Multiprocessor Scheduling For Real Time Systems

This book provides a comprehensive overview of both theoretical and pragmatic aspects of resource-allocation and scheduling in multiprocessor and multicore hard-real-time systems. The authors derive new, abstract models of real-time tasks that capture accurately the salient features of real application systems that are to be implemented on multiprocessor platforms, and identify rules for mapping application systems onto the most appropriate models. New run-time multiprocessor scheduling algorithms are presented, which are demonstrably better than those currently used, both in terms of run-time efficiency and tractability of off-line analysis. Readers will benefit from a new design and analysis framework for multiprocessor real-time systems, which will translate into a significantly enhanced ability to provide formally verified, safety-critical real-time systems at a significantly lower cost.

The theory of supervisory control of discrete-event systems has been applied to real-time systems. The contribution of our proposed work lies in the development of a formal constructive method for controlling the preemptive execution of real-time tasks on both uniprocessor and multiprocessor systems. The set of all possible timed traces of the system is specified by a discrete timed automaton, where each transition is associated with an event occurrence or the passage of one unit of time. This approach allows a unified view of scheduling theory based on the timing analysis of models of real-time applications, meaning that the problem of determining schedulability and finding out a suitable scheduling algorithm are assumed to be intermingled issues, with the solution of one in turn is a solution to the other too. First, a framework for designing universal schedulers for real-time tasks on uniprocessors based on Supervisory Control Theory (SCT) is presented. For this purpose, priorities are introduced in SCT and applied to the setting of discrete timed automata in order to develop a formal and unified framework for task scheduling on a single CPU. A universal scheduler nondeterministically selects a task for execution in such a way that all timing constraints are met in a minimally restrictive fashion, while it contains all feasible deterministic scheduling policies. We then extend that framework by providing a formal constructive method for controlling the preemptive and migrative execution of hard real-time tasks while scheduling them on a set of uniform processors. The methodology relies on the idea that the model of the scheduled system can be obtained by successive and appropriate restrictions of controllable actions of a model representing the real-time application. In uniform multiprocessors, each processor is characterized by its own computing capacity, with the interpretation that a task that executes on a processor of computing capacity s for? time units completes s? units of execution. Since we represented explicitly discrete time in our scheduler design, model sizes were considerably large. The complexity in the synthesis of a scheduler using supervisory control [62] stems from the fact that, with the synchronous product, the number of states of a composite Timed Discrete-Event System (TDES) increases exponentially with the number of real-time tasks. We have attempted to alleviate some of the state explosion problems we had faced while designing schedulers for real-time systems using supervisory control of discrete-event systems framework [39, 41], by providing an informal procedure to design schedulers with reduced state space.

The aim of this handbook is to summarize the recent rapidly developed real-time computing technologies, from theories to applications. This handbook benefits the readers as a full and quick technical reference with a high-level historic review of technology, detailed technical descriptions and the latest practical applications. In general, the handbook is divided into three main parts (subjected to be modified): theory, design, and application covering different but not limited to the following topics: - Real-time operating systems - Real-time scheduling - Timing analysis - Programming languages and run-time systems - Middleware systems - Design and analysis tools - Real-time aspects of wireless sensor networks - Energy aware real-time methods

Techniques for Optimizing Multiprocessor Implementations of Signal Processing Applications An indispensable component of the information age, signal processing is embedded in a variety of consumer devices, including cell phones and digital television, as well as in communication infrastructure, such as media servers and cellular base stations. Multiple programmable processors, along with custom hardware running in parallel, are needed to achieve the computation throughput required of such applications. Reviews important research in key areas related to the multiprocessor implementation of multimedia systems Embedded Multiprocessors: Scheduling and Synchronization, Second Edition presents architectures and design methodologies for parallel systems in embedded digital signal processing (DSP) applications. It discusses application modeling techniques for multimedia systems, the incorporation of interprocessor communication costs into multiprocessor scheduling decisions, and a modeling methodology (the synchronization graph) for multiprocessor system performance analysis. The book also applies the synchronization graph model to develop hardware and software optimizations that can significantly reduce the interprocessor communication overhead of a given schedule. Chronicles recent activity dealing with single-chip multiprocessors and dataflow models This edition updates the background material on existing embedded multiprocessors, including single-chip multiprocessors. It also summarizes the new research on dataflow models for signal processing that has been carried out since the publication of the first edition. Harness the power of multiprocessors This book explores the optimization of interprocessor communication and synchronization in embedded multiprocessor systems. It shows you how to design multiprocessor computer systems that are streamlined for multimedia applications.

