FogSys: A System and Framework for Fault Tolerant Real-Time Fog Computing
Abstract
Tactical clouds play a crucial role in providing reliable backend support for devices and autonomous vehicles deployed in the field. The current cloud computing model, which focuses on running applications and processing data in data centers, is ineffective for geographically dispersed applications that demand intelligence at the edge and must meet strict real-time constraints. Relaying data to the remote cloud leads to high propagation delay, which cannot be overcome without moving computation closer to the edge devices. This is the premise of fog computing, which offers an alternative by bringing computational resources physically closer to the data and reducing its dependence on a backend cloud. Distributed fog resources near the edge can process real-time data from devices with lower latency. However, fog computing introduces its own challenges in effectively managing the large network of heterogeneous fog resources that interface with both devices and the backend cloud. USC/ISI will develop a real-time fog framework capable of maintaining fault tolerance in the face of timing errors and physical faults while providing a scalable set of resources. USC/ISI will then extend this work by leveragingredundancy and binary diversification to provide a cyber-resilient fog platform.
Document Details
- Document Type
- DoD Grant Award
- Publication Date
- Mar 11, 2020
- Source ID
- N000142012143
Entities
People
- John Paul Walters
Organizations
- Office of Naval Research
- United States Navy
- University of Southern California