A5100772689- Distributed and Parallel Computing Systems
- Scientific Computing and Data Management
- Metaheuristic Optimization Algorithms Research
- Caching and Content Delivery
- Peer-to-Peer Network Technologies
- Optical Network Technologies
- Video Coding and Compression Technologies
- Advanced Multi-Objective Optimization Algorithms
- Neural Networks and Reservoir Computing
- Image and Video Quality Assessment
- Service-Oriented Architecture and Web Services
- Parallel Computing and Optimization Techniques
- Advanced Algorithms and Applications
- Advanced Data Storage Technologies
- Neural Networks and Applications
- Computer Graphics and Visualization Techniques
- Advanced Sensor and Control Systems
- Multimedia Communication and Technology
- Process Optimization and Integration
- Web Data Mining and Analysis
- Complex Network Analysis Techniques
- Advanced Vision and Imaging
- Advanced Adaptive Filtering Techniques
- Advanced Image Processing Techniques
- Evolutionary Algorithms and Applications
University of Electronic Science and Technology of China
2009-2024
Neusoft (China)
2024
Huaiyin Normal University
2018-2019
Shanghai Industrial Technology Institute
2019
Jiangnan University
2018-2019
Nanjing Tech University
2018
Tencent (China)
2012
Baidu (China)
2012
PLA Information Engineering University
2011
Shandong University of Science and Technology
2010
P2P file downloading and streaming have already become very popular Internet applications. These systems dramatically reduce the server loading, provide a platform for scalable content distribution, as long there is interest content. P2P-based video-on-demand (P2P-VoD) new challenge technology. Unlike live content, P2P-VoD has less synchrony in users sharing video therefore it much more difficult to alleviate loading at same time maintaining performance. To compensate, small storage...
Despite the increasing popularity, Internet video streaming to mobile devices is still challenging. In particular, there has been a format and resolution "gap" between videos devices, so users have high demand on transcoding facilitate their specific devices. However, as computation-intensive work, greatly challenged by limited battery capacity of this paper we propose implement "Cloud Transcoder", which utilizes an intermediate cloud platform bridge via its special practical designs....
This paper describes a process for the semi-automatic conversion of numerical and scientific routines written in C programming language into Triana-based computational services that can be used within distributed service-oriented architecture such as being adopted Grid computing. involves two separate but related tools, JACAW MEDLI. is wrapper tool based on Java Native Interface (JNI) automatically generate interface files any routine, or library routines. The MEDLI then to assist user...
This paper describes how to extend a Jini system into an OGSA-compliant infrastructure for Grid computing by introducing Web service techniques the system. The Jini-based Service-oriented Architecture (JISGA) is presented, functionality of described, and main services components are discussed. shows present as service, application composed interacting can be described using workflow description language (SWFL). conversion composite specified in SWFL document executable Java code, its...
Content distribution on today's Internet operates primarily in two modes: server-based and peer-to-peer (P2P). To leverage the advantages of both modes while circumventing their key limitations, a third mode: peer-to-server/peer (P2SP) has emerged recent years. Although P2SP can provide efficient hybrid server-P2P content distribution, generally works closed manner by only utilizing its private owned servers to accelerate organized peer swarms. Consequently, still limitations abundance...
This paper presents a variant of multiscale quantum harmonic oscillator algorithm for multimodal optimization named MQHOA-MMO. MQHOA-MMO has only two main iterative processes: process and process. In the iterations, does one thing: sampling according to wave function at different scales. A set benchmark test functions including some challenging are used performance Experimental results demonstrate good in solving problems. For 12 functions, all global peaks can be found without being trapped...
As a powerful tool in optimization, particle swarm optimizers have been widely applied to many different optimization areas and drawn much attention. However, for large-scale problems, the algorithms exhibit poor ability pursue satisfactory results due lack of diversity maintenance. In this paper, an adaptive multi-swarm optimizer is proposed, which adaptively divides into several sub-swarms competition mechanism employed select exemplars. way, on one hand, exemplars increases, helps...
