Abstract
dc:description.abstractDiminishing returns in technology scaling has motivated a resurgence of exploration into new computer architectures. While Coarse Grained Reconfigurable Arrays show promise in accelerating commonly used complex operations, their overall capacity remains fairly limited. While there is pressure on general purpose systems to support wide operations, the typicalworkload mostly exercises the lower 10 to 15 bits. This leaves most of the array on and unused during normal operation. This thesis presents adaptive digit-serial arithmetic as a plug-and-play method to support a variety of bitwidth requirements, showing decreased energy and area alongside increased throughput.
Degree
thesis:*- Name thesis:degree_name
- Ph. D., Electrical and Computer Engineering
- Level thesis:degree_level
- Doctor of Philosophy
- Discipline thesis:degree_discipline
- Electrical and Computer Engineering
- Grantor
- Cornell University
- Year dc:date.issued
- 2020
Author and committee
dc:creator, dc:contributor.*- Author dc:creator
-
- Bingham, Edward Arthur
- Chair dc:contributor.chair
-
- Manohar, Rajit
- Committee members dc:contributor.committeemember
-
- Batten, Christopher
- Zhang, Zhiru
Subjects
dc:subject × 4Rights
- Language dc:language.iso
- en
Identifiers
dc:identifier.*- Dc Identifier Bibid
- 16969961
- Dc Identifier Other
-
ProQuest Submission ID: 12312
ProQuest Publication ID: 28154646