TU Darmstadt / ULB / TUbiblio

Optimizing a geometric algebra compiler for parallel architectures using a table-based approach

Steinmetz, Christian (2011):
Optimizing a geometric algebra compiler for parallel architectures using a table-based approach.
TU Darmstadt, [Bachelor Thesis]

Abstract

Geometric Algebra is an algebra, which permits a calculation in an intuitive and an elegant way. Since a CPU is not able to calculate with Geometric Algebra directly, one need a compiler. Gaalop is such a compiler, which compiles a CLUScript, so that it can be used for instance in C-programs or FPGA implementations. Gaalop recently needs Maple for optimizing Geometric Algebra algorithms. The first main issue of this work is to implement a table-based approach in Gaalop in order to eliminate the need of Maple, since Maple is a commercial software. As a second main issue, we implemented a tool in Gaalop for optimizing Geometric Algebra algorithms for using on parallel architectures.

Item Type: Bachelor Thesis
Erschienen: 2011
Creators: Steinmetz, Christian
Title: Optimizing a geometric algebra compiler for parallel architectures using a table-based approach
Language: English
Abstract:

Geometric Algebra is an algebra, which permits a calculation in an intuitive and an elegant way. Since a CPU is not able to calculate with Geometric Algebra directly, one need a compiler. Gaalop is such a compiler, which compiles a CLUScript, so that it can be used for instance in C-programs or FPGA implementations. Gaalop recently needs Maple for optimizing Geometric Algebra algorithms. The first main issue of this work is to implement a table-based approach in Gaalop in order to eliminate the need of Maple, since Maple is a commercial software. As a second main issue, we implemented a tool in Gaalop for optimizing Geometric Algebra algorithms for using on parallel architectures.

Divisions: 20 Department of Computer Science
Date Deposited: 30 Mar 2017 06:47
Alternative Abstract:
Alternative abstract Language

Geometric Algebra ist eine Algebra, die eine Berechnung in einer intuitiven und eleganten Weise erlaubt. Da eine CPU nicht in der Lage ist, direkt mit geometrischer Algebra zu rechnen, benötigt man einen Compiler. Gaalop ist so ein Compiler, der ein CLUScript kompiliert, so dass es zum Beispiel in C-Programmen oder FPGA-Implementierung genutzt werden kann. Bisher brauchte Gaalop Maple, um Geometrische Algebra Algorithmen zu optimieren. Der erste Kernpunkt dieser Arbeit ist, einen tabellenbasierten Ansatz in Gaalop zu implementieren, um die Notwendigkeit von Maple zu eliminieren, da Maple eine kommerzielle Software ist. Als zweiter Kernpunkt wurde ein Werkzeug in Gaalop implementiert, um Geometrische Algebra Algorithmen für die Benutzung auf parallelen Architekturen zu optimieren.

German
Export:
Suche nach Titel in: TUfind oder in Google
Send an inquiry Send an inquiry

Options (only for editors)
Show editorial Details Show editorial Details