By Topic

Parallel, Distributed and Network-based Processing, 2002. Proceedings. 10th Euromicro Workshop on

Date 11-11 Jan. 2002

Filter Results

Displaying Results 1 - 25 of 76
  • Proceedings 10th Euromicro Workshop on Parallel, Distributed and Network-based Processing

    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (344 KB)  

    The following topics are dealt with: parallel, distributed and network-based processing; performance analysis; Web computing; failure handling; Java and Jini; parallel and distributed programming tools for grids; unorthodox computing architectures; systems and applications; message passing; scheduling; algorithms; and mobile ad hoc networks. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Session 1: performance analysis [breaker page]

    Page(s): 5
    Save to Project icon | Request Permissions | PDF file iconPDF (289 KB)  
    Freely Available from IEEE
  • Session 1: performance analysis [breaker page]

    Page(s): 5
    Save to Project icon | Request Permissions | PDF file iconPDF (289 KB)  
    Freely Available from IEEE
  • Session 2: web computing [breaker page]

    Page(s): 47
    Save to Project icon | Request Permissions | PDF file iconPDF (269 KB)  
    Freely Available from IEEE
  • Meta/spl psi/: a Web-based metacomputing environment to build a computational chemistry problem solving environment

    Page(s): 49 - 54
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (793 KB) |  | HTML iconHTML  

    The increasing complexity of large distributed scientific applications raises the problem of the coordination of diverse computational resources (computers, data bases, etc.). Multi-disciplinary applications often make use of coupled computational resources that cannot be replicated at a single site. There is the need for smart and user-friendly problem-solving environments that free scientists from concerns related to the location and complexity of the computing platform being used. We describe the main features of the Meta/spl psi/ software tool developed at CNUCE-CNR to build PSEs for the execution of complex applications on a Web-based metacomputer. This tool is designed to supply completely transparent support to the user who thus does not need to be aware of the location and the allocation of computing resources. The paper introduces the real application's needs coming from a specific scientific community, and motivates the realization of Meta/spl psi/. It then focuses on the architectural and implementation aspects of Meta/spl psi/ and finally summarizes the work. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Session 3: failure handling [breaker page]

    Page(s): 79
    Save to Project icon | Request Permissions | PDF file iconPDF (265 KB)  
    Freely Available from IEEE
  • Session 4: Java and Jini [breaker page]

    Page(s): 107
    Save to Project icon | Request Permissions | PDF file iconPDF (289 KB)  
    Freely Available from IEEE
  • Session 5: parallel and distributed programming tools for grids [breaker page]

    Page(s): 137
    Save to Project icon | Request Permissions | PDF file iconPDF (289 KB)  
    Freely Available from IEEE
  • Session 6: programming [breaker page]

    Page(s): 171
    Save to Project icon | Request Permissions | PDF file iconPDF (262 KB)  
    Freely Available from IEEE
  • Session 7: unorthodox computing architectures [breaker page]

    Page(s): 209
    Save to Project icon | Request Permissions | PDF file iconPDF (288 KB)  
    Freely Available from IEEE
  • Session 8: systems and applications [breaker page]

    Page(s): 251
    Save to Project icon | Request Permissions | PDF file iconPDF (262 KB)  
    Freely Available from IEEE
  • Session 9: message passing [breaker page]

    Page(s): 291
    Save to Project icon | Request Permissions | PDF file iconPDF (288 KB)  
    Freely Available from IEEE
  • Session 10: scheduling [breaker page]

    Page(s): 317
    Save to Project icon | Request Permissions | PDF file iconPDF (261 KB)  
    Freely Available from IEEE
  • Session 11: algorithms I [breaker page]

    Page(s): 351
    Save to Project icon | Request Permissions | PDF file iconPDF (264 KB)  
    Freely Available from IEEE
  • Session 12: mobile ad hoc networks [breaker page]

    Page(s): 393
    Save to Project icon | Request Permissions | PDF file iconPDF (266 KB)  
    Freely Available from IEEE
  • Session 13: algorithms 2 [breaker page]

    Page(s): 427
    Save to Project icon | Request Permissions | PDF file iconPDF (288 KB)  
    Freely Available from IEEE
  • Session 14: algorithms 3 [breaker page]

    Page(s): 453
    Save to Project icon | Request Permissions | PDF file iconPDF (261 KB)  
    Freely Available from IEEE
  • Author index

    Page(s): 479 - 480
    Save to Project icon | Request Permissions | PDF file iconPDF (207 KB)  
    Freely Available from IEEE
  • Efficient implementation of cellular algorithms on reconfigurable hardware

    Page(s): 211 - 218
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (301 KB) |  | HTML iconHTML  

