Four of the articles in this special issue were the recipients of the Best Paper Awards at the 2006 International Parallel and distributed processing Symposium (IPDPS). There is one award winner for each of the four t...
详细信息
Four of the articles in this special issue were the recipients of the Best Paper Awards at the 2006 International Parallel and distributed processing Symposium (IPDPS). There is one award winner for each of the four tracks of IPDPS: Algorithms, Applications, Architecture, and Software. Although the papers were solicited for this issue, they underwent the same rigorous reviewing process as do all papers submitted to this journal. We are grateful to the authors for submitting their excellent work to this issue, and we offer our heartfelt thanks to the anonymous reviewers of the papers, who selflessly contributed to the quality of this issue.
作者:
Nakamura, NYamada, YMurashige, ANTT
Multimedia Network Labs Yokosuka Kanagawa 2390847 Japan NTT
Multimedia Network Labs Musashino Tokyo 1800012 Japan
Expansion of LP network scale usually results in an increase of the processing loads related to network management, making it difficult to implement management systems based on a single unit of hardware. Therefore, im...
详细信息
Expansion of LP network scale usually results in an increase of the processing loads related to network management, making it difficult to implement management systems based on a single unit of hardware. Therefore, implementation of network management systems using distributed processing technologies is actively investigated. Important issues in the design of network management systems based on distributed processing are the correspondence of distributed functions with the managed agents supported by the network, as well as the procedure calls between distributed functions. In order to realize fully the performance of a network management system, it is necessary to design a system such that the correspondence be tween the functions and the network agents can be easily maintained as the network is expanded acid the overhead due to the distributed processing is kept as low as possible. In this paper, we propose an architecture of a network management system intended for an efficient management of large-scale networks that makes it possible to increase the number of subsystems by designing the functionalities aimed at the distribution of IP network management. Using such an architecture, we built an actual distributed IP network management system based on a DCE distributed environment and demonstrated that such a system exhibits adequate performance characteristics. (C) 2000 Scripta Technica.
We consider the problem of distributed representation of signals in sensor networks, where sensors exchange quantized information with their neighbors. The signals of interest are assumed to have a sparse representati...
详细信息
We consider the problem of distributed representation of signals in sensor networks, where sensors exchange quantized information with their neighbors. The signals of interest are assumed to have a sparse representation with spectral graph dictionaries. We further model the spectral dictionaries as polynomials of the graph Laplacian operator. We first study the impact of the quantization noise in the distributed computation of matrix-vector multiplications, such as the forward and the adjoint operator, which are used in many classical signal processing tasks. It occurs that the performance is clearly penalized by the quantization noise, whose impact directly depends on the structure of the spectral graph dictionary. Next, we focus on the problem of sparse signal representation and propose an algorithm to learn polynomial graph dictionaries that are both adapted to the graph signals of interest and robust to quantization noise. Simulation results show that the learned dictionaries are efficient in processing graph signals in sensor networks where bandwidth constraints impose quantization of the messages exchanged in the network.
This paper considers advanced IN systems in which the data for each customer are allocated in a distributed way among multiple processor modules. A method is presented in which the traffic imbalance for customer data ...
详细信息
This paper considers advanced IN systems in which the data for each customer are allocated in a distributed way among multiple processor modules. A method is presented in which the traffic imbalance for customer data is represented. Based on such a representation, a method to evaluate the load imbalance among the processor modules, as well as a method to estimate its upper limit, are presented. Using the presented method, even if customer data composed of number conversion and screening information, and with a large traffic imbalance factor of several thousand to several tens of thousands, are allocated at random to the processor modules, the load imbalance can be reduced to 10 to 20% at the present level of processor performance and number of dynamic steps in the transaction. It is shown that dynamic load balancing can be achieved for this kind of imbalance by combining the processing that can be executed independently of the customer data, such as directory retrieval, which appears in ordinary distributed processing. (C) 1999 Scripta Technica.
J. Aguado and V. Quintana comment on the paper by R. Ebrahimian and R. Baldick (see ibid., vol.15, no.4, p.1240-6, 2000), and ask for comments on five points relating to their own work in this area of power network st...
详细信息
J. Aguado and V. Quintana comment on the paper by R. Ebrahimian and R. Baldick (see ibid., vol.15, no.4, p.1240-6, 2000), and ask for comments on five points relating to their own work in this area of power network state estimation. The original authors reply to the comments.
This paper considers an advanced intelligent network (IN) service control node, which combines function distribution and load distribution, where processing performance can be expanded by adding processor modules when...
详细信息
This paper considers an advanced intelligent network (IN) service control node, which combines function distribution and load distribution, where processing performance can be expanded by adding processor modules when the load is increased. A highly reliable configuration for the connection mechanism, in which a large number of processor modules are connected by ATM, a control scheme that improves the communication processing performance of each processor module, and a performance evaluation procedure are presented. As to the ATM connection mechanism, a configuration is presented in which the reliability is not much reduced even if a large-scale system is constructed. In the proposed configuration, the function of connecting processor modules, the function of multiplex/demulitplexing signals, and the function of cell routing are realized by separate hardware, and double redundancy is applied to each function. For improvement of the communication processing performance of the processor module, a control scheme is considered in which the messages are grouped and processed for each time unit. The relation between the response time and the throughput is analyzed. A performance evaluation program is presented, which examines the time required for sending/receiving messages between the constructed processor modules. (C) 1999 Scripta Technica.
Efficient distributed processing is vital for collaborative searching tasks of robotic swarm systems. Typically, those systems are decentralized, and the members have only limited communication and processing capaciti...
详细信息
Efficient distributed processing is vital for collaborative searching tasks of robotic swarm systems. Typically, those systems are decentralized, and the members have only limited communication and processing capacities. What is illustrated in this paper is a distributed processing paradigm for robotic swarms moving in a line or v-shape formation. The introduced concept is capable of exploits the line and v-shape formations for 2-D filtering and processing algorithms based on a modified multi-dimensional Roesser model. The communication is only between nearest adjacent members with a simple state variable. As an example, we applied a salient region detection algorithm to the proposed framework. The simulation results indicate the designed paradigm can detect salient regions by using a moving line or v-shape formation in a scanning way. The requirement of communication and processing capability in this framework is minimal, making it a good candidate for collaborative exploration of formatted robotic swarms.
A one dimensional (1D) and a two dimensional (2D) Kalman filters as means of distributed processing are developed for two single input multiple output (SIMO) AWGN channels, i.e., one without and the other with inter s...
详细信息
A one dimensional (1D) and a two dimensional (2D) Kalman filters as means of distributed processing are developed for two single input multiple output (SIMO) AWGN channels, i.e., one without and the other with inter symbol interference (ISI), respectively. We show the capacity of the distributed and centrally processed channels to be the same for channels with and without ISI.
For middle-speed and four-stroke marine diesel engine, a new Main Engine Remote Control System based on distributed processing and dual-redundant CAN network communication technology was presented. It includes engine ...
详细信息
For middle-speed and four-stroke marine diesel engine, a new Main Engine Remote Control System based on distributed processing and dual-redundant CAN network communication technology was presented. It includes engine control unit, indicating panel unit, engine safety unit, main engine interface, digital governor unit and distributed processing unit. System data can be exchanged by dual-redundant CAN network. The distributed processing unit is independent in physics completely, so it has less impact on other units when some units malfunctioned. The communication of system units is designed by open communication protocol of CAN, it can self-test and troubleshoot. The System has function of self-insurance when loss power, and can delay shut-down when loss of air. Its operating mode and man-machine interface are convenient and friendly.
暂无评论