The 0-1 knapsack problem is a typical discrete combinatorial optimization with numerous applications. In this paper, binary multi-scale quantum harmonic oscillator algorithm (BMQHOA) genetic operator proposed for solving problem. framework of BMQHOA consisted three nested phases including energy level stablization, decline and scale adjustment. BMQHOA, the number different bits between solutions defined as distance to map continuous search space into space. Repair greedy strategy adopted in...
Given a data layout of large walkthrough scene, we present novel and simple spatial hierarchy on the disk-pages that has notable advantages over conventional scene geometry. Assume each disk-page consists set triangles whose bounding boxes are computed. A space is constructed, not with given but disk-pages. The leaf nodes spatial-hierarchy refer directly to page numbers pages box it contains. We call this as hierarchy. also propose self-contained format would suit structure well. Further,...
A multi-scale quantum harmonic oscillator algorithm (MQHOA) is a population-based proposed recently. It utilizes the wave function to locate global optimum of numerical optimization problem. As MQHOA employs elitism replace worst particle in each iteration cycle, it reduces one particles run, which will cripple diversity population and slow down convergence speed. Therefore, be easily trapped into local optima. In this paper, we suggest new with truncated mean stabilization (TS-MQHOA) policy...
An approach to dynamic workflow management and optimisation using near‐realtime performance data is presented. Strategies are discussed for choosing an optimal service (based on user‐specified criteria) from several semantically equivalent Web services. Such may involve finding "similar" services, by first pruning the set of discovered services based metadata, subsequently selecting data. The current implementation prototype framework described, demonstrated with a simple workflow....
This paper presents MPFL, an extension to SWFL that provides a message passing flow language for describing parallel scientific applications with composed from Web services. The background developing the is discussed. Then focuses on four main issues: communicators, point-to-point communications, collective and application topologies. semantics of its model representing are also described.
In this paper efficient hardware architecture for deblocking filter in H.264 is presented. A parallel filtering order proposed without violating the standard, and implemented pipelined dataptah. With vertical edges horizontal can be processed simutaneously, efficiency improved. The memory arranged orgnized carefully, data reused efficiently reference reduced. verified on FPGA, it takes only 116 clock cycle to process a 4:2:0 macroblock.
Grid computing is converging on a service-oriented architecture in which applications are composed from geographically dispersed, interacting Web services, and expressed workflow description language. The discovery of resources key importance to the effective and flexible use such an architecture. This paper describes work date Workflow Optimisation Services for e-Science Applications (WOSE) project investigating techniques discovering services optimising execution service composite workflows.
With the deep integration of information and communication, people's demand for more reliable network topology structure is becoming higher higher. Multi-objective model switched industrial Ethernet a recently proposed mode, which has been proved to have flexible high communication efficiency. However, fast non-dominated sorting algorithm II (NSGA-II), used tackle mode above, drawback that multiple repeatedly selected parent individuals may result in low diversity offspring population. To...
In this paper, we propose a fast mode decision algorithm for H.264/AVC. It is based on the spatio-temporal coherency of local neighborhood including and around current block. We first build histograms block co-located in reference frame, respectively. If difference between two small, use large-block-size modes. Otherwise, subdivide into four equal-sized sub-blocks estimate motion vector (MV) each. general, if there high degree those MVs, modes otherwise small-block-size addition, number...
Peer-assisted offline download service provides a new Internet content delivery approach, in which the system coordinate dedicated servers and peers to finish file download. This has attracted large number of subscribers. The related research, however, is still missing. paper presents our preliminary measurement analysis users' behavior QQXuanfeng, one most popular peer-assisted downloading applications China. Based on set real- world user data, we validate performance system, cluster users...
An enhanced non-dominated sorting genetic algorithm (NSGA-II) was applied to carry out the integer optimized design in this article. The steady simulation established based on reaction kinetics analysis using Aspen Plus, and feasible region of each variable calculated through sensitive analysis. Aiming at total cost TAC as objective function, NSGA-II used directly obtain Pareto optimal solutions under constraints variables. results showed that can effectively reduce TAC.