    Reconfigurable architectures represent an innovative approach to the computer system design paradigm, which tries to cope with a problem of inefficiency of conventional computing systems, due to their general purpose nature. On the other hand, cellular automata are attractive computing models due to their fine grain parallelism, simple computational structures and local communication patterns. The inherently parallel cellular automata model is well suited to be implemented on reconfigurable hardware architectures such as field programmable gate arrays (FPGA) that can provide significant speedup. This paper describes the CAREM system that provides an efficient implementation of cellular automata algorithms on FPGA systems exploiting their reconfigurable features for executing different cellular automata rules. Its application to an image processing application and a forest fire simulation are presented and discussed. Performance evaluation and comparison with different implementations of cellular automata are presented View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Performance sensitivity of routing algorithms to failures in networks of workstations with regular and irregular topologies

    Page(s): 81 - 90
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (326 KB) |  | HTML iconHTML  

    Networks of workstations (NOWs) provide a cost-effective alternative to parallel computers. Components in NOWs may fail, degrading the network operation until the faults are repaired. In this paper, we analyze the influence of both switch and link failures on the network performance. In particular, given that network performance in NOWs strongly depends on the applied routing algorithm, we quantify the sensitivity to failures of two routing algorithms: flexible routing and up*/down* routing algorithms. In the case of up*/down* routing, two methodologies to compute routing tables are evaluated. Evaluation results modeling a Myrinet network show that, in general, up*/down* routing is more robust to failures, although its behavior strongly depends on the type of network topology, regular or irregular, and the methodology used to compute routing tables. However, the flexible routing algorithm presents a better performance, regardless of the network topology, even in presence of failures, but at expense of a larger sensitivity View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Beowulf performance in CFD multigrid applications

    Page(s): 7 - 14
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (298 KB) |  | HTML iconHTML  

    Computational fluid dynamics is probably one of the most computationally demanding disciplines, a driving force behind the development of new computer architectures. In fact, the design and evaluation of high-performance parallel systems is commonly based on CFD workloads. One of the most remarkable examples of such workloads is the NAS parallel benchmark, which aims to mimic the computation and data-movement characteristics of large scale CFD applications. We have paid specific attention to the NAS-MG (multigrid) kernel, since these methods represent one of the most promising solvers in the field of CFD. Nevertheless, practical flow computations demand robust multigrid algorithms which differ from the NAS-MG kernel. The paper presents a performance evaluation of a Beowulf system using both a state-of-the-art multigrid solver and the NAS-MG benchmark. These two codes have been used to compare several of its design choices, namely, the interconnection network (GigaNet versus Fast-Ethernet) as well as the node configuration (dual nodes versus single nodes). The results highlight that the optimal combination strongly depends on the target application View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Eventually consistent failure detectors

    Page(s): 91 - 98
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (296 KB) |  | HTML iconHTML  

    The concept of unreliable failure detector was introduced by T.D. Chandra and S. Toueg (1996) as a mechanism that provides information about process failures. This mechanism has been used to solve different problems in asynchronous systems, in particular the Consensus problem. In this paper, we present a new class of unreliable failure detectors, which we call Eventually Consistent and denote by □C. This class adds to the failure detection capabilities of other classes an eventual leader election capability. This capability allows all correct processes to eventually choose the same correct process as leader. We study the relationship between □C and other classes of failure detectors. We also propose an efficient algorithm to trans form □C into □P in models of partial synchrony. Finally, to show the power of this new class of failure detectors, we present a Consensus algorithm based on □C. This algorithm successfully exploits the leader election capability of the failure detector and performs better in number of rounds than all the previously proposed algorithms for failure detectors with eventual accuracy View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • SICOSYS: an integrated framework for studying interconnection network performance in multiprocessor systems

    Page(s): 15 - 22
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (409 KB) |  | HTML iconHTML  

    An environment has been developed which is capable of determining the impact that a multiprocessor interconnection subsystem causes on real application execution time. A general-purpose interconnection network simulator, called SICOSYS, able to capture essential aspects of the low-level implementation, has been integrated into two execution driven simulators for multiprocessors: RSIM and SimOS. The enhancement of both tools allows the analysis of new proposals for the interconnection subsystem of a cc-NUMA machine, from the VLSI level up to the real application level. Any new proposal can be translated to a specific message router architecture and by using a low-level implementation tool, the parameter delays of a detailed router model to be used by SICOSYS can be obtained View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • On the impossibility of implementing perpetual failure detectors in partially synchronous systems

    Page(s): 99 - 105
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (273 KB) |  | HTML iconHTML  

    In this paper we study the implementability of different classes of failure detectors in several models of partial synchrony. We show that no failure detector with perpetual accuracy (namely, P, Q, S, and W) can be implemented in any of the models of partial synchrony proposed previously in systems with even a single failure. We also show that, in these models of partial synchrony, it is necessary for a majority of correct processes to implement a failure detector of class Θ View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Improving induction decision trees with parallel genetic programming

    Page(s): 181 - 187
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (262 KB) |  | HTML iconHTML  

    A parallel genetic programming approach to induce decision trees in large data sets is presented. A population of trees is evolved by employing the genetic operators and every individual is evaluated by using a fitness function based on the J-measure. The method is able to deal with large data sets since it uses a parallel implementation of genetic programming through the grid model. Experiments on data sets from the UCI machine learning repository show better results with respect to C5. Furthermore, performance results show a nearly linear speedup View full abstract»

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