Parallel Processing and Applied Mathematics

1. Front Matter

   PDF

2. Parallel and Distributed Architectures

   1. Session Guarantees to Achieve PRAM Consistency of Replicated Shared Objects

      • Jerzy Brzezinski, Cezary Sobaniec, Dariusz Wawrzyniak

      Pages 1-8

   2. An Extended Atomic Consistency Protocol for Recoverable DSM Systems

      • Jerzy Brzezinski, Michal Szychowiak

      Pages 9-16

   3. Hyper-Threading Technology Speeds Clusters

      • Kazimierz Wackowski, Pawel Gepner

      Pages 17-26

   4. Building Autonomic Clusters: A Response to IBM’s Autonomic Computing Challenge

      • Andrzej Goscinski, Jackie Silcock, Michael Hobbs

      Pages 27-35

   5. Configurable Microprocessor Array for DSP Applications

      • Oleg Maslennikow, Juri Shevtshenko, Anatoli Sergyienko

      Pages 36-41

   6. On Generalized Moore Digraphs

      • Michael Sampels

      Pages 42-49

   7. RDMA Communication Based on Rotating Buffers for Efficient Parallel Fine-Grain Computations

      • Adam Smyk, Marek Tudruj

      Pages 50-58

   8. Communication on the Fly in Dynamic SMP Clusters – Towards Efficient Fine Grain Numerical Computations

      • Marek Tudruj, Lukasz Masko

      Pages 59-68

   9. Wait-Free Publish/Subscribe Using Atomic Registers

      • Pradeep Varma

      Pages 69-76

3. Scheduling and Load Balancing

   1. Accelerated Diffusion Algorithms on General Dynamic Networks

      • Jacques Bahi, Raphaël Couturier, Flavien Vernier

      Pages 77-82

   2. Suitability of Load Scheduling Algorithms to Workload Characteristics

      • Eunmi Choi, Dugki Min

      Pages 83-88

   3. Minimizing Time-Dependent Total Completion Time on Parallel Identical Machines

      • Stanisław Gawiejnowicz, Wiesław Kurc, Lidia Pankowska

      Pages 89-96

   4. Diffusion Based Scheduling in the Agent-Oriented Computing System

      • Marek Grochowski, Robert Schaefer, Piotr Uhruski

      Pages 97-104

   5. Approximation Algorithms for Scheduling Jobs with Chain Precedence Constraints

      • Klaus Jansen, Roberto Solis-Oba

      Pages 105-112

   6. Combining Vector Quantization and Ant-Colony Algorithm for Mesh-Partitioning

      • Jurij Šilc, Peter Korošec, Borut Robič

      Pages 113-118

   7. Wavelet-Neuronal Resource Load Prediction for Multiprocessor Environment

      • Paweł Hajto, Marcin Skrzypek

      Pages 119-124

   8. Fault-Tolerant Scheduling in Distributed Real-Time Systems

      • Nguyen Duc Thai

      Pages 125-130

   9. Online Scheduling of Multiprocessor Jobs with Idle Regulation

      • Andrei Tchernykh, Denis Trystram

      Pages 131-144

4. Performance Analysis and Prediction

   1. Predicting the Response Time of a New Task on a Beowulf Cluster

      • Marta Beltrán, Jose L. Bosque

      Pages 145-152

It is our pleasure to provide you with the volume containing the proceedings of the 5th International Conference on Parallel Processing and Applied Mathe- tics, which was held in Cz¸ estochowa, a Polish city famous for its Jasna Gora Monastery, on September 7–10, 2003. The ?rst PPAM conference was held in 1994 and was organized by the Institute of Mathematics and Computer Science of the Cz¸ estochowa University of Technology in its hometown. The main idea behind the event was to provide a forum for researchers involved in applied and computational mathematics and parallel computing to exchange ideas in a relaxed atmosphere. Conference organizers hoped that this arrangement would result in cross-pollination and lead to successful research collaborations. In - dition, they hoped that the initially mostly Polish conference would grow into an international event. The fact that these assumptions were correct was proven by the growth of the event. While the ?rst conference consisted of 41 presen- tions, the conference reached 150 participants in Na l¸ ecz´ ow in 2001. In this way the PPAM conference has become one of the premiere Polish conferences, and de?nitely the most important one in the area of parallel/distributed computing andappliedmathematics. This year’s meeting gathered almost 200 participants from 32 countries. A strict refereeing process resulted in the acceptance of approximately 150 cont- buted presentations, while the rejection rate was approximately 33%.

• Scheduling
• algorithms
• bioinformatics
• distributed computing
• distributed systems
• grid computing
• high-performance computing
• load balancing
• non-numerical algorithms
• numerical algorithms
• parallel algorithms
• parallel architectures
• parallel computing
• performance eval
• scientific computing
• algorithm analysis and problem complexity

• Institute of Computational and Information Sciences, Czestochowa University of Technology,  

  Roman Wyrzykowski

• Computer Science Department, University of Tennessee, Knoxville, USA

  Jack Dongarra

• Systems Research Institute, Polish Academy of Science, Warsaw, Poland

  Marcin Paprzycki

• Informatics & Mathematical Modeling, Technical University of Denmark, Lyngby, Denmark

  Jerzy Waśniewski

Policies and ethics