检索结果-内蒙古大学图书馆

7th international Conference on High-Performance computing and Networking, HPCN Europe 1999

作者： Antony Louis Piriyakumar, D. Levi, Paul Siva Ram Murthy, C. Institute of Parallel and Distributed High-Performance Systems University of Stuttgart Stuttgart70565 Germany Department of Computer Science and Engineering Indian Institute of Technology Chennai600 036 India

ISBN: (纸本)3540658211

the problem of optimal compile-time multiprocessor scheduling of iterative data-flow programs with feedback (delay elements) is addressed in this paper, unlike the earlier studies assumed the availability of a large number of processors and complete interconnection among them along with the interprocessor communication (IPC) to be non-negligible to be more realistic. We first explain the effects of including IPC in non-overlapped, overlapped, fully-static, and cyclo-static multiprocessor schedules with LMS filter as a realistic example. the effect of IPC in the rate-optimal schedules with the transformation techniques viz. unfolding and retiming in scheduling data-flow programs with optimal unfolding is discussed with an example. We then propose an algorithm, based on the well-known A* algorithm, for optimal scheduling of data-flow programs onto multiprocessors, which uses only minimum number of processors. To alleviate the impediments of large requirements of memory space and CPU time for the optimal scheduling algorithm, we present an effective technique, branch join path isomorphism (BJP) which relies on our previously defined processor isomorphism, task isomorphism, and node isomorphism apart from the lower bound theory and upper bound on the completion time. the schedules produced by our algorithm are superior to those obtained by the earlier algorithms despite considering IPC as non-negligible and also not completely connected multiprocessor systems. © Springer-Verlag Berlin Heidelberg 1999.

关键词： Scheduling

来源：评论

学校读者我要写书评

暂无评论

A Large-Scale Elastic Environment for Scientific computing

A Large-Scale Elastic Environment for Scientific Computing

引用

7th international Conference on Software Paradigm Trends, ICSOFT 2012

作者： Marshall, Paul Tufo, Henry Keahey, Kate LaBissoniere, David Woitaszek, Matthew Department of Computer Science University of Colorado Boulder CO United States CSS National Center for Atmospheric Research Boulder CO United States Computation Institute University of Chicago Chicago IL United States MCS Argonne National Laboratory Argonne IL United States

ISBN: (纸本)9783642454035

the relatively recent introduction of infrastructure-as-a-service (IaaS) clouds, such as Amazon Elastic Compute Cloud (EC2), provide users with the ability to deploy custom software stacks in virtual machines (VMs) across different cloud providers. Users can leverage IaaS clouds to create elastic environments that outsource compute and storage as needed. Additionally, these environments can adapt dynamically to demand, scaling up as demand increases and scaling down as demand decreases. In this paper, we present a large-scale elastic environment that extends cluster resources managers (e.g. Torque) with IaaS resources. Our solution integrates with an open-source elastic manager, the Elastic Processing Unit (EPU), and includes the ability to periodically recontextualize the environment with a light-weight REST-based recontextualization broker. We deploy the Gluster file system to provide a shared file system for all nodes in the environment. though our implementation currently only supports Torque, we also thoroughly discuss how our architecture can interface with different workflows, including Hadoop's MapReduce workflows and Condor's match-making and high-throughput capabilities. For evaluation, we demonstrate the ability to recontextualize 256-node environments within one second of the recontextualization period, scale to over 475 nodes in less than 15 minutes, and support parallel IO from distributed nodes. © Springer-Verlag Berlin Heidelberg 2013.

关键词： Managers

来源：评论

学校读者我要写书评

暂无评论

Evolving Big Data Stream Classification with MapReduce

Evolving Big Data Stream Classification with MapReduce

引用

IEEE international Conference on Cloud computing, CLOUD

作者： Ahsanul Haque Brandon Parker Latifur Khan Bhavani thuraisingham Department of computer Science The University of Texas at Dallas Richardson USA

Big Data Stream mining has some inherent challenges which are not present in traditional data mining. Not only Big Data Stream receives large volume of data continuously, but also it may have different types of features. Moreover, the concepts and features tend to evolve throughout the stream. Traditional data mining techniques are not sufficient to address these challenges. In our current work, we have designed a multi-tiered ensemble based method HSMiner to address aforementioned challenges to label instances in an evolving Big Data Stream. However, this method requires building large number of AdaBoost ensembles for each of the numeric features after receiving each new data chunk which is very costly. thus, HSMiner may face scalability issue in case of classifying Big Data Stream. To address this problem, we propose three approaches to build these large number of AdaBoost ensembles using MapReduce based parallelism. We compare each of these approaches from different aspects of design. We also empirically show that, these approaches are very useful for our base method to achieve significant scalability and speedup.

关键词： Big data Indexes Data mining Scalability Sorting distributed databases parallel processing

来源：评论

学校读者我要写书评

暂无评论

XQ-Index: A distributed Spatial Index for Cloud Storage Platforms

XQ-Index: A Distributed Spatial Index for Cloud Storage Plat...

