technology Changsha Hunan 410073 China

ISBN: (纸本)9780769545011

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%. © 2011 IEEE.

关键词： Power management

来源：评论

学校读者我要写书评

暂无评论

Failure-divergence refinement of compensating communicating processes

Failure-divergence refinement of compensating communicating ...

引用

Lecture Notes in Computer science

作者： Chen, Zhenbang Liu, Zhiming Wang, Ji National Laboratory for Parallel and Distributed Processing Changsha China International Institute for Software Technology United Nations University China

ISBN: (纸本)9783642214363

Compensating CSP (cCSP) extends CSP for specification and verification of long running transactions. The original cCSP is a modest extension to a subset of CSP that does not consider non-deterministic choice, synchronized composition, and recursion. There are a few further extensions. However, it remains a challenge to develop a fixed-point theory of process refinement in cCSP. This paper provides a complete solution to this problem and develops a theory of cCSP, corresponding to the theory of CSP, so that the verification techniques and their tools, such as FDR, can be extended for compensating processes. © 2011 Springer-Verlag.

关键词： Fixed point arithmetic

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

2011 IEEE 3rd International Conference on Communication Software and Networks, ICCSN 2011

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

ISBN: (纸本)9781612844855

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. © 2011 IEEE.

关键词： Information management

来源：评论

学校读者我要写书评

暂无评论

A construction technique of constant degree P2P systems towards efficient complex queries

引用

Jisuanji Yanjiu yu Fazhan/Computer Research and Development 2011年第3期48卷 374-381页

作者： Wang, Xiaohai Peng, Yuxing Li, Dongsheng Key Laboratory of Science and Technology for Parallel and Distributed Processing National University of Defense Technology Changsha 410073 China

Constant degree peer-to-peer (P2P) system is turning into the P2P domain's promising hotspot due to constant degree digraphs having good propertis. However, it is often hard to convert a stardard constant degree digraph to a DHT schema. Thus, most researches focus on DHT's construction and maintenance, while leaving optimization and supporting to complex query behind. Underlying topology affects upper-layers' character a lot. For constant degree P2P topologies, their inherent property makes a constant degree P2P system built using classical technique be poor in the data locality, and unfit for efficient, low-cost complex queries. Aiming at this shortage, a general-purpose construction technique towards efficient complex queries is proposed, which adds an embedding transformation layer between data layer and DHT overlay. In this way, adjacent data are stored in overlay's adjacent peers and the data locality is improved, so that the number of peers referred in complex queries can be minimized with a limited time overhead. To validate this technology, the first constant degree P2P system based on Kautz digraph FissionE is reconstructed as a typical example, which includes re-allocating of resources, query algorithm and locality maintenance strategies. Experimental results show that this construction technique can ensure data locality, reduce query cost and lead to systems' efficiency without changing the underlying DHT layer.

关键词： Peer to peer networks

来源：评论

学校读者我要写书评

暂无评论

parallelization of the training for face detection with transactional memory

Parallelization of the training for face detection with tran...

引用

2011 International Conference on Automation and Robotics, ICAR 2011

作者： Zeng, Kun Key Laboratory and Technology for National Defence of Parallel and Distributed Processing School of Computer National Univiersity of Defense Technology ChangSha Hunan China

ISBN: (纸本)9783642256455

The development of multi-core processor makes the parallelization of traditional sequential algorithms increasingly important. Meanwhile, transactional memory serves a good parallel programming model. This paper takes the advantage of software transactional memory to parallelize the Multi-Exit Asymmetric Adaboost algorithm for face detection. The parallel version is evaluated on three different implementations of software transactional memory. The experiment results show that the transactional memory based parallelization outperforms the traditional lock based approach. A speedup of nearly seven is achieved on a eight-core machine on an eight-core system. © 2011 Springer-Verlag.

关键词： Face recognition

来源：评论

学校读者我要写书评

暂无评论

Coordinate strip-mining and kernel fusion to lower power consumption on GPU

Coordinate strip-mining and kernel fusion to lower power con...

引用

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

ISBN: (纸本)9783981080179

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. © 2011 EDAA.

关键词： Electric power utilization

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

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

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：