G
Gu-Yeon Wei
Researcher at Harvard University
Publications - 16
Citations - 438
Gu-Yeon Wei is an academic researcher from Harvard University. The author has contributed to research in topics: Hardware acceleration & Multi-core processor. The author has an hindex of 8, co-authored 16 publications receiving 296 citations.
Papers
More filters
Proceedings ArticleDOI
DeepRecSys: a system for optimizing end-to-end at-scale neural recommendation inference
Udit Gupta,Samuel Hsia,Vikram Saraph,Xiaodong Wang,Brandon Reagen,Gu-Yeon Wei,Hsien-Hsin S. Lee,David Brooks,Carole-Jean Wu +8 more
TL;DR: DeepRecSched as mentioned in this paper is a recommendation inference scheduler that maximizes latency-bounded throughput by taking into account characteristics of inference query size and arrival patterns, model architectures, and underlying hardware systems.
Journal ArticleDOI
ReVIVaL: A Variation-Tolerant Architecture Using Voltage Interpolation and Variable Latency
TL;DR: ReVIVaL is presented, which combines two fine-grained post-fabrication tuning techniques---voltage interpolation(VI) and variable latency(VL) and shows that the frequency variation between chips, between cores on one chip, and between functional units within cores can be reduced to a very small range.
Journal ArticleDOI
Revival: A Variation-Tolerant Architecture Using Voltage Interpolation and Variable Latency
TL;DR: The revival technique combines the post-fabrication tuning techniques voltage interpolation and variable latency (VL) to reduce such frequency variations in process variations.
Posted Content
DeepRecSys: A System for Optimizing End-To-End At-scale Neural Recommendation Inference.
Udit Gupta,Samuel Hsia,Vikram Saraph,Xiaodong Wang,Brandon Reagen,Gu-Yeon Wei,Hsien-Hsin S. Lee,David Brooks,Carole-Jean Wu +8 more
TL;DR: DeepRecSched is proposed, a recommendation inference scheduler that maximizes latency-bounded throughput by taking into account characteristics of inference query size and arrival patterns, model architectures, and underlying hardware systems.
Journal ArticleDOI
SMAUG: End-to-End Full-Stack Simulation Infrastructure for Deep Learning Workloads
TL;DR: SMAUG as discussed by the authors is a DNN framework that is purpose-built for simulation of end-to-end deep learning applications, which can be used to evaluate the performance and energy efficiency of deep learning workloads.