This monograph coherently presents a series of research results on concurrent production systems recently contributed by the author and several co-authors.
Before going into the details of concurrent production systems the performance of a single system is discussed. Based on these considerations it is explained how to estimate the efficiency of production systems programs and how to automatically determine efficient join structures. Parallel systems, including algorithms for parallel rule firings, distributed systems under distributed control, and particularly multiagent production systems and the related control issues are investigated. Finally a meta-level control architecture is applied to the construction of the multiagent system CoCo for public telecommunication network control.

The articles in this volume are revised versions of the best papers presented at the Fifth Workshop on Languages and Compilers for Parallel Computing, held at Yale University, August 1992. The previous workshops in this series were held in Santa Clara (1991), Irvine (1990), Urbana (1989), and Ithaca (1988). As in previous years, a reasonable cross-section of some of the best work in the field is presented. The volume contains 35 papers, mostly by authors working in the U.S. or Canada but also by authors from Austria, Denmark, Israel, Italy, Japan and the U.K.

The REX School/Symposium "A Decade of Concurrency - Reflections and Perspectives" was the final event of a ten-year period of cooperation between three Dutch research groups working on the foundations of concurrency.
Ever since its inception in 1983, the goal of the project has been to contribute to the cross-fertilization between formal methods from the fields of syntax, semantics, and proof theory, aimed at an improved understanding of the nature of parallel computing. The material presented in this volume was prepared by the lecturers (and their coauthors) after the meeting took place.
In total, the volume constitutes a thorough state-of-the-art report of the research activities in concurrency.

Parallel processing offers a solution to the problem of providing the processing power necessary to help understand and master the complexity of natural phenomena and engineering structures. By taking several basic processing devices and connecting them together the potential exists of achieving a performance many times that of an individual device. However, building parallel application programs is today recognized as a highly complex activity requiring specialist skills and in-depth knowledge. PARLE is an international, European based conference which focuses on the parallel processing subdomain of informatics and information technology. It is intended to become THE European forum for interchange between experts in the parallel processing domain and to attract both industrial and academic participants with a technical programme designedto provide a balance between theory and practice. This volume contains the proceedings of PARLE '93. The PARLE conference came into existence in 1987 as an initiative from the ESPRIT I programme and the format was revised in 1991/92. PARLE '93 is the second conference with the new format and was held in Munich.

The prefix operation on a set of data is one of the simplest and most useful building blocks in parallel algorithms. This introduction to those aspects of parallel programming and parallel algorithms that relate to the prefix problem emphasizes its use in a broad range of familiar and important problems. The book illustrates how the prefix operation approach to parallel computing leads to fast and efficient solutions to many different kinds of problems. Students, teachers, programmers, and computer scientists will want to read this clear exposition of an important approach.

The 18th International Workshop on Graph-Theoretic Concepts in Computer Science (WG '92) was held in Wiesbaden-Naurod, Germany, June 18-20, 1992. Itwas organized by the Department of Computer Science, Johann Wolfgang Goethe University, Frankfurt am Main. Contributions with original results inthe study and application of graph-theoretic concepts in various fields of computer science were solicited, and 72 papers were submitted and reviewed, from which 29 were selected for presentation at the workshop. The workshop was attended by 61 scientists from 16 countries. All 29 papers in the volume have undergone careful revision after the meeting, based on the discussions and comments from the audience and the referees. The volume is divided into parts on restricted graph classes, scheduling and related problems, parallel anbd distributed algorithms, combinatorial graph problems, graph decomposition, graph grammars and geometry, and modelling by graphs.

Distributed-memory multiprocessing systems (DMS), such as Intel's hypercubes, the Paragon, Thinking Machine's CM-5, and the Meiko Computing Surface, have rapidly gained user acceptance and promise to deliver the computing power required to solve the grand challenge problems of Science and Engineering. These machines are relatively inexpensive to build, and are potentially scalable to large numbers of processors. However, they are difficult to program: the non-uniformity of the memory which makes local accesses much faster than the transfer of non-local data via message-passing operations implies that the locality of algorithms must be exploited in order to achieve acceptable performance. The management of data, with the twin goals of both spreading the computational workload and minimizing the delays caused when a processor has to wait for non-local data, becomes of paramount importance. When a code is parallelized by hand, the programmer must distribute the program's work and data to the processors which will execute it. One of the common approaches to do so makes use of the regularity of most numerical computations. This is the so-called Single Program Multiple Data (SPMD) or data parallel model of computation. With this method, the data arrays in the original program are each distributed to the processors, establishing an ownership relation, and computations defining a data item are performed by the processors owning the data.

