Predicting Critical Warps in Near-Threshold GPGPU Applications Using a Dynamic Choke Point Analysis
General purpose graphics processing units (GP-GPU), owing to their enormous thread-level parallelism, can significantly improve the power consumption at the near-threshold (NTC) operating region, while offering close to a super-threshold performance. However, process variation (PV) can drastically r...
Main Author: | Sanyal, Sourav |
---|---|
Format: | Others |
Published: |
DigitalCommons@USU
2019
|
Subjects: | |
Online Access: | https://digitalcommons.usu.edu/etd/7545 https://digitalcommons.usu.edu/cgi/viewcontent.cgi?article=8676&context=etd |
Similar Items
-
Enhancing GPGPU Performance through Warp Scheduling, Divergence Taming and Runtime Parallelizing Transformations
by: Anantpur, Jayvant P
Published: (2018) -
Exploiting Application Behaviors for Resilient Static Random Access Memory Arrays in the Near-Threshold Computing Regime
by: Mugisha, Dieudonne Manzi
Published: (2015) -
Tackling Choke Point Induced Performance Bottlenecks in a Near-Threshold GPGPU
by: Shabanian, Tahmoures
Published: (2018) -
Near-Threshold L1 Data Cache for Yield Management Under Process Variations
by: Joonho Kong, et al.
Published: (2020-01-01) -
Split Latency Allocator: Process Variation-Aware Register Access Latency Boost in a Near-Threshold Graphics Processing Unit
by: Pal, Asmita
Published: (2018)