This paper presents the first detailed evaluation of Careful Resume, a new Congestion Control mechanism for Internet transports designed to accelerate connection startup based on retained path state. Traditional Congestion Control relies on Slow Start, which introduces significant latency for paths with a high Bandwidth-Delay Product. Careful Resume employs safety mechanisms to prevent excessive congestion when end-to-end signals suggest that the path characteristics, such as bandwidth or round-trip time (RTT), may have changed. We evaluate this safety using a testbed, followed by experiments using real-world Internet paths, and finally present initial experiences from use in a content delivery node. Results show that this new mechanism can significantly reduce the Time To Last Byte for a connection. In our tested scenarios, the reduction was by more than three RTTs. This finding aligns with results from testing Internet paths, where we observed reductions ranging from 2.8 to 3.4 RTTs on geostationary satellite and 4G/5G mobile paths. Our results also show that Careful Resume can be effectively used with HyStart++ to avoid further path capacity overshoot. This analysis demonstrates the benefit of using Careful Resume, and provides evidence that this mechanism does not unduly harm other connections that share the capacity of a network bottleneck.

Open Access via the Elsevier agreement The authors appreciate the valuable comments provided by Justin Iurman and Benoit Donnet and Bob Hinden. Elizabeth Boswell received funding from the University of Aberdeen to help analyse paths using Paris Traceroute. This work was supported by the University of Aberdeen’s School of Engineering Department , and experiments using Atlas probes were funded by the RIPE NCC Community Fund, Project ID 619935.

We propose XRC, an explicit rate control algorithm that overcomes the poor performance of commonly used TCP variants in cellular networks. XRC exploits explicit feedback from the radio access network that is aware of the physical, network and transport layer information of all UEs as well as resource distribution policies for users with different traffic characteristics. XRC co-exists fairly with other XRC and non-XRC flows at the wireless and non-wireless bottlenecks while it strictly controls queuing delay within a small threshold. We implement XRC in NS-3 and examine its performance across a range of network loads and dynamics. When competing with CUBIC at a wireless bottleneck, XRC achieves a Jain’s fairness index of 99.7% while providing a 3x lower median queuing delay compared to when CUBIC competes with CUBIC in the same setup.

“Roam like Home” is the initiative of the European Commission (EC) to end the levy of extra charges when roaming within the European region. As a result, people can use data services more freely across Europe. However, the implications of roaming solutions on network performance have not been carefully examined yet. This paper provides an in-depth characterization of the implications of international data roaming within Europe. We build a unique roaming measurement platform using 16 different mobile networks deployed in six countries across Europe. Using this platform, we measure different aspects of international roaming in 4G networks in Europe, including mobile network configuration, performance characteristics, and quality of experience. We find that operators adopt a common approach to implement roaming called Home-routed roaming (HR). This results in additional latency penalties of 60 ms or more, depending on geographical distance. This leads to worse browsing performance, with an increase in the metrics related to Quality of Experience (QoE) of users (Page Load time and Speed Index) in the order of 15-20 percent. We further analyze in isolation the impact of latency on QoE metrics and find that the penalty imposed by HR leads to a degradation on QoE metrics up to 150 percent in case of intercontinental roaming.

"Roam like Home" is the initiative of the European Commission (EC) to end the levy of extra charges when roaming within the European region. As a result, people are able to use data services more freely across Europe. However, the implications roaming solutions have on performance have not been carefully examined. This paper provides an in-depth characterization of the implications of international data roaming within Europe. We build a unique roaming measurement platform using 16 different mobile networks deployed in six countries across Europe. Using this platform, we measure different aspects of international roaming in 3G and 4G networks, including mobile network configuration, performance characteristics, and content discrimination. We find that operators adopt common approaches to implementing roaming, resulting in additional latency penalties of ∼60 ms or more, depending on geographical distance. Considering content accessibility, roaming poses additional constraints that leads to only minimal deviations when accessing content in the original country. However, geographical restrictions in the visited country make the picture more complicated and less intuitive.

