Saba: Rethinking Datacenter Network Allocation from Application’s Perspective

Siavash Katebzadeh, Paolo Costa, Boris Grot

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract / Description of output

Today’s datacenter workloads increasingly comprise distributed data-intensive applications, including data analytics, graph processing, and machine-learning training. These applications are bandwidth-hungry and often congest the datacenter network, resulting in poor network performance,which hurts application completion time. Efforts made to address this problem generally aim to achieve max-min fairness at the flow or application level. We observe that splitting the bandwidth equally among workloads is sub-optimal for aggregate application-level performance because various workloads exhibit different sensitivity to network bandwidth:for some workloads, even a small reduction in the available bandwidth yields a significant increase in completion time; for others, the completion time is largely insensitive to the available bandwidth.Building on this insight, we propose Saba, an application aware bandwidth allocation framework that distributes network bandwidth based on application-level sensitivity. Saba combines ahead-of-time application profiling to determine bandwidth sensitivity with runtime bandwidth allocation using lightweight software support with no modifications to network hardware or protocols. Experiments with a 32-server hardware testbed show that Saba improves average completion time by 1.88× (and by 1.27× in a simulated 1,944-server cluster).
Original languageEnglish
Title of host publicationEuroSys '23: Eighteenth European Conference on Computer Systems
Number of pages16
ISBN (Electronic) 9781450394871
Publication statusPublished - 8 May 2023
EventEuroSys '23: Eighteenth European Conference on Computer Systems - Rome, Italy
Duration: 8 May 202312 May 2023
Conference number: 18


ConferenceEuroSys '23: Eighteenth European Conference on Computer Systems
Internet address


Dive into the research topics of 'Saba: Rethinking Datacenter Network Allocation from Application’s Perspective'. Together they form a unique fingerprint.

Cite this