The Austrian Center for Parallel Computation (ACPC) is a cooperative research organization founded in 1989 to promote research and education in the field of software for parallel computer systems. The areas in which the ACPC is active include algorithms, languages, compilers, programming environments, and applications for parallel and high-performance computing systems. This volume contains the proceedings of the Second International Conference of the ACPC, held in Gmunden, Austria, October 1993. Authors from 17 countries submitted 44 papers, of which 15 were selected for inclusion in this volume, which also includes 4 invited papers by distinguished researchers. The volume is organized into parts on architectures (2 papers), algorithms (7 papers), languages (6 papers), and programming environments (4 papers).

The substantial effort of parallelizing scientific programs is only justified if the resulting codes are efficient. Thus, all types of performance tuning are important to parallel software development. But performance improvements are much more difficult to achieve with parallel programs than with sequential programs. One way to overcome this difficulty is to bring in graphical tools. This monograph covers recent developments in parallel program visualization techniques and tools and demonstrates the application of specific visualization techniques and software tools to scientific parallel programs. The solution of initial value problems of ordinary differential equations, and numerical integration are treated in detail as two important examples.

Research in the field of parallel computer architectures and parallel algorithms has been very successful in recent years, and further progress isto be expected. On the other hand, the question of basic principles of the architecture of universal parallel computers and their realizations is still wide open. The answer to this question must be regarded as mostimportant for the further development of parallel computing and especially for user acceptance. The First Heinz Nixdorf Symposium brought together leading experts in the field of parallel computing and its applications to discuss the state of the art, promising directions of research, and future perspectives. It was the first in a series of Heinz Nixdorf Symposia, intended to cover varying subjects from the research spectrum of the Heinz Nixdorf Institute of the University of Paderborn. This volume presents the proceedings of the symposium, which was held in Paderborn in November 1992. The contributions are grouped into four parts: parallel computation models and simulations, existing parallel machines, communication and programming paradigms, and parallel algorithms.

This volume presents the proceedings of the joint meeting CONPAR 92 - VAPP V, held in Lyon, France, September 1992. The international Conferences on Parallel Processing (CONPAR) and the meetings on Vector and Parallel Processors in computational science (VAPP) have been held jointly since CONPAR 90 - VAPP IV, held in Zurich. The aim of the meeting presented in this volume is to review hardware and architecture developmentstogether with languages and software tools for supporting parallel processing and to highlight advances in models, algorithms, andapplications software on vector and parallel architectures. The papers in the volume are organized into sections on networks, software tools, distributed algorithms, dedicated architectures, numerical applications, systolic algorithms, parallel linear algebra, architectures, shared virtual memory, load balancing, data parallelism, parallel algorithms, image processing, compiling and scheduling, simulation and performance analysis, parallel artificialintelligence, dataflow architectures, parallel programming, and poster presentations.

Code Generation - Concepts, Tools, Techniques is based upon the proceedings of the Dagstuhl workshop on code generation which took place from 20-24 May 1991. The aim of the workshop was to evaluate current methods of code generation and to indicate the main directions which future research is likely to take. It provided an excellent forum for the exchange of ideas and had the added advantage of bringing together European and American experts who were unlikely to meet at less specialised gatherings. This volume contains 14 of the 30 papers presented at the Dagstuhl workshop. The papers deal mainly with the following four topics: tools and techniques for code generation, code generation for parallel architectures, register allocation and phase ordering problems, and formal methods and validations. Most of the papers assess the progress of on-going research work, much of which is published here for the first time, while others provide a review of recently completed projects. The volume also contains summaries of two discussion groups which looked at code generation tools and parallel architectures. As a direct result of one of these discussions, a group of the participants have collaborated to make a pure BURS system available for public distribution. This system, named BURG, is currently being beta-tested. Code Generation - Concepts, Tools, Techniques provides a representative summary of state-of-the-art code generation techniques and an important assessment of possible future innovations. It will be an invaluable reference work for researchers and practitioners in this important area.

Petri nets are a well-known model for parallel systems, used for both applications and theoretical studies. They can be used for specification, modelling, and analysis, and offer a graphical representation and a clear view of concurrency. For the design of large systems, modular construction is indispensible, and considerable effort has been spent on studying themodular construction of Petri nets. This book studies the modular construction of nets, and in particular the top-down design of nets by action refinement. Suitable behavior descriptions are presented and special care is taken to justify these descriptions by showing that theyare necessary under reasonable specification requirements. In particular, itis shown that partial-order semantics is necessary to support action refinement.

