The hierarchical decomposition of programs into smaller ones is generally considered imperative to master the complexity of large programs. The impact of this principle of program decomposition on the specification and verification of parallel executed programs is the subject of this monograph. Two important yardsticks for verification methods, those of compositionality and modularity, are made precise. The problem of reusing software is addressed by the introduction of the notion of specification adaptation. Within this context, different methods for specifying the observable behavior with respect to partial correctness of communicating processes are considered, and in particular the contrast between the "programs are predicates" and the "programs are predicate transformers" paradigms is shown. The associated formal proof systems are proven sound and complete in various senses with respect to the denotational semantics of the programming language, and they are related to each other to give an in-depth comparison between the different styles of program verification. The programming language TNP used here is near to actual languages like Occam. It combines CCS/CSP style communication based programming with state based programming, and allows dynamically expanding and shrinking networks of processes.

This is a systematic and comprehensive introduction both to compositional proof methods for the state-based verification of concurrent programs, such as the assumption-commitment and rely-guarantee paradigms, and to noncompositional methods, whose presentation culminates in an exposition of the communication-closed-layers (CCL) paradigm for verifying network protocols. Compositional concurrency verification methods reduce the verification of a concurrent program to the independent verification of its parts. If those parts are tightly coupled, one additionally needs verification methods based on the causal order between events. These are presented using CCL. The semantic approach followed here allows a systematic presentation of all these concepts in a unified framework which highlights essential concepts. This 2001 book is self-contained, guiding the reader from advanced undergraduate level. Every method is illustrated by examples, and a picture gallery of some of the subject's key figures complements the text.

This is a systematic and comprehensive introduction both to compositional proof methods for the state-based verification of concurrent programs, such as the assumption-commitment and rely-guarantee paradigms, and to noncompositional methods, whose presentation culminates in an exposition of the communication-closed-layers (CCL) paradigm for verifying network protocols. Compositional concurrency verification methods reduce the verification of a concurrent program to the independent verification of its parts. If those parts are tightly coupled, one additionally needs verification methods based on the causal order between events. These are presented using CCL. The semantic approach followed here allows a systematic presentation of all these concepts in a unified framework which highlights essential concepts. The book is self-contained, guiding the reader from advanced undergraduate level to the state-of-the-art. Every method is illustrated by examples, and a picture gallery of some of the subject's key figures complements the text.