We introduce ParaDrop, a specific edge computing platform that provides computing and storage resources at the "extreme" edge of the network allowing third-party developers to flexibly create new types of services. This extreme edge of the network is the WiFi Access Point (AP) or the wireless gateway through which all end-device traffic (personal devices, sensors, etc.) passes through. ParaDrop's focus on WiFi APs also stems from the fact that the WiFi AP has unique contextual knowledge of its end-devices (e.g., proximity, channel characteristics) that are lost as we get deeper into the network. While different variations and implementations of edge computing platforms have been created over the last decade, ParaDrop focuses on specific design issues around how to structure an architecture, a programming interface, and orchestration framework through which such edge computing services can be dynamically created, installed, and revoked. ParaDrop consists of the following three main components: a flexible hosting substrate in the WiFi APs that supports multi-tenancy, a cloud-based backend through which such computations are orchestrated across many ParaDrop APs, and an API through which third-party developers can deploy and manage their computing functions across such different ParaDrop APs. We have implemented and deployed the entire ParaDrop framework and, in this paper, describe its overall architecture and our initial experiences using it as an edge computing platform.


    Access

    Check access

    Check availability in my library

    Order at Subito €


    Export, share and cite



    Title :

    ParaDrop: Enabling Lightweight Multi-tenancy at the Network’s Extreme Edge


    Contributors:
    Liu, Peng (author) / Willis, Dale (author) / Banerjee, Suman (author)


    Publication date :

    2016-10-01


    Size :

    1490296 byte




    Type of media :

    Conference paper


    Type of material :

    Electronic Resource


    Language :

    English



    Software Network's missing link moves closer

    British Library Online Contents | 2005



    Enabling Edge-based Federated Learning through MQTT and OMA Lightweight-M2M

    Genovese, Giacomo / Singh, Gurtaj / Campolo, Claudia et al. | IEEE | 2022



    A new RAN slicing strategy for multi-tenancy support in a WLAN scenario

    Koutlia, Katerina / Umbert, Anna / Riggio, Roberto et al. | IEEE | 2018