We consider the distributed setting of N autonomous mobile agents that operate in Look-Compute-Move (LCM) cycles and communicate with other agents using colored lights (the agents with lights model). We study the fund...
详细信息
ISBN:
(纸本)9780769561493
We consider the distributed setting of N autonomous mobile agents that operate in Look-Compute-Move (LCM) cycles and communicate with other agents using colored lights (the agents with lights model). We study the fundamental COMPLETE VISIBILITY problem of repositioning N agents on a plane so that each agent is visible to all others. We assume obstructed visibility under which an agent cannot see another agent if a third agent is positioned between them on the straight line connecting them. We are interested in fault-tolerant algorithms;all existing algorithms for this problem are not fault-tolerant. We study fault-tolerance with respect to failures on the mobility of agents. Therefore, any algorithm for COMPLETE VISIBILITY is required to provide visibility between all non-faulty agents, independently of the behavior of the faulty ones. We model mobility failures as crash faults in which each faulty agent is allowed to stop its movement at any time and, once the faulty agent stopped moving, that agent will remain stationary indefinitely. In this paper, we present and analyze an algorithm that solves COMPLETE VISIBILITY tolerating one crash-faulty agent in a system of N >= 3 agents.
This paper describes the design philosophy for the Grid system being developed by Japan Committee on High-Performance Computing for Bioinformatics and Initiative for parallel Bioinformatics (IPAB). Grid is one of attr...
详细信息
ISBN:
(纸本)0769516599
This paper describes the design philosophy for the Grid system being developed by Japan Committee on High-Performance Computing for Bioinformatics and Initiative for parallel Bioinformatics (IPAB). Grid is one of attractive solutions to achieve distributed bioinformtics environment with high performance parallel computers, large genomic databases, computation intensive applications such as homology search and molecular simulation. However, much has been remained in Grid system design especially in the wide area network environment. OBIGrid emphasizes the virtual organization aspect of the Grid system and gives more priority on security and scalability rather than performance.
The use of the DEEP development environment to analyze parallel program performance is described. The full integrated environment contains tools for the creation, analysis and debugging of parallel programs. All infor...
详细信息
ISBN:
(纸本)0818684038
The use of the DEEP development environment to analyze parallel program performance is described. The full integrated environment contains tools for the creation, analysis and debugging of parallel programs. All information is related back to the original parallel source code. This paper describes the program analysis portion of DEEP, with examples of its use on data parallel programs (HPF and Data parallel C) and shared memory parallel (SMP) programs.
paralleldistributed systems in which multiple computers are connected through LAN and WAN are widely used at present. One of the important function of paralleldistributed systems is broadcasting to deliver data over...
详细信息
ISBN:
(纸本)0780388348
paralleldistributed systems in which multiple computers are connected through LAN and WAN are widely used at present. One of the important function of paralleldistributed systems is broadcasting to deliver data over an entire system. Multicasting is an expanded function of broadcasting. Broadcast/multicast processing greatly affects the performance of paralleldistributed systems, and it is desirable to develop a delivering procedure that completes the processing in a short time. There have been many approaches to the minimum broadcast time problem, aiming at the optimal delivering procedure. In this paper, we consider the minimum broadcast/multicast time problem for a paralleldistributed system whose topology can be represented by a subclass of split graphs.
We propose a new parallel performance visualization scheme, based on a simple moment analysis of processor utilization data. This method combines the scalability advantages of statistical summaries with the more revea...
详细信息
ISBN:
(纸本)0818684038
We propose a new parallel performance visualization scheme, based on a simple moment analysis of processor utilization data. This method combines the scalability advantages of statistical summaries with the more revealing processor utilization information of Gantt charts. it scales well to large numbers of processors, requiring only storage constant in execution time.
A synergistic approach utilizing compilation, compaction, and parallelization is described to achieve real-time computing throughput from rule-based expert systems. The methodology involves synthesizing a set of concu...
详细信息
暂无评论