Scheduled System Maintenance:
Some services will be unavailable Sunday, March 29th through Monday, March 30th. We apologize for the inconvenience.
By Topic

Parallel and Distributed Computing, Applications and Technologies, 2003. PDCAT'2003. Proceedings of the Fourth International Conference on

Date 27-29 Aug. 2003

Filter Results

Displaying Results 1 - 25 of 213
  • Distributed content-aware request distribution in cluster-based Web servers

    Publication Year: 2003 , Page(s): 99 - 103
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (454 KB) |  | HTML iconHTML  

    While content-aware distribution policies getting more popular in cluster-based Web systems, they make the dispatching node a bottleneck. To achieve scalable server performance, we present a completely distributed architecture in which all hosts participate in request dispatching. To cope with the architecture, we propose a distributed dispatching policy named DWARD. With DWARD, all the server nodes may participate in request dispatching on the basis of local access pattern. In addition, the DWARD policy inherits several desirable features of the centralized algorithms, including load balance, high locality and less forwarding overhead. Finally, a testbed is implemented on the basis of Linux kernel to benchmark various dispatching algorithms. The performance results show that DWARD can achieve favorable throughput compared with state-of-the-art dispatching policies. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Congestion control in ATM networks using additive-multiplicative fuzzy neural network

    Publication Year: 2003 , Page(s): 306 - 310
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (332 KB) |  | HTML iconHTML  

    Based on additive-multiplicative fuzzy neural network (AMFNN), a novel congestion control scheme for ATM network is presented. This scheme uses AMFNN to accurately predict the traffic arrival patterns. The predicted traffic with the current queue information of the buffer can be used as a measure of congestion. When the congestion level is reached, a control signal is generated to throttle the input arrival rate. Here, the AMFNN model and its learning algorithm are discussed. The simulation results show that this method can improve the congestion processing capability in real time, and raise the utilization of the network resource at the same time. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Comments on Shao's blind signature scheme

    Publication Year: 2003 , Page(s): 940 - 941
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (209 KB) |  | HTML iconHTML  

    A general attack on Shao's blind signature scheme [Pecora L. M et al. (1990)] is presented. It is pointed out that, in Shao's scheme, the signer can trace the identity of the signature requester, which shows that the Shao's scheme is not really blind. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • An architecture of Internet based data processing based on multicast and anycast protocols

    Publication Year: 2003 , Page(s): 104 - 110
    Cited by:  Papers (1)
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (546 KB) |  | HTML iconHTML  

    Most of the current Web-based application systems suffer from poor performance and costly heterogeneous accessing. Distributed or replicated strategies can alleviate the problem in some degree, but there are still some problems of the distributed or replicated model, such as data synchronization, load balance, and so on. We propose a novel architecture for Internet based data processing system based on multicast and anycast protocols. The proposed architecture breaks the functionalities of existing data processing system, in particular, the database functionality, into several agents. These agents communicate with each other using multicast and anycast mechanisms. We show that the proposed architecture provides better scalability, robustness, automatic load balance, and performance than the current distributed architecture of Internet-based data processing. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Improving service for service systems with different arriving rate

    Publication Year: 2003 , Page(s): 315 - 318
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (334 KB) |  | HTML iconHTML  

    We investigate the performance of service systems such as telephone call centers or Internet access points. Previously, birth-death models were developed to compare the throughput of a system that informs customers about anticipated delays and a one that does not give state information as customer arrive. The results were obtained based on the assumption that the two systems have the same parameters (arrival rate, service rate, mean reneging time, mean waiting-time threshold). In reality, it is more likely that they have different arrival rates. We prove results on system throughput with the condition that the parameters of the two systems have the same ratio. We also give sufficient conditions for a system to have a higher throughput. Comparisons for throughput, server utilization and mean response time are also done by simulation. We present the results from the simulation and a brief conclusion is given. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Infrastructure of unified network management system driven by Web technology

    Publication Year: 2003 , Page(s): 111 - 115
    Cited by:  Papers (2)
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (632 KB) |  | HTML iconHTML  

    As distributed network management systems play an increasingly important role in telecommunication network, the flexible, efficient and low-cost unified NMS infrastructure is a key issue in top-level system design. We propose a new infrastructure for a Web-driven, distributed unified network management system. The key technologies in practical application are investigated in detail. The practical application has been implemented using Java and tested on a unified telecommunication network management system. The experiments and application have demonstrated that the infrastructure is feasible and scalable for operation in current and future telecommunication network management. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • An automatic compiler for extracting inter-block parallelism

    Publication Year: 2003 , Page(s): 502 - 506
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (435 KB) |  | HTML iconHTML  

