This book is an organized proposal of the various components of a system able to integrate grid computing and the semantic Web, and is mostly based on the research activity of the Advanced Computing and System (CCNT) Laboratory at Zhejiang University in China.

The ten chapters of the book are organized in three parts: background information and general definitions (chapters 1 to 4); a presentation of more detailed problems and their solutions, provided by the authors in their DartGrid system (chapters 5 to 8); and applications (chapter 9 and 10).

The introduction includes a general presentation, terminology, and standards. Wu and Chen discuss how grid computing, originally developed for computation-intensive programs, provided the basis for “coordinated resource sharing and problem solving in dynamic, multi-institutional virtual organization” computing. A basic milestone has been the convergence of grid and Web services standards in the Web services resource framework (WSRF) family of specifications. The semantic Web, pushed by the needs of e-science, added the concept of metadata to grid computing and provided new languages and standards, such as the Web ontology language (OWL).

Chapter 2 is about knowledge representation. It reviews classical themes and approaches in artificial intelligence (AI)--mathematical logic, semantic network, and frames--and continues with ontology and description logic. Then, it shows how to integrate the above concepts in the semantic grid, starting with Extensible Markup Language (XML), and then adding layers with the resource definition framework (RDF) schema and OWL. It also presents the main application under development by the authors, traditional Chinese medicine (TCM), and its ontology.

Chapter 3 presents the dynamic problem-solving issue for the semantic grid. After providing terminology, it defines and develops an architecture for problem solving based on ontologies. First, the ontology is extracted and built, and then it is applied to problem solving. Distributed problem solving grows from sub-ontologies, distributed over the nodes in the grid. The reuse of ontologies presents problems similar to the case-based reasoning (CBR) approach, and is originally developed here in a model based on sub-ontologies manipulation.

Chapter 4 is about maintaining trust over the semantic grid. The authors present a trust model for scenarios, such as finding a suitable service provider and choosing it on the basis of its reputation. A closed trust model and an open trust model, built on probability theory, provide an answer to the problem.

Chapter 5 presents the data integration problem: namely, the integration of relational databases and the semantic grid environment. Wu and Chen present their algorithm based on description logic.

In chapter 6, service flow management is defined and evaluated. The model is derived from semantic-based service discovery and uses OWL.

Data mining and knowledge discovery in the semantic grid is the topic of chapter 7. Extracting useful patterns from data is one of the main targets of e-science. Starting from knowledge discovery in databases (KDD), the grid technology offers the possibility to extend it to very large, heterogeneous, and dynamic databases. A general architecture is presented and TCM serves again as an example.

The rest of the book is devoted mostly to the DartGrid application. Chapter 8 describes the implementation, while chapter 9 is about TCM. Another application for intelligent transportation systems, a case study of the Yangtze River Delta, is presented in chapter 10.

There are no conclusions. The references are listed at the end of each chapter. A short analytical index is included.

A notable point of this book is the organized presentation of many topics that occur in such a large area. A large part of the book is tutorial in nature, and combines a large body of references. Most of the themes have been previously approached in a variety of funded projects, such as the OntoGrid of the European Union (EU), but the material is still sparse in the various reports and papers. The most important Web sites about semantic grids also offer material but, again, sparsely. This book is a definite asset for developers and students. However, the topics that relate more to applications are not illustrated at a level that would make them directly usable, and the solutions proposed are still incomplete (as recognized by the authors).