Large-scale scanning of TCP's initial window

Rüth, Jan; Bormann, C.; Hohlfeld, Oliver

doi:10.1145/3131365.3131370

Cited by 22 publications

(17 citation statements)

References 11 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In this paper, we broadly probe CDNs to gather an empirical understanding on how IW customization already takes place in today's Internet. By scanning CDN IW configurations, also from globally distributed vantage points, we extend previous work [14] and shine a light on the degree of customization that CDNs show today. Our results show that IW customization beyond standardized practices is already common practice and there exists a gap between standardization, research, and Internet reality.…”

Section: Introductionmentioning

confidence: 70%

Demystifying TCP Initial Window Configurations of Content Distribution Networks

Rüth

Hohlfeld

2018

2018 Network Traffic Measurement and Analysis Conference (TMA)

Self Cite

View full text Add to dashboard Cite

Driven by their quest to improve web performance, Content Delivery Networks (CDNs) are known adaptors of performance optimizations. In this regard, TCP congestion control and particularly its initial congestion window (IW) size is one long-debated topic that can influence CDN performance. Its size is, however, assumed to be static by IETF recommendationsdespite being network-and application-dependent-and only infrequently changed in its history. To understand if the standardization and research perspective still meets Internet reality, we study the IW configurations of major CDNs. Our study uses a globally distributed infrastructure of VPNs giving access to residential access links that enable to shed light on networkdependent configurations. We observe that most CDNs are well aware of the IW's impact and find a high amount of customization that is beyond current Internet standards. Further, we find CDNs that utilize different IWs for different customers and content while others resort to fixed values. We find various initial window configurations, most below 50 segments yet with exceptions of up to 100 segments-the tenfold of current standards. Our study highlights that Internet reality drifted away from recommended and standardized practices.

show abstract

Section: Introductionmentioning

confidence: 70%

Demystifying TCP Initial Window Configurations of Content Distribution Networks

Rüth

Hohlfeld

2018

2018 Network Traffic Measurement and Analysis Conference (TMA)

Self Cite

View full text Add to dashboard Cite

show abstract

“…RFC6928 [22] recommended in 2013 to increase the initial Congestion Window (IW) to 10 segments. In 2017 Ruth et al [23] observed how 85% of HTTP servers and 80% of TLS servers from Alexa top 1M list already supported this recommendation. In order to realize the full benefit of the large IW on server side, implementations on client side need to advertise an initial Receive Window of at least 10 segments.…”

Section: Resultsmentioning

confidence: 99%

Hic Sunt Proxies: Unveiling Proxy Phenomena in Mobile Networks

Zullo

Pescapè

Edeline

et al. 2019

2019 Network Traffic Measurement and Analysis Conference (TMA)

View full text Add to dashboard Cite

Over the years middleboxes have established themselves as a solution to a wide range of networking issues, progressively changing network landscape and turning the endto-end principle into a reminder of an Arcadian age of the Internet. Among them, proxies have found breeding ground especially in mobile networks that, moreover, have become the most popular way to access the Internet. In this paper, we present Mobile Tracebox, an Android measurement tool, and describe how its methodology, coping with the lack of privileges of mobile devices, can not only detect proxies but also characterize different facets, from their transport layer behavior to their location inside the network. Data collected from a crowdsourced deployment over more than 90 carriers and 350 Wi-Fi networks contributes to describe the potential of the tool and to draw a panorama of proxies across mobile networks. Our study confirms their prevalence and reveals that their scope is not limited to HTTP but can include several TCP services and even non standard ports. We detail the different implementations observed and delve into specific aspects of their configuration, like the initial Receive Window, the Window Scale factor or the set of Options supported, to understand how proxies can affect performance or obstruct extensions. Finally, we focus on fingerprinting and attempt to draw a dividing line between packet modifications performed by proxies and those performed by other classes of middleboxes.

show abstract

“…Recent large-scale measurements [55], [56] show that while most web servers use the default values of their TCP stacks, CDN operators usually apply much larger values of IW [56]. These researches also suggested that some CDNs customize their IW configuration based on the network type and/or the content type.…”

Section: Use Case: Option To Request Initial Congestion Windowmentioning

confidence: 99%

Beyond socket options: making the linux TCP stack truly extensible

Tran

Bonaventure

2019

2019 IFIP Networking Conference (IFIP Networking)

View full text Add to dashboard Cite

The Transmission Control Protocol (TCP) is one of the most important protocols in today's Internet. Its specification and implementations have been refined for almost forty years. The Linux TCP stack is one of the most widely used TCP stacks given its utilisation on servers and Android smartphones and tablets. However, TCP and its implementations evolve very slowly. In this paper, we demonstrate how to leverage the eBPF virtual machine that is part of the recent versions of the Linux kernel to make the TCP stack easier to extend.We demonstrate a variety of use cases where the eBPF code is injected inside a running kernel to update or tune the TCP implementation. We first implement the TCP User Timeout Option. Then we propose a new option that enables a client to request a server to use a specific congestion control scheme. Our third extension is a TCP option that sets the initial congestion window. We then demonstrate how eBPF code can be used to tune the acknowledgment strategy.

show abstract

Large-scale scanning of TCP's initial window

Cited by 22 publications

References 11 publications

Demystifying TCP Initial Window Configurations of Content Distribution Networks

Demystifying TCP Initial Window Configurations of Content Distribution Networks

Hic Sunt Proxies: Unveiling Proxy Phenomena in Mobile Networks

Beyond socket options: making the linux TCP stack truly extensible

Contact Info

Product

Resources

About