    We present an automatic parallelization scheme involving extracting inter-block parallelism within sequential programs, scheduling block graphs, and generating appropriate target codes for message passing multiprocessors. Compared with existing parallelization compilers, it is capable of exploiting more parallelism because potential parallelism existing within blocks is take into consideration. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • A new architecture of matched-filter bank for high-speed code acquisition of ZCZ-CDMA system

    Publication Year: 2003 , Page(s): 391 - 395
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (421 KB) |  | HTML iconHTML  

    We propose a new architecture of matched-filter (MF) bank for the code acquisition of ZCZ-CDMA system. The ZCZ-CDMA system can accurately detect multipath properties and greatly improve data transmission efficiency by using the special properties of the zero correlation zones (ZCZ) sequence. As a traditional matched-filter bank is used to carry out the code acquisition and the despreading operations to realize the information transmission at 20Mbps in the ZCZ-CDMA system, the circuit scale is too large and the power consumption is too high. Using our proposed new matched-filter bank, the circuit scale and the power consumption is greatly reduced. The high-speed and lower power VLSI design of the new matched-filter bank is implemented on FPGA. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Multipath banyan-based multistage networks for ATM switches

    Publication Year: 2003 , Page(s): 319 - 323
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (415 KB) |  | HTML iconHTML  

    Owing to the single path property of the banyan network, cell losses caused by the internal conflicts is well pronounced. To overcome this problem, multipath ATM switches have been investigated. We propose a class of ATM switch architectures that are constructed based on 3×3 switch elements (SEs). The proposed structures are constructed by adding lateral links to the banyan network to provide many paths between each input/output pair. As a result, the cell losses caused by internal conflicts are significantly reduced. And the proposed switch models are very modular, easy to expand, and use self-routing. The proposed switches use output queues to deal with the problem of cell sequence. From the analysis and simulation results, the proposed switches have the features of low cell loss probability and low cell delay. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • An agent-based approach for grid computing

    Publication Year: 2003 , Page(s): 608 - 611
    Cited by:  Papers (5)
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (409 KB) |  | HTML iconHTML  

    We present an agent-based computational grid (ACG), which applies the concept of computational grid to agents. The ACG system is to implement a uniform higher-level management of the computing resources and services on the grid, and provide users with a consistent and transparent interface for accessing such services. All entities in the grid environment including computing resources and services can be represented as agents. Each entity is registered with a grid service manager. An ACG grid service can be a service agent that provides the actual grid service to the other grid member. Grid members communicate with each other by communication space that is an implementation of tuple space. The design and implementation of four grid protocols for grid service discovery are given. Finally, some conclusions are given. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • A different approach for allocating tasks in a DCS using A

    Publication Year: 2003 , Page(s): 577 - 581
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (454 KB) |  | HTML iconHTML  

    In a distributed computing systems (DCS) tasks may consist of multiple modules. Tasks submitted to a DCS, are usually partitioned into different modules and the modules may be allocated to different processing nodes so as to achieve minimum turn around time of the tasks utilizing the maximum resources of the existing system such as CPU speed, memory capacities etc. The problem lies on how to obtain the optimal allocation of these multiple tasks by keeping in mind that no processing node is overloaded due to this allocation. We propose an algorithm A*RS using well-known A* which aims to reduce the search space and time for allocating the tasks by minimizing the turn around time of tasks in the way so that processing nodes do not become overloaded due to this allocation. Our experimental results justify the claims with necessary supports by comparing it with the earlier algorithm for multiple tasks allocation. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • The adaptive technology of pervasive computing terminals

    Publication Year: 2003 , Page(s): 116 - 121
    Cited by:  Papers (2)
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (453 KB) |  | HTML iconHTML  

    In pervasive computing, the terminal's available resources will change dynamically and unpredictably. The terminal should be adaptive to changes in the supply of resources and respond to these changes. We propose the adaptive architecture based on an embedded operating system and analyze the designing method of adaptive application using fuzzy control theory. We demonstrate the adaptation of the system according to an embedded browser. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Fault-tolerant cycle embedding in dual-cube with node faulty

    Publication Year: 2003 , Page(s): 71 - 75
    Cited by:  Papers (1)
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (449 KB) |  | HTML iconHTML  

    A low-degree dual-cube was proposed as an alternative to the hypercubes. A dual-cube DC(m) has m+1 links per node where m is the degree of a cluster (m-cube) and one more link is used for connecting to a node in another cluster. There are 2m+1 clusters and hence the total number of nodes is 22m+1 in a DC(m). By using Gray code, we show that a DC(m) is (m-1)-link Hamiltonian and there exists a faulty-free cycle containing at least 22m+1-2f nodes with f≤m-1 faulty nodes. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • A sequential delimitative and combinatorial algorithm for a class of integer programming

    Publication Year: 2003 , Page(s): 801 - 804
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (297 KB) |  | HTML iconHTML  