Due to the decreasing production costs of IT systems, applications that had to be realised as expensive PCBs formerly, can now be realised as a system-on-chip. Furthermore, low cost broadband communication media for wide area communication as well as for the realisation of local distributed systems are available. Typically the market requires IT systems that realise a set of specific features for the end user in a given environment, so called embedded systems. Some examples for such embedded systems are control systems in cars, airplanes, houses or plants, information and communication devices like digital TV, mobile phones or autonomous systems like service- or edutainment robots. For the design of embedded systems the designer has to tackle three major aspects: The application itself including the man-machine interface, The (target) architecture of the system including all functional and non-functional constraints and, the design methodology including modelling, specification, synthesis, test and validation. The last two points are a major focus of this book. This book documents the high quality approaches and results that were presented at the International Workshop on Distributed and Parallel Embedded Systems (DIPES 2000), which was sponsored by the International Federation for Information Processing (IFIP), and organised by IFIP working groups WG10.3, WG10.4 and WG10.5. The workshop took place on October 18-19, 2000, in Schloß Eringerfeld near Paderborn, Germany. Architecture and Design of Distributed Embedded Systems is organised similar to the workshop. Chapters 1 and 4 (Methodology I and II) deal with different modelling and specification paradigms and the corresponding design methodologies. Generic system architectures for different classes of embedded systems are presented in Chapter 2. In Chapter 3 several design environments for the support of specific design methodologies are presented. Problems concerning test and validation are discussed in Chapter 5. The last two chapters include distribution and communication aspects (Chapter 6) and synthesis techniques for embedded systems (Chapter 7). This book is essential reading for computer science researchers and application developers.

The first book to provide a comprehensive overview of the subject rather than a collection of papers. The author is a recognized authority in the field as well as an outstanding teacher lauded for his ability to convey these concepts clearly to many different audiences. A handy reference for practitioners in the field.

This classroom-tested textbook describes the design and implementation of software for distributed real-time systems, using a bottom-up approach. The text addresses common challenges faced in software projects involving real-time systems, and presents a novel method for simply and effectively performing all of the software engineering steps. Each chapter opens with a discussion of the core concepts, together with a review of the relevant methods and available software. This is then followed with a description of the implementation of the concepts in a sample kernel, complete with executable code. Topics and features: introduces the fundamentals of real-time systems, including real-time architecture and distributed real-time systems; presents a focus on the real-time operating system, covering the concepts of task, memory, and input/output management; provides a detailed step-by-step construction of a real-time operating system kernel, which is then used to test various higher level implementations; describes periodic and aperiodic scheduling, resource management, and distributed scheduling; reviews the process of application design from high-level design methods to low-level details of design and implementation; surveys real-time programming languages and fault tolerance techniques; includes end-of-chapter review questions, extensive C code, numerous examples, and a case study implementing the methods in real-world applications; supplies additional material at an associated website. Requiring only a basic background in computer architecture and operating systems, this practically-oriented work is an invaluable study aid for senior undergraduate and graduate-level students of electrical and computer engineering, and computer science. The text will also serve as a useful general reference for researchers interested in real-time systems.

Real-time systems are used in a wide range of applications, including control, sensing, multimedia, etc. Scheduling is a central problem for these computing/communication systems since responsible of software execution in a timely manner. This book provides state of knowledge in this domain with special emphasis on the key results obtained within the last decade. This book addresses foundations as well as the latest advances and findings in Real-Time Scheduling, giving all references to important papers. But nevertheless the chapters will be short and not overloaded with confusing details. Coverage includes scheduling approaches for mono-core as well as multi-core platforms, dependent tasks, networks, and notably very tremendous recent advances in scheduling of energy constrained embedded systems. Other sophisticated issues such as feedback control scheduling and timing analysis of critical applications are also addressed. This volume can serve as a textbook for courses on the topic in bachelor and in more advanced master programs. It also provides a reference for computer scientists and engineers involved in the design or the development of Cyber-Physical Systems which require up-to-date real-time scheduling solutions.