The differentiated service architecture (DiffServ) provides a means for network devices to classify traffic based on the DiffServ codepoint (DSCP) and to map the traffic to a specific QoS forwarding treatment. Successful use beyond the local network depends on consistent remarking and forwarding of the DSCP value inside and at the boundaries of DiffServ domains. This paper provides the results of a new widescale measurement campaign to examine how the DSCP value is altered as packets travel along a set of Internet paths. This allows us to infer whether a packet is likely to receive an appropriate QoS treatment and to comment the opportunities for more widely deploying DiffServ QoS. Our results identify a set of remarking pathologies, revealing that many deployed routers continue to use the previous semantics of the deprecated Type of Service (ToS) field. We also note that is not common to observe clearing of the bits in the DiffServ field, as previously believed for routers in the core of the Internet, although this varies significantly depending on the type of network studied. Our results are related to recent IETF work that recommends use of specific DSCP values.

The deployment of encrypted transport protocols imposes new challenges for network operations. Key in-network functions such as those implemented by firewalls and passive measurement devices currently rely on information exposed by the transport layer. Encryption, in addition to improving privacy, helps to address ossification of network protocols caused by middleboxes that assume certain information to be present in the clear. However, “encrypting it all” risks diminishing the utility of these middleboxes for the traffic management tasks for which they were designed. A middlebox cannot use what it cannot see. We propose an architectural solution to this issue, by introducing a new “path layer” for transport-independent, in-band signaling between Internet endpoints and network elements on the paths between them, and using this layer to reinforce the boundary between the hop-by-hop network layer and the end-to-end transport layer. We define a path layer header on top of UDP to provide a common wire image for new, encrypted transports. This path layer header provides information to a transportindependent on-path state machine that replaces stateful handling currently based on exposed header flags and fields in TCP; it enables explicit measurability of transport layer performance; and offers extensibility by sender-to-path and path-to-receiver communications for diagnostics and management. This provides not only a replacement for signals that are not available with encrypted traffic, but also allows integrity-protected, enhanced signaling under endpoint control. We present an implementation of this wire image integrated with the QUIC protocol, as well as a basic stateful middlebox built on Vector Packet Processing (VPP) provided by FD.io.

There is a growing concern that the Internet transport layer has become less adaptive to the requirements of new applications, and that further evolution has become very difficult. This is because a fundamental assumption no longer holds: it can no longer be assumed that the transport layer is only in the scope of end-hosts. The success of TCP and UDP and the ubiquity of middleboxes have led to ossification of both the network infrastructure and the API presented to applications. This has led to the development of workarounds and point solutions that fail to cover many facets of the problem. To address this issue, this paper identifies requirements for a new transport layer and then proposes a conceptual architecture that we argue is both flexible and evolvable. This new architecture requires that applications interface to the transport at a higher abstraction level, where an application can express communication preferences via a new richer API. Protocol machinery can use this information to decide which of the available transport protocols is used. By placing the protocol machinery in the transport layer, the new architecture can allow for new protocols to be deployed and enable evolution of the transport layer.

Delay-sensitive applications, such as live and interactive video, are mainstream in today's Internet, and set to increase with the emergence of web-based video conference. For such applications quality cannot be captured solely by a flow's throughput: interactive video needs to avoid fluctuations in both visual quality and delay. This is far more valuable than achieving a fleeting increase in throughput. Emerging real-time protocols are being designed with these goals in mind, but it is important to evaluate these methods when sharing the network with real-world traffic. Much of today's Internet traffic uses TCP CUBIC, we therefore quantify its impact on interactive video experience (visual quality and delay): we measure the degradation imposed to interactive video when sharing a network bottleneck with TCP CUBIC traffic. To understand this impact, we compare performance when two delay-based congestion control schemes are used, one for interactive video and another for file transfer and show that these algorithms can assure good video experience and appropriate download time to their respective users. In contrast, we suggest that achieving such a “low-delay coexistence” with TCP CUBIC would require use of Active Queue Management (AQM) techniques. The paper therefore provides quantitative evidence that AQM can force loss-based TCP and delay-sensitive flows to reach a stable equilibrium point that is similar to the one naturally achieved when TCP flows are governed by delay-based mechanisms.

With multimedia and Internet enabled devices being ubiquitous, mechanisms that ensure multimedia flows do not congest the Internet are crucial components of multimedia systems that are embraced rather than opposed by network service providers. The Real-time Transport Protocol (RTP) Circuit Breaker is designed to terminate RTP/UDP flows that cause excessive congestion in the network. Multimedia users congesting the network have their flows terminated, as dictated by the RTP circuit breaker congestion rule. Users who obtain little quality from a multimedia session, and consume network resources to no avail, should also cease transmission. This is the mandate of the RTP circuit breaker media usability rule. We propose an algorithm for this rule, and show that it avoids wasting network resources on flows that deliver no quality to the user.

