检索结果-内蒙古大学图书馆

18th IEEE International conference on high performance Computing and Communications (HPCC) / 14th IEEE International conference on Smart City (Smart City) / 2nd IEEE International conference on Data Science and Systems (DSS)

作者： Zhao, Meiting Liu, Rui Liu, Yi Song, Kaida Qian, Depei Beihang Univ Sinogerman Joint Software Inst Beijing Peoples R China State Key Lab Math Engn & Adv Comp Wuxi Peoples R China

ISBN: (纸本)9781509042975

Many image processing algorithms have been parallelized successfully on many-core processors, such as GPU and Intel Xeon Phi. In this paper, we choose the Sunway many-core processor SW26010, which is a new processor designed and made in China that constitutes the current NO. 1 supercomputer Sunway TaihuLight. This paper firstly introduces the architecture of Sunway SW26010 processor and two representative image processing algorithms: local binary pattern (LBP) and histogram of oriented gradient (HOG). Furthermore we propose a method of parallel implementation, and the experimental results of this method show that the speedup can be up to 170 for LBP and 33 for HOG. Then two optimized methods are brought forward based on this parallel implementation, including the optimization of program and parallel design. We optimize the program by using the method that combined step transmission and software prefetching. From the experiment results of the first optimization we can know that the maximum speedup can reach 310 for LBP with processing high-resolution images and 83 for HOG. Then we optimize the parallel design by using a coarse grained parallel method, and the experimental results show that the speedup can be up to 370 for LBP and 95 for HOG when processing low-resolution images. Finally, we investigate the scalability of our parallelism on the Sunway TaihuLight with different number of processor nodes, and the experiment results prove that the two algorithms' parallel design and implementation have better expansibility.

关键词： image processing Sunway SW26010 processor the Sunway TaihuLight supercomputer parallel implementation optimization

来源：评论

学校读者我要写书评

暂无评论

Dynamic Approach to Scheduling Reconfigurable Scientific Workflows in Heterogeneous HPC Environments 10

Dynamic Approach to Scheduling Reconfigurable Scientific Wor...

引用

10th International conference on Complex, Intelligent, and software Intensive Systems (CISIS)

作者： Cheptsov, Alexey High Performance Comp Ctr Stuttgart HLRS Stuttgart Germany

ISBN: (纸本)9781509009879

high performance Computing infrastructures are getting increasingly heterogeneous, which offers opportunities to the applications in terms of performance improvement. However, the "hardware-concise" heterogeneous resource allocation requires a deep knowledge of the scheduled applications' characteristics. The challenge is getting more difficult if the scheduler has to balance the optimization policies between the application-and the infrastructure-specific policies, e.g. the overall energy consumption, resource utilization, etc. We introduce a heuristic-based approach to adaptive scheduling, enabled by in-depths monitoring technologies. Our solution is complementary to the native schedulers like Torque or Maui and is open per design and thus can be seamlessly integrated into the core scheduling algorithms.

关键词： scheduling performance workflow monitoring heuristics DreamCloud

来源：评论

学校读者我要写书评

暂无评论

Joint Route-Server Load Balancing in software Defined Networks using Ant Colony optimization 14

Joint Route-Server Load Balancing in Software Defined Networ...

引用

International Workshop on Parallel optimization using/for Multi and Many-Core high performance Computing (POMCO) in conjunction with 14th International conference on high performance Computing and Simulation (HPCS)

作者： Sathyanarayana, Sushma Moh, Melody San Jose State Univ Dept Comp Sci San Jose CA 95192 USA

ISBN: (纸本)9781509020881

Managing resource utilization can greatly improve system and network performance. Load balance (LB) is one important means for controlling resources. Traditional LB methods are usually inefficient, inflexible since there are no effective ways to obtain network statistics from each network device. This can be overcome by adopting the software Defined Networking (SDN) approach. We present a dynamic LB for SDN, which aims to load balancing both the servers and the paths leading to the servers. To choose the best server, we use the least-loaded server LB policy. To select the best path to reach the chosen server, we adopt the Ant Colony System (ACS) algorithm, a successful variation of the nature-inspired ant colony optimization method. The proposed work therefore implements the joint ACS routing and dynamic server LB algorithm as a LB module in the SDN controller. This algorithm uses server-load and network statistics collected by the controller to find both the best server and the best path for network flows. As a proof of concept, the proposed algorithm is implemented as a module in the OpenDayLight controller, and experiments are carried over Mininet, an OpenFlow network emulator. The experimental results demonstrate that the proposed algorithm performs significantly well as compared to two other benchmark algorithms, achieving both higher network throughput and lower delay. This work would serve as an illustration of adopting SDN for system and network performance management, as well as a good example of using nature-inspired cooperative methods for SDN.

