Giorgio Buttazzo is Full Professor of Computer Engineering at the Scuola Superiore Sant'Anna of Pisa. He graduated in Electronic Engineering at the University of Pisa in 1985, received a Master in Computer Science at the University of Pennsylvania in 1987, and a Ph.D. in Computer Engineering at the Scuola Superiore Sant'Anna of Pisa in 1991. From 1987 to 1988, he worked on active perception and real-time control at the G.R.A.S.P. Laboratory of the University of Pennsylvania, Philadelphia. From 1991 to 1998, he held a position of Assistant Professor at the Scuola Superiore Sant'Anna of Pisa, where he founded and coordinated the RETIS Laboratory on real-time systems. From 1998 to 2005, he held a position of Associate Professor at the University of Pavia, where he directed the robotics laboratory of the Computer Science department. His main research interests include real-time operating systems, dynamic scheduling algorithms, quality of service control, multimedia systems, advanced robotics applications, and neural networks.

This monograph presents state-of-the-art techniques for building flexible, predictable soft real-time systems, that, in comparison to hard real-time systems, optimize resources and reduce costs. It provides background and basic terminology of soft real-time computing, and goes on to discuss overload management techniques, temporal protection, the execution of multi-thread applications, synchronization protocols, reclaiming techniques and quality of service issues. Accompanied by numerous illustrations and comprehensive references, this text is an invaluable reference for researchers, students and system developers alike.

Preface
1: Introduction
1.1 Basic terminology
1.2 From hard to soft real-time systems
1.3 Providing support for soft real-time systems
2: Overload Management
2.1 Introduction
2.2 Load definitions
2.3 Admission control methods
2.4 Performance degradation methods
2.5 Service adaptation
2.6 Job skipping
2.7 Period adaptation
3: Temporal Protection
3.1 Problems without temporal protection
3.2 Providing temporal protection
3.3 The GPS model
3.4 Proportional share scheduling
3.5 Resource reservation techniques
3.6 Resource reservations in dynamic priority systems
3.7 Temporal guarantees
3.8 Resource reservations in operating system kernels
4: Multi-thread Applications
4.1 The thread model
4.2 Global approaches
4.3 Partition-based approaches
4.4 Concluding remarks and open problems
5: Synchronization Protocols
5.1 Terminology and notation
5.2 Shared resource in real-time systems
5.3 Synchronization protocols for hard real-time systems
5.4 Shared resources in soft real-time systems
5.5 Extending resource reservation with the SRP
5.6 Resource constraints in dynamic systems
5.7 Concluding remarks
6: Resource Reclaiming
6.1 Problems with reservations
6.2 The CASH algorithm
6.3 The GRUB algorithm
6.4 Other forms of reclaiming
7: QoS Management
7.1 The QoS-based resource allocation model
7.2 Static vs. dynamic resource management
7.3 Integrating design and scheduling issues
7.4 Smooth rate adaptation
8: Feedback Scheduling
8.1 Controlling the number of missed deadlines
8.2 Adaptive reservations
8.3 Application level adaptation
8.4 Workload estimators
9: Stochastic Scheduling
9.1 Background and definitions
9.2 Statistical analysis of classical algorithms
9.3 Real-time queueing theory
9.4 Novel algorithms for stochastic scheduling
9.5 Reservations and stochastic guarantee
References
Index

Hard real-time systems are very predictable, but not sufficiently flexible to adapt to dynamic situations. They are built under pessimistic assumptions to cope with worst-case scenarios, so they often waste resources. Soft real-time systems are built to reduce resource consumption, tolerate overloads and adapt to system changes. They are also more suited to novel applications of real-time technology, such as multimedia systems, monitoring apparatuses, telecommunication networks, mobile robotics, virtual reality, and interactive computer games. This unique monograph provides concrete methods for building flexible, predictable soft real-time systems, in order to optimize resources and reduce costs. It is an invaluable reference for developers, as well as researchers and students in Computer Science.