AdaInf: Data Drift Adaptive Scheduling for Accurate and SLO-guaranteed Multiple-Model Inference Serving at Edge Servers.
Sudipta Saha ShubhaHaiying ShenPublished in: SIGCOMM (2023)
Keyphrases
- data sets
- experimental data
- prior knowledge
- data analysis
- input data
- test data
- computational model
- high quality
- probabilistic model
- data collection
- database
- learning models
- prior information
- data points
- accurate models
- simulation data
- training data
- measured data
- high level
- data distribution
- probability distribution
- data sources
- upper layer
- edge information
- bayesian inference
- data mining
- sensor data
- data mining techniques
- scheduling problem
- image data