SOAR: Minimizing Network Utilization with Bounded In-Network Computing

    Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

    12 Scopus citations

    Abstract

    In-network computing via smart networking devices is a recent trend for modern datacenter networks. State-of-the-art switches with near line rate computing and aggregation capabilities are developed to enable, e.g., acceleration and better utilization for modern applications like big data analytics, and large scale distributed and federated machine learning. We formulate and study the problem of activating a limited number of in-network computing devices within a network, aiming at reducing the overall network utilization for a given workload. Such limitations on the number of in-network computing elements per workload arise, e.g., in incremental upgrades of network infrastructure, and are also due to requiring specialized middleboxes, or FPGAs, that should support heterogeneous workloads, and multiple tenants. We present an optimal and efficient algorithm for placing such devices in tree networks with arbitrary link rates, and further evaluate our proposed solution in various scenarios and for various tasks. Our results show that having merely a small fraction of network devices support in-network aggregation can lead to a significant reduction in network utilization. Furthermore, we show that various intuitive strategies for performing such placements exhibit significantly inferior performance compared to our solution, for varying workloads, tasks, and link rates.

    Original languageEnglish
    Title of host publicationCoNEXT 2021 - Proceedings of the 17th International Conference on emerging Networking EXperiments and Technologies
    PublisherAssociation for Computing Machinery, Inc
    Pages16-29
    Number of pages14
    ISBN (Electronic)9781450390989
    DOIs
    StatePublished - 2 Dec 2021
    Event17th ACM International Conference on emerging Networking EXperiments and Technologies, CoNEXT 2021 - Virtual, Online, Germany
    Duration: 7 Dec 202110 Dec 2021

    Conference

    Conference17th ACM International Conference on emerging Networking EXperiments and Technologies, CoNEXT 2021
    Country/TerritoryGermany
    CityVirtual, Online
    Period7/12/2110/12/21

    ASJC Scopus subject areas

    • Hardware and Architecture
    • Computer Science Applications
    • Computer Networks and Communications

    Fingerprint

    Dive into the research topics of 'SOAR: Minimizing Network Utilization with Bounded In-Network Computing'. Together they form a unique fingerprint.

    Cite this