    The algorithm was given for solving a class of integer programming problems, this algorithm generates all combinations according to a certain order of the magnitude of objective function by using a multilevel generating method of preceding for higher place. In the procedure of generating combinations, the algorithm eliminates a majority of infeasible and nonoptimum combinations by using sequential delimitative and combinatorial method so that the algorithm has a higher computational efficiency. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • The non-eye typhoon center locating with parallel computing

    Publication Year: 2003 , Page(s): 488 - 491
    Cited by:  Papers (3)
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (285 KB) |  | HTML iconHTML  

    A new mathematical morphology-based algorithm is proposed to achieve automatic center location of noneye typhoon. The center of a noneye typhoon is near the geometric center of the cloud system and has higher temperature. For each infrared satellite cloud image, the locating procedures are as follows: a) noises filtering, b) main cloud systems segmenting, c) center locating and d) multispectral image verification. The algorithms are based on operations of mathematical morphology, and designed in IBM SP2 massively parallel computer. The experiment results show that the algorithm locates the centers of most noneye typhoons successfully and achieves faster, more precise and nonhuman interactive noneye typhoon center positioning. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • QoS guaranteeing during UMTS packet-domain handover

    Publication Year: 2003 , Page(s): 387 - 390
    Cited by:  Patents (1)
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (377 KB) |  | HTML iconHTML  

    We focus on the problem of guaranteeing real-time service quality during the interRNS handover/relocation in packet-domain/all-IP UMTS. We propose fully seamless SRNS relocation mechanisms for the conversational and streaming services respectively. The multimedia service quality can be guaranteed by packets bicasting in backbone and radio signal macrodiversity. By adapting and extending the data link protocols GTP-U/PDCP/RLC, we synchronize the packet streams in the new and old backbone routes. For the streaming class we propose a new RLC mode called no-sequencing acknowledged RLC, assuring packet lossless delivery and throughput steady during relocation. Our mechanisms are based on existing basic interRNS procedure, and could be easily integrated into it. Synchronous multicast/bicast requires backbone links to be overprovisioned with bandwidth. In multicast-tree mode, the bandwidth overprovision amount of Iu-PS link is the sum of capacities of the neighbor cells. In multicast-on-demand/bicast mode, the bandwidth overprovision amount of Iu-PS link is 2 Mbps. There is no need to overprovision the Iur link bandwidth. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • An instruction cache architecture for parallel execution of Java threads

    Publication Year: 2003 , Page(s): 180 - 184
    Cited by:  Papers (1)
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (460 KB) |  | HTML iconHTML  

    Designing a Java processor supporting horizontal multithreading has become more attractive as network computing gains importance. Different from the traditional superscalar processors that issue multiple instructions from a single instruction stream to exploit the instruction level parallelism (ILP), the horizontal multithreading Java processors issue multiple instructions (bytecodes) from multiple threads in parallel to exploit not only the ILP but the thread level parallelism (TLP). Such processors have multiple dispatch slots and require the instruction fetch unit to supply instructions with much higher bandwidth than superscalar processors. Using a traditional superscalar cache architecture in a horizontal multithreading Java processor results in high cache miss ratio caused by the interference among the threads. We investigate multibank instruction cache architecture for horizontal multithreading Java processor to meet the requirements of the high instruction fetch bandwidth. In order to evaluate the cache performance as well as the horizontal multithreading Java processor performance, we developed a trace driven simulator. The simulator consists of a trace generator that generates the Java bytecode execution traces and an architectural simulator that reads the traces and evaluates the performance of the instruction cache and the overall performance of the Java processor. Our simulation results show that the performance improvements are obtained by the low cache miss ratio and the high instruction fetch bandwidth of the proposed cache architecture. The IPC performance is about 19 when both the number of slots and the number of banks are 8, about 5 times better than one bank cache. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • An optimized multiple access control protocol for a 'little LEO' satellite store-and-forward global data network

    Publication Year: 2003 , Page(s): 413 - 415
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (288 KB) |  | HTML iconHTML  

    The reservation method used in a TD-SRMA protocol determines the protocol's efficiency. The typical reservation methods proposed are SALOHA and VCR-SALOHA (varying calling rate SALOHA). SALOHA is relatively simple to realize but it cannot adapt well to the dynamic traffic pattern in a 'little LEO' satellite network. VCR-SAOHA can better adapt the dynamic traffic, but its highest efficiency on average is still 0.38. To improve the reservation efficiency, a new reservation method, AP-SALOHA (adaptive polling and SALOHA), is proposed. In AP-SALOHA, polling is used for the stations, which have high probability to upload a message, and SALOHA is used for the stations, which have low probability to upload messages. Different from SALOHA and VCR-SALOHA, the length of a reservation period in AP-SALOHA varies according to the network traffic. Analysis and simulation show that AP-SALOHA provides better throughput performance. About 19% improvement over VCR-SALOHA can be achieved. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • A clustering algorithm applied to the satellite networks management

