Elements Of Distributed Computing

Mit der Verfügbarkeit verteilter Systeme wächst der Bedarf an einer fundamentalen Diskussion dieses Gebiets. Hier ist sie! Abgedeckt werden die grundlegenden Konzepte wie Zeit, Zustand, Gleichzeitigkeit, Reihenfolge, Kenntnis, Fehler und Übereinstimmung. Die Betonung liegt auf der Entwicklung allgemeiner Mechanismen, die auf eine Vielzahl von Problemen angewendet werden können. Sorgfältig ausgewählte Beispiele (Taktgeber, Sperren, Kameras, Sensoren, Controller, Slicer und Syncronizer) dienen gleichzeitig der Vertiefung theoretischer Aspekte und deren Umsetzung in die Praxis. Alle vorgestellten Algorithmen werden mit durchschaubaren, induktionsbasierten Verfahren bewiesen.

Concurrent and Distributed Computing in Java addresses fundamental concepts in concurrent computing with Java examples. The book consists of two parts. The first part deals with techniques for programming in shared-memory based systems. The book covers concepts in Java such as threads, synchronized methods, waits, and notify to expose students to basic concepts for multi-threaded programming. It also includes algorithms for mutual exclusion, consensus, atomic objects, and wait-free data structures. The second part of the book deals with programming in a message-passing system. This part covers resource allocation problems, logical clocks, global property detection, leader election, message ordering, agreement algorithms, checkpointing, and message logging. Primarily a textbook for upper-level undergraduates and graduate students, this thorough treatment will also be of interest to professional programmers.

Distributed Computing is rapidly becoming the principal computing paradigm in diverse areas of computing, communication, and control. Processor clusters, local and wide area networks, and the information highway evolved a new kind of problems which can be solved with distributed algorithms. In this textbook a variety of distributed algorithms are presented independently of particular programming languages or hardware, using the graphically suggestive technique of Petri nets which is both easy to comprehend intuitively and formally rigorous. By means of temporal logic the author provides surprisingly simple yet powerful correctness proofs for the algorithms. The scope of the book ranges from distributed control and synchronization of two sites up to algorithms on any kind of networks. Numerous examples show that description and analysis of distributed algorithms in this framework are intuitive and technically transparent.

This book studies algorithmic issues associated with cooperative execution of multiple independent tasks by distributed computing agents including partitionable networks. It provides the most significant algorithmic solution developed and available today for do-all computing for distributed systems (including partitionable networks), and is the first monograph that deals with do-all computing for distributed systems. The book is structured to meet the needs of a professional audience composed of researchers and practitioners in industry. This volume is also suitable for graduate-level students in computer science.

Distributed processing has a strong theoretical foundation, but many day-to-day practitioners make limited use of the advantages this theory can give them. The result includes unreliable systems with obscure and intermittent failures, that can cost time, money and in extreme cases, lives. Reliable construction of distributed and concurrent systems must incorporate theory in practice. This book provides a concise presentation of the theory closely linked to the practical realization of these concepts. This highly practical presentation contains all the elements needed for a complete development of a distributed system. The book includes examples from C, Java and Eiffel, and sample code is available online.

Designed for introductory parallel computing courses at the advanced undergraduate or beginning graduate level, Elements of Parallel Computing presents the fundamental concepts of parallel computing not from the point of view of hardware, but from a more abstract view of algorithmic and implementation patterns. The aim is to facilitate the teaching of parallel programming by surveying some key algorithmic structures and programming models, together with an abstract representation of the underlying hardware. The presentation is friendly and informal. The content of the book is language neutral, using pseudocode that represents common programming language models. The first five chapters present core concepts in parallel computing. SIMD, shared memory, and distributed memory machine models are covered, along with a brief discussion of what their execution models look like. The book also discusses decomposition as a fundamental activity in parallel algorithmic design, starting with a naive example, and continuing with a discussion of some key algorithmic structures. Important programming models are presented in depth, as well as important concepts of performance analysis, including work-depth analysis of task graphs, communication analysis of distributed memory algorithms, key performance metrics, and a discussion of barriers to obtaining good performance. The second part of the book presents three case studies that reinforce the concepts of the earlier chapters. One feature of these chapters is to contrast different solutions to the same problem, using select problems that aren't discussed frequently in parallel computing textbooks. They include the Single Source Shortest Path Problem, the Eikonal equation, and a classical computational geometry problem: computation of the two-dimensional convex hull. After presenting the problem and sequential algorithms, each chapter first discusses the sources of parallelism then surveys parallel algorithms.

In 1992 we initiated a research project on large scale distributed computing systems (LSDCS). It was a collaborative project involving research institutes and universities in Bologna, Grenoble, Lausanne, Lisbon, Rennes, Rocquencourt, Newcastle, and Twente. The World Wide Web had recently been developed at CERN, but its use was not yet as common place as it is today and graphical browsers had yet to be developed. It was clear to us (and to just about everyone else) that LSDCS comprising several thousands to millions of individual computer systems (nodes) would be coming into existence as a consequence both of technological advances and the demands placed by applications. We were excited about the problems of building large distributed systems, and felt that serious rethinking of many of the existing computational paradigms, algorithms, and structuring principles for distributed computing was called for. In our research proposal, we summarized the problem domain as follows: “We expect LSDCS to exhibit great diversity of node and communications capability. Nodes will range from (mobile) laptop computers, workstations to supercomputers. Whereas mobile computers may well have unreliable, low bandwidth communications to the rest of the system, other parts of the system may well possess high bandwidth communications capability. To appreciate the problems posed by the sheer scale of a system comprising thousands of nodes, we observe that such systems will be rarely functioning in their entirety.