关键词： Ant Colony System software Defined Networking OpenDayLight OpenFlow Load Balancing Routing

来源：评论

学校读者我要写书评

暂无评论

Credibilistic clustering algorithms via alternating cluster estimation

引用

JOURNAL OF INTELLIGENT MANUFACTURING 2017年第3期28卷 727-738页

作者： Zhou, Jian Wang, Qina Hung, Chih-Cheng Yang, Fan Shanghai Univ Sch Management Shanghai 200444 Peoples R China Anyang Normal Univ Anyang 455000 Peoples R China Southern Polytech State Univ Sch Comp & Software Engn Marietta GA 30060 USA

Credibilistic clustering is a new clustering method using the credibility measure in fuzzy clustering. Zhou et al. (2014) presented the clustering model of credibilistic clustering together with a credibilistic clustering algorithm for solving the optimization model. In this paper, a further investigation on credibilistic clustering is made. Within the solution architecture of alternating cluster estimation, a family of general credibilistic clustering algorithms are designed for solving the credibilistic clustering model. Moreover, a new credibilistic clustering algorithm is recommended for the real applications. Numerical examples based on randomly generated data sets and real data sets are presented to illustrate the performance and effectiveness of the credibilistic clustering algorithms from different aspects. Results comparing with the fuzzy -means algorithm and the possibilistic clustering algorithms show that the proposed credibilistic clustering algorithms can survive from the coincident problem and the noisy environments, and provide the clustering results with high overall accuracy.

关键词： Fuzzy clustering Credibilistic clustering Alternating cluster estimation Credibility measure

来源：评论

学校读者我要写书评

暂无评论

Predictive Adaptive Control of nonlinear Multivariable Systems Using Fuzzy CMAC

Predictive Adaptive Control of Nonlinear Multivariable Syste...

引用

International conference on Computational Intelligence for Modelling, Control and Automation

作者： Mohajeri, Kamran Teshnehlab, Mohammad Nekoui, Mohammad Ali Moaveni, Bijan Islamic Azad Univ Dept Elect Engn Sci & Res Unit Tehran Iran

ISBN: (纸本)9781424433292

CMAC computational model that is based on the cerebellum structure is known as a Neural Network with high computation and learning speed. Fuzzy CMAC, by introducing fuzzy reasoning to CMAC, converts it from a black box to a white box whose performance can be interpreted using fuzzy rules. Fuzzy CMAC compared to CMAC has higher approximation and modeling capability and allows for model gradient computation necessary for applications such as predictive control. This paper presents two nonlinear predictive control algorithms based on fuzzy CMAC. The first algorithm uses a numerical method for optimization computations and the second one uses model gradient. Furthermore, their performance is shown by applying them to control a nonlinear multivariable system.

关键词： Numerical methods

来源：评论

学校读者我要写书评

暂无评论

A resource allocation model with cost-performance ratio in Data Grid

A resource allocation model with cost-performance ratio in D...

引用

8th ACIS International conference on software Engineering, Artificial Intelligence, Networking and Parallel/Distributed Computing/3rd ACIS International Workshop on Self-Assembling Wireless Networks

作者： Zhao, Xiangang Xu, Liutong Wang, Bai Beijing Univ Posts & Telecommun Sch Comp Sci & Technol Beijing Key Lab Intelligent Telecommun Software & Beijing 100876 Peoples R China

ISBN: (纸本)0769529097

Resource allocation for data transfer is a fundamental issue for achieving high performance in Data Grid environments. In this paper, we survey the existing researches on allocation problems in Grid environment and propose a resource allocation model based on the cost-performance ratio for commerce environments. This ratio takes both price and quality of data resource into consideration at the same time. According to the optimization objective we define two types of ratios: Price-aware ratio and Quality-aware ratio. They are suitable for the environments where allocations put more emphasis on quality or price of resource respectively. Based on the maximal cost-efficiency, we formulize the allocation optimization problem and present two theorems and a deduction about its solutions. The corresponding proofs are also given in the paper. Finally, we present two algorithms to allocate and re-allocate data resources. The analysis shows that the algorithms can satisfy our requirements.

关键词： Resource allocation

来源：评论

学校读者我要写书评

暂无评论

Study on roundness error evaluation with least-squares method based on nonlinear optimization

引用

2013 International conference on Advances in Materials Science and Manufacturing Technology, AMSMT 2013

作者： Liu, Zhen Tao Yang, Jian Xin Zhao, Ben Beijing University of Technology Beijing 100124 China School of Mechanical Engineering Tsinghua University Beijing 100084 China Xian XD Switchgear Electric Co. Ltd Shaanxi 710077 China

ISBN: (纸本)9783037857984

Roundness error evaluation software is developed based on two-dimensional circle fitting with least-squares method based on nonlinear optimization with constraints. The local derivative-free optimization algorithms of NLopt can solve nonlinear constraint problems by combining with augmented Lagrangian algorithm. The fitting precision and convergence time of each algorithm are analyzed by calculating the fitting results with same test data to find its advantages and disadvantages. It is shown that each algorithm has different behaviors from others on performance and stability. This work provides a good basis for choosing the appropriate algorithm for roundness error evaluation. © (2013) Trans Tech Publications, Switzerland.

