2026-09-22T07:06:58Z https://www.ideals.illinois.edu/oai-pmh

oai:www.ideals.illinois.edu:2142/78785 2023-07-11 col_2142_5131 col_2142_8888 com_2142_5130 com_2142_8887 com_2142_234

Patel, Sanjay J. Patel, Sanjay J. Chen, Deming Frank, Matthew I. Lumetta, Steven S. Johnson, Matthew Robert 2015-07-22T22:46:03Z 2015-07-22T22:46:03Z 2017-07-23T09:15:27Z 2015-05 2015-04-24 Power and energy consumption have become important for all computers, but the tools used to measure and optimize power on physical hardware lag far behind performance focused tools. Existing measurement apparata have low analog bandwidth, do not explicitly correlate power data with processor activity, and are not explained in sufficient detail to quantify uncertainty in their data. We present the design, implementation, and application of Jouler’s Loupe, a measurement device that overcomes these obstacles and enables a new generation of fast, fundamentally sound energy-efficiency-focused tools. We demonstrate substantial opportunity for energy-focused software optimizations on a mobile CPU core. Submission published under a 24 month embargo labeled 'Closed Access', the embargo will last until 2017-05-01 The student, Matthew Johnson, accepted the attached license on 2015-04-24 at 09:42. The student, Matthew Johnson, submitted this Dissertation for approval on 2015-04-24 at 09:49. This Dissertation was approved for publication on 2015-04-24 at 13:48. DSpace SAF Submission Ingestion Package generated from Vireo submission #8079 on 2015-07-22 at 14:26:28 Made available in DSpace on 2015-07-22T22:46:03Z (GMT). No. of bitstreams: 2 JOHNSON-DISSERTATION-2015.pdf: 6455818 bytes, checksum: 32e23499cf0abf4815bb8949f0016bac (MD5) LICENSE.txt: 4212 bytes, checksum: ae5ab981c861a12240aa954846e7daff (MD5) Previous issue date: 2015-04-24 Embargo set by: Seth Robbins for item 80026 Lift date: 2017-07-22T22:46:21Z Reason: Author requested closed access (OA after 2yrs) in Vireo ETD system Limited Restriction Lifted for Item 80026 on 2017-07-23T09:15:27Z. application/pdf http://hdl.handle.net/2142/78785 en Copyright 2015 Matthew Robert Johnson Power measurement Current measurement Energy efficiency Software optimization Field-Programmable Gate Array (FPGA) Power optimization Fast, accurate power measurement and optimization for microprocessor platforms text text 2015-5 Electrical & Computer Eng Electrical & Computer Engr University of Illinois at Urbana-Champaign Dissertation Ph.D.