الفهرس | Only 14 pages are availabe for public view |
Abstract The thesis is divided into five chapters as listed below: Chapter 1 Introduction: provides an introduction to the problem, the motivation for this work, objectives and the approach used to achieve the outcome of this thesis. Chapter 2 Literature Review: provides a literature review of the real number with their different representations and the problems of representing those real numbers on the digital systems, also review the previous work done by other researchers to have error free transformation algorithms for floating-point calculations to have accurate operations. Chapter 3 Acceleration of accurate floating point operations using Single Instruction Multiple Data: provides detailed explanation of the work done to accelerate the MN-System using both SSE and AVX, and complexity analysis of the accelerated operations. Chapter 4 Experimental results: provides some case studies where the performance of the accelerated version of MN-System was tested in the applications that require high accuracy and speedup was measured against the original MN version. Chapter 5 Conclusion and Future Work: contains the conclusion of the whole work documented in this thesis and also Future work suggestions that can enhance the performance. |