T
Trevor Mudge
Researcher at University of Michigan
Publications - 467
Citations - 27174
Trevor Mudge is an academic researcher from University of Michigan. The author has contributed to research in topics: Cache & Cache pollution. The author has an hindex of 74, co-authored 452 publications receiving 25870 citations. Previous affiliations of Trevor Mudge include Stanford University & Intel.
Papers
More filters
Proceedings ArticleDOI
MiBench: A free, commercially representative embedded benchmark suite
Matthew R. Guthaus,Jeff Ringenberg,Daniel J. Ernst,Todd Austin,Trevor Mudge,Richard B. Brown +5 more
TL;DR: A new version of SimpleScalar that has been adapted to the ARM instruction set is used to characterize the performance of the benchmarks using configurations similar to current and next generation embedded processors.
Journal ArticleDOI
Leakage current: Moore's law meets static power
Nam Sung Kim,Todd Austin,D. Baauw,Trevor Mudge,Krisztian Flautner,Jie Hu,Mary Jane Irwin,Mahmut Kandemir,Vijaykrishnan Narayanan +8 more
TL;DR: The other source of power dissipation in microprocessors, dynamic power, arises from the repeated capacitance charge and discharge on the output of the hundreds of millions of gates in today's chips.
Proceedings ArticleDOI
Razor: a low-power pipeline based on circuit-level timing speculation
Daniel J. Ernst,Nam Sung Kim,Shidhartha Das,Sanjay Pant,Rajeev R. Rao,Toan Pham,Conrad H. Ziesler,David Blaauw,Todd Austin,Krisztian Flautner,Trevor Mudge +10 more
TL;DR: A solution by which the circuit can be operated even below the ‘critical’ voltage, so that no margins are required and thus more energy can be saved.
Proceedings ArticleDOI
Neurosurgeon: Collaborative Intelligence Between the Cloud and Mobile Edge
TL;DR: Neurosurgeon, a lightweight scheduler to automatically partition DNN computation between mobile devices and datacenters at the granularity of neural network layers is designed, finding that a fine-grained, layer-level computation partitioning strategy based on the data and computation variations of each layer within a DNN has significant latency and energy advantages over the status quo approach.
Journal ArticleDOI
Drowsy caches: simple techniques for reducing leakage power
TL;DR: It is argued that the use of drowsy caches can simplify the design and control of low-leakage caches, and avoid the need to completely turn off selected cache lines and lose their state.