引用

international Conference on Agro-Geoinformatics

作者： Yulong Hu Lun Luo Lin Yin China Transport Telecommunications & Information Center Beijing China

Currently the cloud storage platforms (CSPs) based on key-value model only support simple keyword-based queries and can't answer complex queries efficiently due to lacking of efficient index techniques. In this paper we propose a novel distributed spatial index (XQ-index) for CSPs to solve this problem. In the process of index building, we proposed a new space partitioning method based on an improved Quad-tree algorithm that can make full use of MapReduce parallel computing model. Our approach can process typical spatial queries including range queries and k-NN queries efficiently based on MapReduce framework. Besides, frequent change of data on big number of machines makes the index maintenance a challenging problem, and to cope with this problem we proposed an index update strategy that can effectively update the index structure. In the experiments, we implement the XQ-index on Hadoop platform to test the effect of parameter on the efficiency of index, and also test the efficiency of index building, querying and updating under data of different sizes. Experiments show that the XQ-index is quite efficient and independent from the underlying infrastructure and can be easily carried over for implementation on various CSPs.

关键词： Spatial databases Partitioning algorithms Buildings distributed databases Task analysis Spatial indexes

来源：评论

学校读者我要写书评

暂无评论

the Application of Spark in the Power Grid Intelligent Decision Analysis Platform

The Application of Spark in the Power Grid Intelligent Decis...

引用

international Conference on Intelligent Human-Machine Systems and Cybernetics, IHMSC

作者： Wei Li Qian Niu Weijia Zhang Jin Pang School of Control And Computer Engineering North China Electric Power University Beijing China

ISBN: (纸本)9781479986477

Intelligent Decision Analysis Platform is an advanced application of decision and analysis based on the construction of "State Grid 186" in China. the platform is committed to achieve the goal of all kinds of business managements and developments. It mainly focuses on providing scientific decisions by data mining and predicting after status data being analyzed. With the auxiliary support of the platform, power enterprises are capable of making better strategies and increasing revenue eventually. According to the characteristics of smart grid monitoring and the requirements of management efficiently as well as the demand of reliable storage of massive grid data, this paper puts forward the decision and analysis platform based on the parallel processing framework which is called Spark and the fault-tolerant abstraction for in-memory cluster computing embedded in Spark as it is known as Resilient distributed Datasets (RDDs). this paper gives detailed analysis of the feasibility and the advantages of the new method, as well as some unsolved problems.

关键词： Sparks Reliability Servers Decision analysis Smart grids parallel processing

来源：评论

学校读者我要写书评

暂无评论

Sensitivity Analysis of Biomolecular Simulations using Symbolic Models

Sensitivity Analysis of Biomolecular Simulations using Symbo...

引用

IEEE symposium on Bioinformatics and Bioengineering (BIBE)

作者： Sadaf R. Alam Nikhil Bhatia Jeffrey S. Vetter Oak Ridge National Laboratory Oak Ridge USA

Performance and scaling of biomolecular simulations frameworks largely depends on not only the workload characteristics of the simulations but also the design of underlying processor architecture and interconnection networks. Because construction of Teraflops and Petaflops scale prototype systems for evaluation alone is impractical and cost-prohibitive, architects use analytical models of workloads and architecture simulators to guide their design decisions and tradeoffs. To address the problem of providing scalable yet precise input for network simulators, we have developed a technique to model symbolically the communication patterns of production-level scientific applications to study workload growth rates and to carry out sensitivity analysis. We apply our symbolic modeling scheme to the particle mesh ewald (PME) implementation in the sander package of the AMBER framework and demonstrate how the increase in computation, memory and communication requirements impact the performance and scaling of the PME method on the next-generation massively-parallel systems.

关键词： Sensitivity analysis Analytical models Computational modeling Predictive models Packaging Large-scale systems Virtual prototyping High performance computing Performance analysis Laboratories

来源：评论

学校读者我要写书评

暂无评论

Methodology for Evaluating DNA Pattern Searching Algorithms on Multiprocessor

Methodology for Evaluating DNA Pattern Searching Algorithms ...

引用

IEEE symposium on Bioinformatics and Bioengineering (BIBE)

作者： Benfano Soewito Ning Weng Department of Electrical and Computer Engineering Southem Illinois University Carbondale IL USA

Pattern matching has been one of the major operations in modern bioengineering especially in Bioinformatics. Prior work on this area have focus on either pursuing mathematically efficient matching algorithms or hardwired approach. As multicore processor are becoming mainstream, developers need to determine how to take advantage of multicore technology for pattern matching. In this paper, we propose a methodology to evaluate pattern search algorithms for DNA on Multiprocessor. Our evaluation methodology is an automatic simulation framework. Starting from a uniprocessor profiling, the framework constructs task graphs for string matching algorithms. then task graphs are mapped onto multiprocessor. the system's performance is determined by the analytical performance model. With this framework, we can evaluate the performance of different algorithms on multiprocessor. Our case studies show that finite automaton based (Aho-Corasick) is more efficient than shift table based algorithms (SFKSearch and Wu-Manber) on uniprocessor, however, Wu-Manber is 3 times efficient than Aho-Corasick on multiprocessor due to its inherent parallelism.