关键词： Constrained optimization

来源：评论

学校读者我要写书评

暂无评论

nonlinear Adaptive Channel Equalization using Genetic algorithms 12

Nonlinear Adaptive Channel Equalization using Genetic Algori...

引用

12th IEEE International New Circuits and Systems conference (IEEE NEWCAS)

作者： Merabti, Hocine Massicotte, Daniel Univ Quebec Trois Rivieres Dept Elect & Comp Engn Lab Signaux & Syst Integres Trois Rivieres PQ Canada

ISBN: (纸本)9781479948857

nonlinear adaptive channel equalization is a well-documented problem. Equalizers based on the complex decision feedback recurrent neural network (CDFRNN) have been intensively studied to address this problem. However, when trained with conventional training algorithms like the real time recurrent learning (RTRL) technique, the equalizer suffers from low convergence speed, requiring very long training sequence to achieve proper performance. In this work, we propose a new approach to equalize nonlinear channels using genetic algorithms. The proposed Volterra decision feedback genetic algorithm (VDFGA) uses a genetic optimization strategy to estimate Volterra kernels in order to model the inverse of the channel response. Simulation results show very high convergence speed, which allowed to achieve interesting bit error rate (BER) using relatively short training symbols, when considering only 8-bits long coded weights.

关键词： Biological cells Bit error rate Convergence Equalizers Genetic algorithms Sociology Training Genetic Algorithm (GA) Equalizers Bit error rate channel equalization cytology Sociology Converge Functional training decision feedback nonlinear channel Training training sequence

来源：评论

学校读者我要写书评

暂无评论

An improved intelligent calibration method for vortex flowmeter

An improved intelligent calibration method for vortex flowme...

引用

2007 American Control conference, ACC

作者： Yan, Yi Wu, Huifeng Institute of Intelligent and Software Technology Hangzhou Dianzi University Hangzhou Zhejiang 310018 China

ISBN: (纸本)1424409888

Because the characteristic curve of vortex flowmeter is nonlinear, the error of traditional calibration method, which didn't take the nonlinear character into account, is large. And the flowmeter calibrated by the traditional method could not measure the flux accurately, especially at the high flux area. In order to resolve this problem a new calibration method based on intelligent optimization algorithms is presented in the article. It uses the improved BP neural network to model the characteristic curve of vortex flowmeter. And then it applies the genetic algorithms to seek two additional optimum calibration points intelligently at the intervals where the curve are nonlinear obviously. At last the vortex flowmeter was calibrated at the new calibration points. The results of analog simulation indicate that the calibration error and measurement error were decreased obviously by using the intelligent calibration method. © 2007 IEEE.

关键词： Flowmeters

来源：评论

学校读者我要写书评

暂无评论

Bacterial-inspired algorithms for solving constrained optimization problems

引用

NEUROCOMPUTING 2015年 148卷 54-62页

作者： Niu, Ben Wang, Jingwen Wang, Hong Shenzhen Univ Coll Management Shenzhen 518060 Peoples R China Chinese Acad Sci Hefei Inst Intelligent Machines Hefei 230031 Peoples R China Hong Kong Polytech Univ Dept Ind & Syst Engn Hong Kong Hong Kong Peoples R China

Bio-inspired optimization techniques using analogy of swarming principles and social behavior in nature have been adopted to solve a variety of problems. In this paper Bacterial Foraging optimization (BFO) is employed to achieve high-quality solutions to the constrained optimization problems. However, the chemotaxis step was set as only a constant in the original BFO, where no mechanism could keep the balance between global search and local search. To further improve the performance of the original BFO, we also come up with two modified BFOs, i.e. BFO with linear decreasing chemotaxis step (BFO-LDC) and BFO with non-linear decreasing chemotaxis step (BFO-NDC). In order to illustrate the efficiency of the proposed method (BFO-LDC and BFO-NDC), six well-known constrained benchmark problems from the optimization literature were selected as testing functions. The experimental results demonstrated that the modified BFOs are of greater efficiency in the speed of convergence as well as fine tune the search in the multidimensional space, and they can be used as a general approach for most nonlinear optimization problems with inequity constrains. (C) 2014 Published by Elsevier B.V.

关键词： Constrained handling optimization Bacterial foraging Linear decreasing chemotaxis Non-linear decreasing chemotaxis

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：