    Publication Year: 2003 , Page(s): 396 - 399
    Cited by:  Papers (1)
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (359 KB) |  | HTML iconHTML  

    As a new type network, the properties of satellite networks are unique comparing with other computer networks and make traditional network management into trouble. Recently, the network management on satellite networks has been on the initial stage and more accurate assumption is not defined. A new clustering algorithm based on graph with the token mechanism and node weights applied to hierarchical satellite networks management is proposed. The unidirectional links and independence of space satellite clustering were considered specially. The algorithm suggests a new idea to the satellite networks management. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • On boundary pixels in seeded region growing segmentation

    Publication Year: 2003 , Page(s): 838 - 839
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (306 KB) |  | HTML iconHTML  

    Boundary pixels can have a subtle but serious impact on the performance of the seeded region growing algorithm [R. Adams et al. (1994)], a fast semiautomatic algorithm for segmentation of intensity images. We have found, however, that as originally described the algorithm can fail if some sections of a connected segment is narrow. The failure results, for example, in a loss of details or an undue dependence of seed placement. The remedy is a careful processing of the boundary pixels. We here present an algorithm whose performance is not affected by such situation. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Using attribute certificate to design role-based access control

    Publication Year: 2003 , Page(s): 216 - 218
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (335 KB) |  | HTML iconHTML  

    Public key infrastructure (PKI), privilege management infrastructure (PMI) and role based access control (RBAC) are well-established approaches to ease the management of secure systems and to enable secure distributed systems. We elaborate on using the RBAC approach for its administrative advantages combined with authentication based on public key certificates (PKCs) and authorization based on attributes certificates (ACs). In particular, we concentrate on how they can be integrated to support the authentication and authorization model in CERNET setting. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Study on cross-layer design and power conservation in ad hoc network

    Publication Year: 2003 , Page(s): 324 - 328
    Cited by:  Papers (4)  |  Patents (1)
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (391 KB) |  | HTML iconHTML  

    As the use of small portable and mobile computing devices increases, ad hoc networks have become an increasingly popular topic of research and development activities. Cross-layer design and power conservation are two new technologies in this area. Both of them can make the ad hoc network more robust and more adaptive. The relation between them is improvable. A power control based cross-layer architecture is presented. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • The adaptive SAR technology in non-ATM network node

    Publication Year: 2003 , Page(s): 311 - 314
    Cited by:  Papers (2)
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (332 KB) |  | HTML iconHTML  

    The IP packet length distribution is changing with Internet development and traffic. A different network also has a different IP packet length distribution. The IP packet processing methods of many nonATM network nodes are similar with segment and reassemble (SAR). The payload length of the cell is fixed. When the network node is placed in a network, a large number of redundant bytes, which result from the zero-pads and headers being used, decrease the processing bandwidth. When a well-designed node is placed in a different network with different distribution, the processing bandwidth of the node may be even lower. We present an adaptive IP packet SAR processing solution, which can adapt the traffic of the network. Computer simulations based on the packet length statistic data are also given. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • On Schnorr-Adleman lattice

    Publication Year: 2003 , Page(s): 946 - 949
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (293 KB) |  | HTML iconHTML  

    Lattice theory has been found numerous applications in mathematics and computer science. The shortest vector problem (SVP) and the closest vector problem (CVP) are basic problems of lattice theory. It has been known that the CVP is NP-hard under Karp reduction. SVP is NP-hard under randomized reduction within any constant factor less than √2. The reduction from large integer factorization and discrete logarithm to CVP under a reasonable hypothesis. The factoring problem was random polynomial time reducible to SVP. These methods were believed to provide the possibility of factoring large integers and solving discrete logarithms by approximate lattice reduction algorithms. This is called Schnorr-Adleman lattice. We discuss some properties of Schnorr-Adleman lattice, and improve a result in [D. Micciancio et al., (2002)]. Our results can also be viewed as generalization of some results of [L. M. Adleman] and [C. P. Schnorr (1993)] in some sense. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • A new approach for performance evaluation of TCP over CDMA-based channel

    Publication Year: 2003 , Page(s): 361 - 363
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (309 KB) |  | HTML iconHTML  

    A new metric for performance evaluation of TCP over CDMA channel based on the interference-limited characteristics of CDMA is proposed. According to the new metric, the performance of TCP over CDMA correlated channel for different protocol parameters and different versions is investigated. The results show that, appropriate selection of protocol parameters and PER operation point can improve significantly the capacity of packet-switched CDMA-based network. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.