Can Machine Learning Benefit Bandwidth Estimation at Ultra-high Speeds?

Yin, Qianwen; Kaur, Jasleen

doi:10.1007/978-3-319-30505-9_30

Cited by 13 publications

(9 citation statements)

References 16 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Another area of research which has considered the shared nature of networks is the research of estimating available bandwidth, where the notion of cross traffic (e.g., a mix of foreground and background traffic) has been investigated [61].…”

Section: Effects Of Sharing Bandwidthmentioning

confidence: 99%

“…Estimations could range from probing available bandwidth, to inferring the workload and the type of the traffic on the network. There are a number of studies in this area, mainly investigating the possibility of applying ML techniques for bandwidth estimation in high-speed networks [61,67]. In a different vein, machine-learning techniques have been used for designing or optimizing network protocols.…”

Section: Machine-learning For Networkingmentioning

confidence: 99%

See 1 more Smart Citation

Machine-Learned Recognition of Network Traffic for Optimization through Protocol Selection

Anvari

2021

Computers

View full text Add to dashboard Cite

We introduce optimization through protocol selection (OPS) as a technique to improve bulk-data transfer on shared wide-area networks (WANs). Instead of just fine-tuning the parameters of a network protocol, our empirical results show that the selection of the protocol itself can result in up to four times higher throughput in some key cases. However, OPS for the foreground traffic (e.g., TCP CUBIC, TCP BBR, UDT) depends on knowledge about the network protocols used by the background traffic (i.e., other users). Therefore, we build and empirically evaluate several machine-learned (ML) classifiers, trained on local round-trip time (RTT) time-series data gathered using active probing, to recognize the mix of network protocols in the background with an accuracy of up to 0.96.

show abstract

Section: Effects Of Sharing Bandwidthmentioning

confidence: 99%

Section: Machine-learning For Networkingmentioning

confidence: 99%

Machine-Learned Recognition of Network Traffic for Optimization through Protocol Selection

Anvari

2021

Computers

View full text Add to dashboard Cite

show abstract

“…Moreover, the authors in [19]- [21] use packet chirps [7], i.e., the probes of several packets sent at increasing rates. They achieve the rate increase by a geometric reduction of the input gap [19], by concatenating several packet trains with increasing rates to a multi-rate probe [20], and by linearly increasing the packet size [21]. The packet chirps, with a single probe, lead to the detection of the turning point, which is the actual available bandwidth.…”

Section: A State-of-the-art Estimation Techniquesmentioning

confidence: 99%

“…Furthermore, the authors in [19] study the packet bursts which are known as back-toback packet probes and conclude that the packet bursts are not enough to estimate the available bandwidth. Also, the authors in [20] consider constant-rate packet trains in an iterative manner to attain the available bandwidth. Here, machine learning solves a classification problem to estimate whether the rate of a packet train exceeds the available bandwidth.…”

Section: A State-of-the-art Estimation Techniquesmentioning

confidence: 99%

See 1 more Smart Citation

Measurement-Based Online Available Bandwidth Estimation Employing Reinforcement Learning

Khangura

Akın

2019

2019 31st International Teletraffic Congress (ITC 31)

View full text Add to dashboard Cite

An accurate and fast estimation of the available bandwidth in a network with varying cross-traffic is a challenging task. The accepted probing tools, based on the fluid-flow model of a bottleneck link with first-in, first-out multiplexing, estimate the available bandwidth by measuring packet dispersions. The estimation becomes more difficult if packet dispersions deviate from the assumptions of the fluid-flow model in the presence of non-fluid bursty cross-traffic, multiple bottleneck links, and inaccurate time-stamping. This motivates us to explore the use of machine learning tools for available bandwidth estimation. Hence, we consider reinforcement learning and implement the single-state multi-armed bandit technique, which follows the εgreedy algorithm to find the available bandwidth. Our measurements and tests reveal that our proposed method identifies the available bandwidth with high precision. Furthermore, our method converges to the available bandwidth under a variety of notoriously difficult conditions, such as heavy traffic burstiness, different cross-traffic intensities, multiple bottleneck links, and in networks where the tight link and the bottleneck link are not same. Compared to the piece-wise linear network a modelbased direct probing technique that employs a Kalman filter, our method shows more accurate estimates and faster convergence in certain network scenarios and does not require measurement noise statistics.

show abstract

Machine-Learned Classifiers for Protocol Selection on a Shared Network

Anvari

Huard

2019

Machine Learning for Networking

View full text Add to dashboard Cite

Can Machine Learning Benefit Bandwidth Estimation at Ultra-high Speeds?

Cited by 13 publications

References 16 publications

Machine-Learned Recognition of Network Traffic for Optimization through Protocol Selection

Machine-Learned Recognition of Network Traffic for Optimization through Protocol Selection

Measurement-Based Online Available Bandwidth Estimation Employing Reinforcement Learning

Machine-Learned Classifiers for Protocol Selection on a Shared Network

Contact Info

Product

Resources

About