Improving TCP performance with bufferless token bucket policing: A TCP friendly policer

Haalen, R. van; Malhotra, R.

doi:10.1109/lanman.2007.4295978

Cited by 12 publications

(5 citation statements)

References 4 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…This is an important result because the static nature of the classic mechanisms is contradictory with the bursty, dynamic nature of video traffic and this results in very high percentages of unfairly marked traffic for conforming users. Additionally, as shown in [31], TCP performance can degrade significantly when policed with classic static mechanisms and the solution of using a large token bucket size creates other problems related with packet loss and latency; the solution of using a dynamic bucket size, in [31], was shown to be TCP friendly. Similarly, our mechanisms, which are dynamically tailored to each transmitted video, can be equally beneficial to TCP.…”

Section: Discussionmentioning

confidence: 99%

Video Activity-Based Traffic Policing: A New Paradigm

Maratsolas

Koutsakis

Lazaris

2014

IEEE Trans. Multimedia

View full text Add to dashboard Cite

The constant development of new multimedia applications, which are "greedy" in terms of bandwidth and Quality of Service (QoS) requirements, calls for new approaches to the traffic policing problem. This paper proposes, analyzes and presents an extensive performance evaluation of such a new approach, namely the activity-based video traffic policing. Using as a motivation recent work, which has shown that the classic traffic policing mechanisms provide unnecessarily strict policing for conforming but bursty video users, we propose five simple and efficient new mechanisms, which take into consideration and exploit video activity and the Group-of Pictures (GoP) pattern of the video traces. Contrary to the classic approach, our mechanisms do not use a token generator based on a fixed rate, but vary the token generation rate according to specific video activity-based algorithms. The results show significant improvement for conforming users, and reveal that dynamic traffic policing can provide much higher efficiency than the widely used static mechanisms. One of the new mechanisms, the Frame Size Aware Token Bucket, is shown to clearly outperform all other policing approaches for conforming users, and to provide comparable policing results with the classic mechanisms for non-conforming video users.

show abstract

Section: Discussionmentioning

confidence: 99%

Video Activity-Based Traffic Policing: A New Paradigm

Maratsolas

Koutsakis

Lazaris

2014

IEEE Trans. Multimedia

View full text Add to dashboard Cite

show abstract

“…Others explored policing for differentiated services [54], fair bandwidth allocation [36], or throughput guarantees [21,64]. One study explored the relationship between TCP performance and token bucket policers in a lab setting and proposed a TCP-friendly version achieving per-flow goodputs close to the policed rate regardless of the policer configuration [60]. Finally, a concurrently published study investigated the impact of traffic policing applied by T-Mobile to content delivery.…”

Section: Related Workmentioning

confidence: 99%

An Internet-Wide Analysis of Traffic Policing

Flach

Papageorge

Terzis

et al. 2016

Proceedings of the 2016 ACM SIGCOMM Conference

View full text Add to dashboard Cite

Large flows like video streams consume significant bandwidth. Some ISPs actively manage these high volume flows with techniques like policing, which enforces a flow rate by dropping excess traffic. While the existence of policing is well known, our contribution is an Internet-wide study quantifying its prevalence and impact on transportlevel and video-quality metrics. We developed a heuristic to identify policing from server-side traces and built a pipeline to process traces at scale collected from hundreds of Google servers worldwide. Using a dataset of 270 billion packets served to 28,400 client ASes, we find that, depending on region, up to 7% of connections are identified to be policed. Loss rates are on average 6× higher when a trace is policed, and it impacts video playback quality. We show that alternatives to policing, like pacing and shaping, can achieve traffic management goals while avoiding the deleterious effects of policing.

show abstract

“…These efforts either adapt the traffic conditioning algorithm to the TCP mechanism [6] or modify the TCP congestion control algorithm according to the characteristics of traffic conditioners [7]. The traffic conditioning algorithm, e.g., the token bucket algorithm, may introduce bursty packet drops when the token bucket has loaned all its tokens [6].…”

Section: Introductionmentioning

confidence: 99%

“…The bursty packet losses beat the TCP congestion window size down to zero. To reduce the bursty packet loss, several solutions have been proposed to adapt the traffic conditioners such as increasing the bucket size [6], adapting the per-flow buffer size according to the traffic burstiness [8], and applying an early dropping policy that drops a packet when the occupancy of the buffer exceeds a predefined threshold but before it is close to overflowing [9]. However, these proposals are not applicable to mobile networks because the parameters of the traffic conditioning algorithm are static and pre-determined according to user service contracts.…”

Section: Introductionmentioning

confidence: 99%

TCP-Mobile Edge: Accelerating delivery in mobile networks

Ansari

2012

2012 IEEE International Conference on Communications (ICC)

View full text Add to dashboard Cite

Owing to the imminent fixed mobile convergence, Internet applications are frequently accessed through mobile nodes. However, service delivery latency is too high to satisfy user expectations. In this paper, we design a new TCP algorithm, TCP-ME (Mobile Edge), to accelerate the service delivery in mobile networks. Considering the QoS (Quality of Service) mechanisms of mobile networks, TCP-ME is designed to differentiate the packet loss caused by wireless errors, traffic conditioning of mobile core networks, and Internet congestion, as well as to react to the packet loss accordingly. To detect wireless errors, we mark the ACK (Acknowledge) packets in the uplink direction at the base station, and the marking threshold is a function of the instantaneous downlink queue length and the number of consecutive HARQ retransmissions. We modify the ECN mechanism with deterministic marking to detect Internet congestion. The packet loss caused by traffic conditioners of mobile networks is detected by whether the incoming DUPACK is marked or not. TCP-ME adapts the inter-packet interval when the packet loss is caused by wireless errors or the admission control mechanism. If the packet loss is due to Internet congestion, TCP-ME applies the TCP-New Reno's congestion window adaptation algorithm. Simulation results show that TCP-ME can speed up web service response time in mobile networks by about 80%.

show abstract

Improving TCP performance with bufferless token bucket policing: A TCP friendly policer

Cited by 12 publications

References 4 publications

Video Activity-Based Traffic Policing: A New Paradigm

Video Activity-Based Traffic Policing: A New Paradigm

An Internet-Wide Analysis of Traffic Policing

TCP-Mobile Edge: Accelerating delivery in mobile networks

Contact Info

Product

Resources

About