I-Corps: AirBox: Bringing the Cloud to the Edge
Georgia Tech Research Corporation, Atlanta GA
Investigators
Abstract
This project addresses Mobile Edge Computing (MEC), a distributed computing architecture that combines cloud capabilities and cellular networks to support real-time applications that demand low latency response. Interactive, data-rich end-user services supported by remote cloud back-end systems face formidable challenges. To provide a responsive user experience, they need low, predictable access latency while processing ever-increasing amounts of data. Concretely, applications such as augmented reality, remote tactile interaction, self-driving cars, etc., require close to 1 millisecond end to end latencies while requiring large data transfers from mobile devices. The aggregate data demand of such interactions is projected to reach 24.2 exabytes per month. Those needs are hard to meet when information must move to and from distant servers across an increasingly congested Internet backbone. Distant services also have limited or no access to local context data such as positioning and signal strength. This project is aimed at exploring the opportunities that Mobile Edge Computing (MEC) offers to address these challenges. MEC is based on distributed infrastructure elements, from wireless access points to high density/low-power micro-datacenters, situated near end users. MEC allows enterprise and consumer services to be delivered from the wireless edge, enabling fresh business opportunities across multiple sectors, targeting conventional services, content delivery engines, on premise enterprise applications, and more. The concrete goals of this project are (1) to investigate the economic models and practical challenges in leveraging the edge infrastructure for deployment and execution of various edge functions, and (2) to understand the technical requirements for provisioning and executing services in this edge tier, with adequate performance and security guarantees. Given the scope and complexity of the mobile ecosystem, this I-Corps team is aware of the fact that a great deal of customer discovery and industry research will be necessary to identify a niche where a startup company will succeed. Additionally, it is important for the team to explore the market risks associated with deployment, ownership and viable economic models for edge infrastructure. The team plans to engage a broad range of customers and stakeholders to uncover one or more potential niches.
View original record on NSF Award Search →