Human-driven application management at the Edge

Brogi, Antonio and Forti, Stefano and Ibrahim, Ahmad (2017) Human-driven application management at the Edge. Technical Report del Dipartimento di Informatica, TR . University of Pisa, Pisa, IT. (Submitted)

Preview

PDF - Submitted Version Available under License Creative Commons Attribution Non-commercial No Derivatives. Download (1545Kb) | Preview

Abstract

The design and management of Edge systems will proactively involve human intelligence at the Edge, according to a human-driven approach that increases productivity and improves usability. Due to its ubiquity and heterogeneity, the Edge will give to application administrators a more decisional role in application deployment and resource management. Final decisions on where to distribute application components should be informedly taken by them during the entire application lifecycle, accounting for compliance to QoS requirements. As a first step, this requires devising new tools that suitably abstract heterogeneity of edge systems, permit simulating different runtime scenarios and ease human-driven management of such systems by providing meaningful evaluation metrics. In this article, we discuss how human decision-making can be supported to solve QoS-aware management related challenges for Edge computing.

Repository staff only actions