Date of Graduation
5-2025
Document Type
Thesis
Degree Name
Bachelor of Science in Electrical Engineering
Degree Level
Undergraduate
Department
Electrical Engineering
Advisor/Mentor
Dix, Jeff
Abstract
With the current era of AI technology, the era of single instruction multiple data has become an increasingly viable solution to accelerate training. The problem is that while software to use GPUs and other hardware accelerators, designing GPUs and ASIC devices has become increasingly more expensive and there aren’t great examples of generic GPUs that anyone can use and modify. In this thesis, there are four design considerations that will be discussed and how they affect the result of a generic GPU. The four considerations that were talked about in the thesis are, word width, arithmetic type, number of stages, and number of memory paths. Throughout, decisions will be made about these design decisions to create an example GPU that will be tested at the end. The set up chosen is to make the GPU an 8-bit word width, integer-based arithmetic computations, single staged pipeline, with 8 memory paths onto and off the chip. The result was a generic GPU that can complete basic calculations.
Keywords
GPU; Design; Hardware; Memory; Bandwidth
Citation
DeVilliers, N. M. (2025). Design Considerations of a GPU. Electrical Engineering and Computer Science Undergraduate Honors Theses Retrieved from https://scholarworks.uark.edu/elcsuht/15
Included in
Computational Engineering Commons, Computer and Systems Architecture Commons, Digital Circuits Commons, Hardware Systems Commons, Other Electrical and Computer Engineering Commons, VLSI and Circuits, Embedded and Hardware Systems Commons
