The peak-hopper: a new end-to-end retransmission timer for reliable unicast transport

Ekstrom, H.; Ludwig, R.

doi:10.1109/infcom.2004.1354671

Cited by 23 publications

(27 citation statements)

References 12 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In RFC 2988 [23] the minimum RTO has been set to a fixed value of Ì second. However, following [8] our results suggest a need for a more flexible implementation, especially for connections with small RTT's. We note that under our algorithm packets are not retransmitted too aggressively since the penalty of a premature retransmission timeout is rather high.…”

Section: Implementation Issuesmentioning

confidence: 76%

See 1 more Smart Citation

Optimizing TCP Retransmission Timeout

Kesselman

Mansour

2005

Networking - ICN 2005

View full text Add to dashboard Cite

Abstract-Delay spikes on Internet paths can cause spurious TCP timeouts leading to significant throughput degradation. However, if TCP is too slow to detect that a retransmission is necessary, it can stay idle for a long time instead of transmitting. The goal is to find a Retransmission Timeout (RTO) value that balances the throughput degradation between both of these cases. In the current TCP implementations, RTO is a function of the Round Trip Time (RTT) alone. We show that the optimal RTO that maximizes the TCP throughput need to depend also on the TCP window size. Intuitively, the larger the TCP window size, the longer the optimal RTO. We derive the optimal RTO for several RTT distributions. An important advantage of our algorithm is that it can be easily implemented based on the existing TCP timeout mechanism.

show abstract

Section: Implementation Issuesmentioning

confidence: 76%

“…Gurtov and Ludwig [12] present an enhanced version of the Eifel algorithm and show its performance benefits on paths with a high bandwidth-delay product. Ekstr and Ludwig [8] propose a new algorithm for calculating the RTO, named the Peak-Hopper-RTO (PH-RTO), that improves upon the performance of TCP in high loss environments.…”

Section: Introductionmentioning

confidence: 99%

Optimizing TCP Retransmission Timeout

Kesselman

Mansour

2005

Networking - ICN 2005

View full text Add to dashboard Cite

show abstract

“…For a wider understanding of congestion control for the IoT, besides comparing default CoAP with CoCoA, in this paper we also analyze other RTO calculation algorithms such as the Linux TCP RTO (Linux-RTO) estimator [4], the peak-hopper TCP RTO estimator (PH-RTO) [5] and also a CoCoA variant that only uses the strong RTO estimator (CoCoA-S).…”

Section: Alternative Congestion Control Mechanismsmentioning

confidence: 99%

“…In this setup, both the CoAP clients and server run the Java Californium (Cf) CoAP [6] implementation. The alternative congestion control mechanisms considered in this paper are implemented and publicly available in Cf 5 . For a fair comparative evaluation, while the PH-RTO and Linux algorithms were designed for TCP, we have implemented these algorithms for CoAP (over UDP).…”

Section: A Testbedsmentioning

confidence: 99%

CoAP congestion control for the internet of things

et al. 2016

View full text Add to dashboard Cite

Abstract-The Constrained Application Protocol (CoAP) is a lightweight RESTful application layer protocol devised for the Internet of Things (IoT). Operating on top of UDP, CoAP must handle congestion control by itself. The core CoAP specification defines a basic congestion control mechanism, which is however not capable of adapting to network conditions. Yet, IoT scenarios exhibit significant resource constraints which pose new challenges on the design of congestion control mechanisms. In this paper we present the CoAP Simple Congestion Control/Advanced (CoCoA), an advanced congestion control mechanism for CoAP being standardized by the Internet Engineering Task Force (IETF) CoRE working group. CoCoA introduces novel Round Trip Time (RTT) estimation techniques, together with a Variable Backoff Factor (VBF) and aging mechanisms in order to provide dynamic and controlled Retransmission Timeout (RTO) adaptation suitable for the peculiarities of IoT communications. We conduct a comparative performance analysis of CoCoA and a variety of alternative algorithms including state-of-the-art mechanisms developed for TCP. The study is based on experiments carried out in real testbeds. Results show that, in contrast with the alternative methods considered, CoCoA consistently outperforms default CoAP congestion control mechanism in all evaluated scenarios.

show abstract

“…Ekström and Ludwig [9] point out that the retransmission timeout algorithm defined in RFC2988 [19] and used in both TCP and SCTP responds sluggishly to sudden fluctuations in the RTT. This leads to extreme estimated RTO values in some cases.…”

Section: Related Workmentioning

confidence: 99%

Improving SCTP retransmission delays for time-dependent thin streams

Petlund

Beskow

Pedersen

et al. 2009

Multimed Tools Appl

View full text Add to dashboard Cite

A large number of network services rely on IP and reliable transport protocols. For applications that provide abundant data for transmission, loss is usually handled satisfactorily, even if the application is latency-sensitive (Wang et al. 2004). For data streams where small packets are sent intermittently, however, applications can occasionally experience extreme latencies (Griwodz and Halvorsen 2006). As it is not uncommon that such thin-stream applications are time-dependent, any unnecessarily induced delay can have severe consequences for the service provided. Massively Multiplayer Online Games (MMOGs) are a defining example of thin streams. Many MMOGs (like World of Warcraft and Age of Conan) use TCP for the benefits of reliability, in-order delivery and NAT/firewall traversal. It has been shown that TCP has several shortcomings with respect to the latency requirements of thin streams because of the way it handles retransmissions (Griwodz and Halvorsen 2006). As such, an alternative to TCP may be SCTP (Stewart et al. 2000), which was originally developed to meet the requirements of signaling transport. In this paper, we evaluate the Linux-kernel SCTP implementation in the context of thin streams. To address the identified latency challenges, we propose sender-side only enhancements that reduce the application-layer latency in a manner that is compatible with unmodified receivers. These enhancements can be switched on by applications and are used only when the system identifies the stream as thin. To evaluate the latency performance, we have performed several tests over various real networks and over an emulated network, varying parameters like RTT, packet loss and amount of competing cross traffic. When comparing our modifications with SCTP on Linux and FreeBSD and TCP New Reno, our results show great latency improvements and indicate the need for a separate handling of thin and thick streams.

show abstract

The peak-hopper: a new end-to-end retransmission timer for reliable unicast transport

Cited by 23 publications

References 12 publications

Optimizing TCP Retransmission Timeout

Optimizing TCP Retransmission Timeout

CoAP congestion control for the internet of things

Improving SCTP retransmission delays for time-dependent thin streams

Contact Info

Product

Resources

About