Self-driving Networks
Building and Deploying Self-Driving Science Networks exploring WAN, Wireless and Beyond
Our Mission: Providing Fast and Reliable Networking for Future Hyperconnected DOE Self-Driving Facilities.
Our group focuses on providing "networks beyond boundaries" for DOE Science. Networks are the essential bloodline to science collaborations across the globe such as in high-energy physics, earth sciences, and genomics. Our mission is to explore machine learning, deep learning, and AI techniques for operational networking and distributed computing problems is efforts to save millions of dollars in continual network upgrades.
Our Research
Network Prediction
NetPredict focuses on predicting Network Traffic and understanding flows to help optimize the network.
Self-driving WAN controller
HECATE focuses on using deep reinforcement learning to help optimize network utilization and performance.
End-to-end automation from 5G Edge to Core
Self-driving 5G network focuses on using Intent and SENSE deployments to ease 5G science networks for many use cases.
5G Security
Exploring zero-touch for 5G security
TCP Research
Studying how TCP congestion control impacts the Science networks. Including WAN and 5G layers.
5G Security
Exploring zero-touch for 5G security
Questions: Contact <mkiran@es.net> and <awiedlea@es.net> for more inputs.
Our Industry and Research Partners
AWS
Verizon
LBNL
DOE / Andrew (other partners)