Characterizing the Performance of Accelerated Jetson Edge Devices for Training Deep Learning Models
Abstract
Supplemental Material
- Download
- 157.42 MB
References
Index Terms
- Characterizing the Performance of Accelerated Jetson Edge Devices for Training Deep Learning Models
Recommendations
Characterizing the Performance of Accelerated Jetson Edge Devices for Training Deep Learning Models
SIGMETRICS '23Deep Neural Network (DNN) models are becoming ubiquitous in a variety of contemporary domains such as Autonomous Vehicles, Smart cities and Healthcare. They help drones to navigate, identify suspicious activities from safety cameras, and perform ...
Characterizing the Performance of Accelerated Jetson Edge Devices for Training Deep Learning Models
POMACSDeep Neural Networks (DNNs) have had a significant impact on domains like autonomous vehicles and smart cities through low-latency inferencing on edge computing devices close to the data source. However, DNN training on the edge is poorly explored. ...
PowerTrain: Fast, generalizable time and power prediction models to optimize DNN training on accelerated edges
AbstractAccelerated edge devices, like Nvidia’s Jetson with 1000+ CUDA cores, are increasingly used for DNN training and federated learning, rather than just for inferencing workloads. A unique feature of these compact devices is their fine-grained ...
Highlights- ML-based prediction models for estimating runtime and power of edge DNN training.
- Demonstrate the generalizability of prediction models.
- Optimization of training time using the Pareto front.
Comments
Information & Contributors
Information
Published In
- General Chair:
- Evgenia Smirni,
- Program Chairs:
- Konstantin Avrachenkov,
- Phillipa Gill,
- Bhuvan Urgaonkar
Sponsors
Publisher
Association for Computing Machinery
New York, NY, United States
Publication History
Check for updates
Author Tags
Qualifiers
- Abstract
Data Availability
Funding Sources
- Ministry of Education, India
- Department of Science and Technology, India
Conference
Acceptance Rates
Contributors
Other Metrics
Bibliometrics & Citations
Bibliometrics
Article Metrics
- View Citations1Total Citations
- 158Total Downloads
- Downloads (Last 12 months)98
- Downloads (Last 6 weeks)10
Other Metrics
Citations
View Options
Login options
Check if you have access through your login credentials or your institution to get full access on this article.
Sign in