关键词： Bioinformatics Multicore processing Pattern matching Databases Sequences parallel processing Computer networks DNA computing Genomics Application specific integrated circuits

来源：评论

学校读者我要写书评

暂无评论

Transparent adaptive parallelism on NOWs using OpenMP

Proceedings of the ACM SIGPLAN Symposium on Principles and P...

引用

Proceedings of the ACM SIGPLAN symposium on Principles and Practice of parallel Programming, PPOPP 1999年 96-106页

作者： Scherer, Alex Lu, Honghui Gross, thomas Zwaenepoel, Willy ETH Zurich Zurich Switzerland

We present a system that allows OpenMP programs to execute on a network of workstations with a variable number of nodes. the ability to adapt to a variable number of nodes allows a program to take advantage of additional nodes that become available after it starts execution, or to gracefully scale down when the number of available nodes is reduced. We demonstrate that the cost of adaptation is modest;the system allows a program to adapt at a moderate rate without much performance loss. Two ideas underlie the efficiency of our design. First, we recognize that OpenMP programs exhibit convenient adaptation points during their execution, points at which the cost of adaptation can be much reduced. Second, by allowing a process a certain grace period before it must leave a node, we insure that most adaptations can occur at these adaptation points, and thus at low cost. Migration of a process, a much more expensive method for providing adaptivity, is used only as a back-up solution, when the process cannot reach an adaptation point within the grace period. Our implementation consists of an OpenMP pre-processor that generates TreadMarks distributed shared memory (DSM) programs, and a version of TreadMarks modified to adapt to a variable number of nodes. Using a DSM as the underlying substrate facilitates the data (re-)distribution necessary after an adaptation.

关键词： parallel processing systems

来源：评论

学校读者我要写书评

暂无评论

Cloud computing and Robotics for Disaster Management

Cloud Computing and Robotics for Disaster Management

引用

international Conference on Intelligent Systems, Modelling and Simulation (ISMS)

作者： Nitesh Jangid Bindu Sharma Information Technology Department Green Research IT Solutions Pvt. Ltd. Jaipur Rajasthan India Amity School of Fashion Technology Amity University Jaipur Jaipur Rajasthan India

ISBN: (纸本)9781509006663

Nowadays, IT community is experiencing great shift in computing and information storage infrastructures by using powerful, flexible and reliable alternative of cloud computing. the power of cloud computing may also be realized for mankind if some dedicated disaster management clouds will be developed at various countries cooperating each other on some common standards. the experimentation and deployment of cloud computing by governments of various countries for mankind may be the justified use of IT at social level. It is possible to realize a real-time disaster management cloud where applications in cloud will respond within a specified time frame. If a Real-Time Cloud (RTC) is available then for intelligent machines like robots the complex processing may be done on RTC via request and response model. the complex processing is more desirable as level of intelligence increases in robots towards humans even more. therefore, it may be possible to manage disaster sites more efficiently with more intelligent cloud robots without great lose of human lives waiting for various assistance at disaster site. Real-time garbage collector, real-time specification for Java, multicore CPU architecture with network-on-chip, parallel algorithms, distributed algorithms, high performance database systems, high performance web servers and gigabit networking can be used to develop real-time applications in cloud.

关键词： Cloud computing Real-time systems Robots Java Multicore processing Hardware Operating systems

来源：评论

学校读者我要写书评

暂无评论

Minimizing Update Bits of NVM-Based Main Memory Using Bit Flipping and Cyclic Shifting

Minimizing Update Bits of NVM-Based Main Memory Using Bit Fl...

引用

IEEE international Conference on High Performance computing and Communications (HPCC)

作者： Wei Dong Xin Li Yanbin Li Meikang Qiu Lei Dou Lei Ju Zhiping Jia School of Computer Science and Technology Shandong University Jinan P.R. China Department of Computer Science Pace University New York NY USA Jinan Vocational College P.R. China

ISBN: (纸本)9781479989386

Non-Volatile Memory (NVM) has many advantages, such as non-volatility, high density, shock-resistivity, and low leakage power, and disadvantages, such as the relatively high write energy and limited write endurance. It also has the feature of writing more slowly than reading. We design a read-modify-write strategy called MinFS for NVM-based main memory to reduce the number of update bits and prolong NVM lifetime. the MinFS sets one flip bit and multiple shift bits for one word to re-encode the word. the flip bit indicates whether the word has been flipped or not, while the shift bits show how many bits the word has been circularly shifted. For a write request, the MinFS first computes all possible encoded data of new data with different flip or shift values and reads out old data. then it calculates hamming distances between old data and each encoded data. the encoded data with the minimal distance is found and written into NVM. In order to reduce the overhead, we design a hardware unit to support parallel computation of multiple hamming distances. Experimental results show that the MinFS scheme can reduce the number of update bits by 16.9% and 14.6% compared with Flip-N-Write and Min-Shift on average under real benchmarks, respectively.

关键词： Conferences High performance computing Cyberspace Safety Security Cascading style sheets Embedded software

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：