دانلود کتاب High Performance Computing in Science and Engineering '21: Transactions of the High Performance Computing Center, Stuttgart

This book presents the state-of-the-art in supercomputer simulation. It includes the latest findings from leading researchers using systems from the High Performance Computing Center Stuttgart (HLRS) in 2021. The reports cover all fields of computational science and engineering ranging from CFD to computational physics and from chemistry to computer science with a special emphasis on industrially relevant applications. Presenting findings of one of Europe’s leading systems, this volume covers a wide variety of applications that deliver a high level of sustained performance. The book covers the main methods in high-performance computing. Its outstanding results in achieving the best performance for production codes are of particular interest for both scientists and engineers. The book comes with a wealth of color illustrations and tables of results.

Developing scalable algorithms is a challenging task that requires careful analysis and extensive experimental evaluation. CPU technology shifts to deliver increasing amounts of cores with relatively low clock rates, since these are cheaper to produce and operate. Parallelizing computationally intensive algorithms at the core of all applications is therefore an important research topic. Developing distributed algorithms on cluster computers such as the ForHLR II is an integral part of this scalability challenge. Our focus is especially on discrete algorithms, such as graph partitioning, text search, and propositional satisfiability (SAT) solving.

In previous years, we studied distributed online sorting and string sorting in our Big Data toolkit Thrill, developed a scalable approach to edge partitioning, developed and evaluated algorithms for maintaining uniform and weighted samples over distributed data streams (reservoir sampling), and designed new approaches to massively parallel malleable job scheduling applied to propositional satisfiability (SAT) solving. Thrill – A High-Performance Big Data Framework in C++. We improved the sorting algorithm of Thrill, our next-generation C++ framework for distributed Big Data batch processing on a cluster of homogeneous machines which enables writing distributed applications conveniently using “dataflow” graph-like computations.