Scaling Bandwidth Estimation to High Speed Networks

Yin, Qianwen; Kaur, Jasleen; Smith, F. Donelson

doi:10.1007/978-3-319-04918-2_25

Cited by 3 publications

(2 citation statements)

References 4 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Note that sampled inter-arrival times may oscillate around the real value due to the effect of cross traffic and interrupt coalescence [9] in the receiver side. In order to avoid the influence of such effects, the inter-arrival times used for CIR and P IR calculation are pre-processed using a smoothing algorithm (similar as the one in [21]) based on exponential weights. Even though this technique minimizes the effects of the outliers, the measurement may still be affected by values that significantly diverge from the average value.…”

Section: Cumulativementioning

confidence: 99%

Estimation of the parameters of token-buckets in multi-hop environments

et al. 2018

View full text Add to dashboard Cite

show abstract

Section: Cumulativementioning

confidence: 99%

Estimation of the parameters of token-buckets in multi-hop environments

et al. 2018

View full text Add to dashboard Cite

show abstract

“…This is because small inter-packet gaps are needed for probing higher bandwidth-such fine-scale gaps are more susceptible to being disturbed by small-scale buffering related noise at shared resources and at end hosts [19].…”

Section: Introductionmentioning

confidence: 99%

Can Bandwidth Estimation Tackle Noise at Ultra-high Speeds?

Yin

Kaur

Smith

2014

2014 IEEE 22nd International Conference on Network Protocols

Self Cite

View full text Add to dashboard Cite

Abstract-While existing bandwidth estimation tools have been shown to perform well on 100Mbps networks, they fail to do so at gigabit and higher network speeds. This is because finer inter-packet gaps are needed to probe for higher ratesfine gaps are more susceptible to be disturbed by small-scale buffering-related noise. In this paper, we evaluate existing noisereduction techniques for tackling the issue, and show that they are ineffective on 10Gbps links. We propose a novel smoothing strategy, Buffering-aware Spike Smoothing (BASS), which can be applied effectively to both single-rate and multi-rate probing frameworks and help significantly in scaling bandwidth estimation to ultra-high speed networks. Besides, we provide first evidence that accurate bandwidth estimation using our strategy can help improve the performance of congestion-control protocols on real 10Gbps networks.

show abstract