WF/sup 2/Q-M : a worst-case fair weighted fair queueing with maximum rate control

Lee, Jeng Farn; Chen, Meng Chang; Sun, Yeali S.

doi:10.1109/glocom.2002.1188463

Cited by 1 publication

(1 citation statement)

References 21 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…To address this issue, over the last decade, there has been a slew of proposals in the literature on how to improve congestion control in datacenters and efficiently share network bandwidth among competing flows and applications. Some of these proposals focus on providing per-flow fairness [18,20,34] with various network-level objectives, such as minimizing per-packet latency [25] or flow completion time [5,6,44] using network-level properties, such as flow size or deadlines [5,6,13,25,44,55]. Other works aim to provide isolation among tenants or applications when running in shared datacenters [11,12,23,27,32,48,57].…”

Section: Introductionmentioning

confidence: 99%

Saba: Rethinking Datacenter Network Allocation from Application's Perspective

Katebzadeh

Costa

Grot

2023

Proceedings of the Eighteenth European Conference on Computer Systems

View full text Add to dashboard Cite

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 applicationaware 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 32server hardware testbed show that Saba improves average completion time by 1.88× (and by 1.27× in a simulated 1,944server cluster).

show abstract