Vector models for data-parallel computing

摘要: "Vector Models for Data-Parallel Computing "describes a model of parallelism that extends and formalizes the on which Connection Machine other supercomputers are based. It presents many algorithms based model, ranging from graph to numerical algorithms, argues data-parallel models not only practical can be applied surprisingly wide variety problems, they also well suited very-high-level languages lead concise clear description their complexity. Many author's ideas have been incorporated into instruction set currently running Machine.The book includes definition parallel vector machine; an extensive uses scan (also called parallel-prefix) operations; introduction segmented data structures trees, graphs, grids; computational-geometry, graph, sorting algorithms; techniques compiling nested parallelism; compiler Paralation Lisp; details implementation operations.Guy E. Blelloch is Assistant Professor Computer Science Principal Investigator with Super Compiler Advanced Language project at Carnegie Mellon University.Contents: Introduction. Parallel Vector Models. The Scan Primitives. Computational-Geometry Algorithms. Graph Numerical Languages Compilers. Correction-Oriented Languages. Flattening Nested Parallelism. A Lisp. Paralation-Lisp Code. Model. Data Structures. Implementing Operations. Conclusions. Glossary.