MECC'18: Proceedings of the 3rd Workshop on Middleware for Edge Clouds & Cloudlets

Rolando Martins (Editor/Coordinator), Hervé Miguel Cordeiro Paulino (Editor/Coordinator), Luís Veiga (Editor/Coordinator)

Research output: Book/ReportBook

Abstract

The Middleware for Edge Clouds & Cloudlets (MECC) workshop aims to bring together researchers and practitioners addressing how to make the full stack of modern cloud computing platforms escape outside of mega-datacenters and operate seamlessly across all geographies and all devices. There is a growing trend of interactive and more resource-intensive (e.g., compute, storage, need for big data) applications on mobile devices today, and currently many such applications are powered resorting only to resources provided by mega-datacenters on infrastructural clouds. This poses increasingly difficult challenges regarding resource costs, application availability, latency, energy and environmental concerns. In fact, it is challenging to provide such applications using just cloud resources when there is limited or intermittent connectivity. Harvesting the resources present at the edge of the network, on nearby mobile devices and/or cloudlets, is a viable solution to this problem. This approach can also improve latency, reduce costs, and help to contribute to diminish the current incentives for larger and larger datacenters. Today, there is also increasing demand for middleware that offers higher level abstractions without hampering expressiveness and performance. However, many distributed systems today are still designed with only datacenters in mind, and their assumptions, such as that nodes use fast wired interconnects, no longer hold in edge environments. In particular, edge clouds, such as those made up of only mobile devices at the edge, use unreliable wireless links. These unreliable links directly translate into unavailability and churn. Simultaneously, since mobile devices have limited energy resources, heavyweight distributed algorithms, such as coordination using a leader-based consensus protocol, are impractical. Conversely, as an effort to offload heavier computation from mobile devices, cloudlets were originally envisioned as server-class hardware deployed in neighborhoods, office buildings or more generally, in close physical proximity to environments with high user density, such as at large public events. It is now adopting more lightweight approaches where the offloading no longer resorts solely to virtual machines, as initially proposed, where cloudlets can operate as execution environments for lightweight containers and can also provide connectivity support to crowd-sourced mobile devices, i.e., edge clouds. With this new trend in sight, the community still needs to define the services that should be offered at each tier of the edge cloud stack. While cloudlets can provide well-defined APIs to support multiple computation offloading methods, new modular and reconfigurable architectures still have to be proposed to support a variety of deployment scenarios, such as edge clouds without cloudlet support, and scenarios with very limited access to infrastructural clouds. In this second edition, we look forward to the presentations and discussions that may shed light on these challenges. The organizers would like to thank all the authors, Middleware organizers, chairs, workshop and publicity chairs.
Original languageEnglish
Place of PublicationNew York
PublisherAssociation for Computing Machinery, Inc
Number of pages19
ISBN (Print)978-1-4503-6117-0
Publication statusPublished - Dec 2018
Event3rd Workshop on Middleware for Edge Clouds & Cloudlets (MECC'18) - Rennes, France
Duration: 10 Dec 201811 Dec 2018

Fingerprint

Dive into the research topics of 'MECC'18: Proceedings of the 3rd Workshop on Middleware for Edge Clouds & Cloudlets'. Together they form a unique fingerprint.

Cite this