Parallel and distributed computing are becoming increasingly important as cost-effective ways to achieve high computational performance. Symbolic computations are notable for their use of irregular data structures and hence parallel symbolic computing has its own distinctive set of technical challenges. The papers in this book are based on presentations made at a workshop at MIT in October 1992. They present results in a wide range of areas including: speculative computation, scheduling techniques, program development tools and environments, programming languages and systems, models of concurrency and distribution, parallel computer architecture, and symbolic applications.

The 1992 Parallel Architectures and Languages Europe conference continues the tradition - of a wide and representative international meeting of specialists from academia and industry in theory, design, and application of parallel computer systems - set by the previous PARLE conferences held in Eindhoven in 1987, 1989, and 1991. This volume contains the 52 regular and 25 poster papers that were selected from 187 submitted papers for presentation and publication. In addition, five invited lectures areincluded. The regular papers are organized into sections on: implementation of parallel programs, graph theory, architecture, optimal algorithms, graph theory and performance, parallel software components, data base optimization and modeling, data parallelism, formal methods, systolic approach, functional programming, fine grain parallelism, Prolog, data flow systems, network efficiency, parallel algorithms, cache systems, implementation of parallel languages, parallel scheduling in data base systems, semantic models, parallel data base machines, and language semantics.

This volume contains the proceedings of the Fourth Workshop on Languages andCompilers for Parallel Computing, held in Santa Clara, California, in August1991. The purpose of the workshop, held every year since 1988, is to bring together the leading researchers on parallel programming language designand compilation techniques for parallel computers. The papers in this book cover several important topics including: (1) languages and structures to represent programs internally in the compiler, (2) techniques to analyzeand manipulate sequential loops in order to generate a parallel version, (3)techniques to detect and extract fine-grain parallelism, (4) scheduling and memory-management issues in automatically generated parallel programs, (5) parallel programming language designs, and (6) compilation of explicitly parallel programs. Together, the papers give a good overview of the research projects underway in 1991 in this field.

The Austrian Center for Parallel Computation (ACPC) is a cooperative research organization founded in 1989 to promote research and education in the field of software for parallel computer systems. The areas in which the ACPC is active include algorithms, languages, compilers, programming environments, and applications for parallel and high-performance computing systems. This volume contains the proceedings of the First International Conference of the SCPC, held in Salzburg, Austria, September 30 - October 2, 1991. Authors from 15 countries submitted 66 papers, from which 28 were selected and presented at the conference. They appear in this volume together with 8 invited papers by distinguished researchers. Based on the quality of the papers presented and the response of the participants, the ACPC has decided to organize a conference every other year. The second ACPC conference will take place in Vienna in June 1993.

This book contains papers presented at a workshop on the use of parallel techniques in symbolic and algebraic computation held at Cornell University in May 1990. The eight papers in the book fall into three groups. The first three papers discuss particular programming substrates for parallel symbolic computation, especially for distributed memory machines. The next three papers discuss novel ways of computing with elements of finite fields and with algebraic numbers. The finite field technique is especially interesting since it uses the Connection Machine, a SIMD machine, to achievesurprising amounts of parallelism. One of the parallel computing substrates is also used to implement a real root isolation technique. One of the crucial algorithms in modern algebraic computation is computing the standard, or Gr-bner, basis of an ideal. The final two papers discuss two different approaches to speeding their computation. One uses vector processing on the Cray and achieves significant speed-ups. The other uses a distributed memory multiprocessor and effectively explores the trade-offs involved with different interconnect topologies of the multiprocessors.

This volume contains the proceedings of an international workshop on parallelism in inference systems held in Germany in December 1990. The topicof the workshop is still rather young and several papers in the book are overview articles intended to provide a first orientation toward some of the more intensively investigated subtopics. The main part of the book is a compilation of research papers on parallelization in special domains ofinference such as rewriting, automatic reasoning, logic programming, andconnectionist inference. Appended to the book is a collection of short project summaries received in response to a worldwide email call. The book is intended primarily for researchers working on inference systems who are interested in parallelizing their systems.

