Infrastructure as Software in Micro Clouds at the Edge

Abstract

Edge computing offers cloud services closer to data sources and end-users, making the foundation for novel applications. The infrastructure deployment is taking off, bringing new challenges: how to use geo-distribution properly, or harness the advantages of having resources at a specific location? New real-time applications require multi-tier infrastructure, preferably doing data preprocessing locally, but using the cloud for heavy workloads. We present a model, able to organize geo-distributed nodes into micro clouds dynamically, allowing resource reorganization to best serve population needs. Such elasticity is achieved by relying on cloud organization principles, adapted for a different environment. The desired state is specified descriptively, and the system handles the rest. As such, infrastructure is abstracted to the software level, thus enabling “infrastructure as software” at the edge. We argue about blending the proposed model into existing tools, allowing cloud providers to offer future micro clouds as a service.

A Hierarchical Namespace Approach for Multi-Tenancy in Distributed Clouds

“…A consequence of this isolation is the proper usage of Linuxbased security access control, known as secure computing mode (Seccomp) [34], [35]. These techniques pave the way for extending namespaces and Seccomp principles to other areas, most notably in micro and distributed clouds [36].…”

“…The main benefits of our model are increased development freedom and agility. While specialized models may outperform distributed clouds in terms of speed, developers are willing to trade off a small degree of speed for increased development freedom and agility [36].…”

“…This architecture allows connecting different data centers, edge nodes, and other computing resources to provide a vast portfolio of services for different use cases such as the Internet of Things (IoT), autonomous vehicles, gaming, or augmented reality. In the µC model, resources are deployed on a variable scale but keep the main advantages of distributed clouds [36]. Pham et al define three enabling technologies for distributed clouds: decentralized software-defined networks, cloud-native applications, and Artificial Intelligence (AI)-powered intent-based networking (IBN) [54].…”

“…Resources organized dynamically into disposable µCs, operating near users first serving their requests locally, present promising ideas because they bring cloud-like infrastructure to users at the network edge, making collaboration between the cloud and the edge far easier [29], [36].…”

“…Our previous work [29], [36] focuses on creating and formally describing the system that would achieve such a feat. The original µC model lacks proper support for multitenancy.…”