A unique investigation of the state of the art in design,architectures, and implementations of advanced computationalinfrastructures and the applications they support
Emerging large-scale adaptive scientific and engineeringapplications are requiring an increasing amount of computing andstorage resources to provide new insights into complex systems. Dueto their runtime adaptivity, these applications exhibit complicatedbehaviors that are highly dynamic, heterogeneous, andunpredictable--and therefore require full-fledgedcomputational infrastructure support for problem solving, runtimemanagement, and dynamic partitioning/balancing. This book presentsa comprehensive study of the design, architecture, andimplementation of advanced computational infrastructures as well asthe adaptive applications developed and deployed using theseinfrastructures from different perspectives, including systemarchitects, software engineers, computational scientists, andapplication scientists. Providing insights into recent researchefforts and projects, the authors include descriptions andexperiences pertaining to the realistic modeling of adaptiveapplications on parallel and distributed systems.
The first part of the book focuses on high-performance adaptivescientific applications and includes chapters that describehigh-impact, real-world application scenarios in order to motivatethe need for advanced computational engines as well as to outlinetheir requirements. The second part identifies popular and widelyused adaptive computational infrastructures. The third part focuseson the more specific partitioning and runtime management schemesunderlying these computational toolkits.
* Presents representative problem-solving environments andinfrastructures, runtime management strategies, partitioning anddecomposition methods, and adaptive and dynamic applications
* Provides a unique collection of selected solutions andinfrastructures that have significant impact with sufficientintroductory materials
* Includes descriptions and experiences pertaining to therealistic modeling of adaptive applications on parallel anddistributed systems
The cross-disciplinary approach of this reference delivers acomprehensive discussion of the requirements, design challenges,underlying design philosophies, architectures, andimplementation/deployment details of advanced computationalinfrastructures. It makes it a valuable resource for advancedcourses in computational science and software/systems engineeringfor senior undergraduate and graduate students, as well as forcomputational and computer scientists, software developers, andother industry professionals.