This volume contains most of the papers presented at the workshop on research directions in high-level parallel programming languages, held at Mont Saint-Michel, France, in June 1991. The motivation for organizing this workshop came from the emergence of a new class of formalisms for describing parallel computations in the last few years. Linda, Unity, Gamma, and the Cham are the most significant representatives of this new class. Formalisms of this family promote simple but powerful language features for describing data and programs. These proposals appeared in different contexts and were applied in different domains, and the goal of the workshop was to review the status of this new field and compare experiences. The workshop was organized into four main sessions: Unity, Linda, Gamma, and Parallel Program Design. The corresponding parts ofthe volume are introduced respectively by J. Misra, D. Gelernter, D. Le M tayer, and J.-P. Ban tre.

This volume presents the proceedings of a workshop on evolutionary models and strategies and another workshop on parallel processing, logic, organization, and technology, both held in Germany in 1989. In the search for new concepts relevant for parallel and distributed processing, the workshop on parallel processing included papers on aspects of space and time, representations of systems, non-Boolean logics, metrics, dynamics and structure, and superposition and uncertainties. The point was stressed that distributed representations of information may share features with quantum physics, such as the superposition principle and the uncertainty relations. Much of the volume contains material on general parallel processing machines, neural networks, and system-theoretic aspects. The material on evolutionary strategies is included because these strategies will yield important and powerful applications for parallel processing machines, and open the wayto new problem classes to be treated by computers.

This special volume contains the Proceedings of a Workshop on "Parallel Algorithms and Transputers for Optimization" which was held at the University of Siegen, on November 9, 1990. The purpose of the Workshop was to bring together those doing research on 2.lgorithms for parallel and distributed optimization and those representatives from industry and business who have an increasing demand for computing power and who may be the potential users of nonsequential approaches. In contrast to many other conferences, especially North-American, on parallel processing and supercomputers the main focus of the contributions and discussion was "problem oriented". This view reflects the following philosophy: How can the existing computing infrastructure (PC's, workstations, local area networks) of an institution or a company be used for parallel and/or distributed problem solution in optimization. This volume of the LECfURE NOTES ON ECONOMICS AND MA THEMA TICAL SYSTEMS contains most of the papers presented at the workshop, plus some additional invited papers covering other important topics related to this workshop. The papers appear here grouped according to four general areas. (1) Solution of optimization problems using massive parallel systems (data parallelism). The authors of these papers are: Lootsma; Gehne. (II) Solution of optimization problems using coarse-grained parallel approaches on multiprocessor systems (control parallelism). The authors of these papers are: Bierwirth, Mattfeld, and Stoppler; Schwartz; Boden, Gehne, and Grauer; and Taudes and Netousek.

Artificial neural networks are massively parallel interconnected networks ofsimple elements which are intended to interact with the objects of the real world in the same way as biological nervous systems do. Interest in these networks is due to the opinion that they are able to perform tasks like image and speech recognition that have only been implemented in limited ways by traditional computing methods. This book includes invited lectures and the full contributions to the International Workshop onArtificial Neural Networks held in Granada, Spain, September 17-19, 1991. The workshop was sponsored by the IEEE Computer Society, the Spanish Association for Computing and Automatics, and the University of Granada. The contributions were selected by an international program committee; the authors of the papers come from 12 countries. The book is organized in six sections, covering: - Neural network theories and neural models - Biological perspectives - Neural network architectures and algorithms - Software developments and tools - Hardware implementations - Applications.

This volume contains the proceedings of the 16th International Symposium on Mathematical Foundations of Computer Science, MFCS '91, held in Kazimierz Dolny, Poland, September 9-13, 1991. The series of MFCS symposia, organized alternately in Poland and Czechoslovakia since 1972, has a long and well established tradition. The purpose of the series is to encourage high-quality research in all branches of theoretical computer science and to bring together specialists working actively in the area. Principal areas of interest in this symposium include: software specification and development, parallel and distributed computing, logic and semantics of programs, algorithms, automata and formal languages, complexity and computability theory, and others. The volume contains 5 invited papers by distinguished scientists and 38 contributions selected from a total of 109 submitted papers.

With the appearance of massively parallel computers, increased attention has been paid to algorithms which rely upon analogies to natural processes. This development defines the scope of the PPSN conference at Dortmund in 1990 whose proceedings are presented in this volume. The subjects treated include: - Darwinian methods such as evolution strategies and genetic algorithms; - Boltzmann methods such as simulated annealing; - Classifier systems and neural networks; - Transfer of natural metaphors to artificial problem solving. The main objectives of the conference were: - To gather theoretical results about and experimental comparisons between these algorithms, - To discuss various implementations on different parallel computer architectures, - To summarize the state of the art in the field, which was previously scattered widely both among disciplines and geographically.