Modeling BBR's Interactions with Loss-Based Congestion Control

Ware, Ranysha; Mukerjee, Matthew K.; Seshan, Srinivasan; Sherry, Justine

doi:10.1145/3355369.3355604

Cited by 74 publications

(43 citation statements)

References 3 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…TCP Congestion Control. Many of the congestion control studies have especially looked at CUBIC [17] and BBR [10] and found that BBR dominates in under-buffered scenarios causing packet loss and making CUBIC back off, while it is disadvantaged in over-buffered scenarios [18,23,26,28]. Here, CUBIC, as a loss-based algorithm, fills the buffer and increases the queuing delay which makes BBR back off.…”

Section: Background and Related Workmentioning

confidence: 99%

Video Conferencing and Flow-Rate Fairness: A First Look at Zoom and the Impact of Flow-Queuing AQM

Sander¹,

Kunze²,

Wehrle³

et al. 2021

Passive and Active Measurement

View full text Add to dashboard Cite

Congestion control is essential for the stability of the Internet and the corresponding algorithms are commonly evaluated for interoperability based on flow-rate fairness. In contrast, video conferencing software such as Zoom uses custom congestion control algorithms whose fairness behavior is mostly unknown. Aggravatingly, video conferencing has recently seen a drastic increase in use -partly caused by the COVID-19 pandemic -and could hence negatively affect how available Internet resources are shared. In this paper, we thus investigate the flow-rate fairness of video conferencing congestion control at the example of Zoom and influences of deploying AQM. We find that Zoom is slow to react to bandwidth changes and uses two to three times the bandwidth of TCP in low-bandwidth scenarios. Moreover, also when competing with delay aware congestion control such as BBR, we see high queuing delays. AQM reduces these queuing delays and can equalize the bandwidth use when used with flow-queuing. However, it then introduces high packet loss for Zoom, leaving the question how delay and loss affect Zoom's QoE. We hence show a preliminary user study in the appendix which indicates that the QoE is at least not improved and should be studied further.

show abstract

Section: Background and Related Workmentioning

confidence: 99%

Video Conferencing and Flow-Rate Fairness: A First Look at Zoom and the Impact of Flow-Queuing AQM

Sander¹,

Kunze²,

Wehrle³

et al. 2021

Passive and Active Measurement

View full text Add to dashboard Cite

show abstract

“…BBR is a delay/model-based CCA which converges on a fair share of bottleneck bandwidth by reducing its rate if the round-trip time increases, while periodically attempting to increase send rate to account for path/load changes. However, BBR traffic can consume 40 % of link capacity when multiplexed with loss-based CCAs, regardless of the number of competing flows [13]. When ensuring fair transit to all flows, this is hardly a desirable outcome; in fact, it's one which may frustrate clients or violate SLAs.…”

Section: I I T C P C O N G E S T I O N C O N T R O L C L a S S I F I C At I O Nmentioning

confidence: 99%

“…As a concrete use-case, we look at fine dynamics of TCP congestion control algorithms. Understanding and classifying them is important for network providers as inadequate choices have severe effects on transfer rates, especially in networks with high bandwidth-delay product [12] and in networks where multiple congestion control algorithms are used [13]. By using accurate congestion control diagnostics, operators will be able to infer sender problems (e.g., backlogged or applicationlimited senders), network inefficiencies (e.g., increased path latency and congestion), as well as receiver issues (e.g., delayed acknowledgements, small receiver windows) and fairness issues between delay-based and loss-based algorithms [13].…”

Section: Introductionmentioning

confidence: 99%

Seiðr: Dataplane Assisted Flow Classification Using ML

Simpson

Cziva

Pezaros

2020

GLOBECOM 2020 - 2020 IEEE Global Communications Conference

View full text Add to dashboard Cite

Real-time, high-speed flow classification is fundamental for network operation tasks, including reactive and proactive traffic engineering, anomaly detection and security enhancement. Existing flow classification solutions, however, do not allow operators to classify traffic based on fine-grained, temporal dynamics due to imprecise timing, often rely on sampled data, or only work with low traffic volumes and rates. In this paper, we present Seiðr, a classification solution that: (i) uses precision timing, (ii) has the ability to examine every packet on the network, (iii) classifies very high traffic volumes with high precision. To achieve this, Seiðr exploits the data aggregation and timestamping functionality of programmable dataplanes. As a concrete example, we present how Seiðr can be used together with Machine Learning algorithms (such as CNN, k-NN) to provide accurate, real-time and high-speed TCP congestion control classification, separating TCP BBR from its predecessors with over 88-96 % accuracy and F1-score of 0.864-0.965, while only using 15.5 MiB of memory in the dataplane.

show abstract

“…It is highly unlikely that many new services will set out to deliberately deploy algorithms that penalize the performance of their own flows -but given the difficulty of achieving perfect fairness ( §2.1.2) it is likely that this outcome may happen in some scenarios. 2 A very deployable, friendly algorithm would err on the side of harming their own connections where a more aggressive algorithm would err on the side of harming those 2 Consider Google's BBR, which, when one BBR flow competes with one Reno flow will only consume 40% of link capacity -less than its fair share [2,24]. Furthermore, 'background' CCA algorithms, such as TCP-Nice [23] and LEDBAT [20] deliberately only consume leftover bandwidth that is underutilized by other connections.…”

Section: 13mentioning

confidence: 99%

Beyond Jain's Fairness Index

Ware

Mukerjee²,

Seshan

et al. 2019

Proceedings of the 18th ACM Workshop on Hot Topics in Networks

Self Cite

View full text Add to dashboard Cite

The Internet community faces an explosion in new congestion control algorithms such as Copa, Sprout, PCC, and BBR. In this paper, we discuss considerations for deploying new algorithms on the Internet. While past efforts have focused on achieving 'fairness' or 'friendliness' between new algorithms and deployed algorithms, we instead advocate for an approach centered on quantifying and limiting harm caused by the new algorithm on the status quo. We argue that a harm-based approach is more practical, more future proof, and handles a wider range of quality metrics than traditional notions of fairness and friendliness.

show abstract

Modeling BBR's Interactions with Loss-Based Congestion Control

Cited by 74 publications

References 3 publications

Video Conferencing and Flow-Rate Fairness: A First Look at Zoom and the Impact of Flow-Queuing AQM

Video Conferencing and Flow-Rate Fairness: A First Look at Zoom and the Impact of Flow-Queuing AQM

Seiðr: Dataplane Assisted Flow Classification Using ML

Beyond Jain's Fairness Index

Contact Info

Product

Resources

About