A recent proposed update to TCP congestion control, TCP-newCWV, has targeted congestion control for rate-limited applications. These methods need to be explored in the context of rate-adaptive applications, such as DASH. The new method enables a client to exploit the persistence of a DASH connection and enables the DASH server to rapidly resume transmission of a series of video segments using a single TCP connection. Another technique, called 'Pacing' smoothes DASH bursts when there is no TCP ACK clock, and is shown to significantly reduce burst loss. These two methods in combination can increase the application performance. This paper investigates the effect of implementing these techniques on a DASH flow in different congestion scenarios and whether the method can promote better capacity sharing while minimizing the latency experienced by other flows sharing a common network bottleneck. The results confirm that newCWV with Pacing provides a benefit as a platform for DASH transport.

For many years Internet routers have been designed and benchmarked in ways that encourage the use of large buffers. When these buffers accumulate a large standing queue, this can lead to high network path latency. Two AQM algorithms: PIE and CoDel, have been recently proposed to reduce buffer latency by avoiding the drawbacks of previous AQM algorithms like RED. This paper explores the performance of these new algorithms in simulated rural broadband networks where capacity is limited. We compared the new algorithms using Adaptive RED as a reference. We observe that to achieve a small queuing delay PIE and CoDel both increase packet loss. We therefore explored this impact on the quality of experience for loss-sensitive unreliable multimedia applications, such as real-time and near-real-time video. The results from simulations show that PIE performs better than CoDel in terms of packet loss rates affecting video quality. We also noted that the performance of ARED is comparable to that of PIE and CoDel in constant capacity links. This suggests that AQM in general is useful for limited capacity network paths.

Real-time media applications often ignore ongoing congestion if there is no option to reduce the rate. These applications pose a threat to themselves and other traffic. Reducing the transmission rate requires reducing the amount of packets rather than spreading the transmission over a longer interval. Loss-based congestion control mechanisms are unsuitable for this requirement. Also this rate reduction with popular video codecs, e.g. MPEG4, is often problematic. This paper investigates the problems associated with real-time video transmission over the Internet. We investigate a rate control method minimizing delay and losses and report preliminary but promising results.

Many modern Internet applications require fast client-server interaction to maximise user interactivity. This need has caused new work to be proposed to working groups of the Internet Engineering Task Force (IETF), which seek to remove bottlenecks when starting sessions. RFC 6298 improves Transmission Control Protocol (TCP) responsiveness by retransmitting lost SYNs, or first data segments sent after the three-way handshake (3WHS), with a reduced initial retransmission timeout (RTO) of 1 second from the previous value of 3 seconds. This benefits a large number of Internet connections. However, there is growing use of wireless and other long-haul radio technologies that yield round trip times (RTT) greater than 1 second. This means a significant number of connections will suffer spurious retransmissions with associated performance penalties. This paper examines whether RFC 6298 can cooperate with a set of proposed methods to extend performance benefits to long delay paths that may experience startup loss. It was found that a slight modification of TCP congestion-state initialization after the 3WHS significantly improves performance, especially if combined with SYN duplication and an increased Initial Window (IW). We further explain how a deprecated proposal to seed the RTO based on the SYN round trip time (RTT) can be updated to become suited for general deployment. Proposed modifications are evaluated, by analysis and using ns2 simulations, showing improved overall responsiveness for short delay paths as well as networks paths with shared bandwidth and appreciable delay (e.g. wireless/satellite networks).

Researches on multicast has explored security challenges faced by group communications. Multicast transport protocols and multicast security protocols need to work closely to provide reliable and secure multicast services. However, there has been comparatively little work carried out to specify how exactly the two technologies can work together efficiently. In this paper, authors present an example of partially integrating the Timed Efficient Stream Loss-Tolerant Authentication (TESLA) protocol and the File Delivery over Unidirectional Transport (FLUTE) protocol. TESLA provides source authorization and data integrity for multicast groups while FLUTE ensures reliability of the multicast transmission. This paper proposes a loose synchronization mechanism for a unidirectional transmission environment, suited to satellite networks. The proposed algorithm was implemented on a testbed with multicast tunnel between University of Surrey and University of Aberdeen and the results are presented in this paper.

Initially designed for TV broadcasting, DVB standard families have become mature technologies for Internet communications via satellite access networks. A generic encapsulation protocol is therefore required for the forward link, the return link, and for mesh satellite systems. Moreover, because of the cost and the scarcity of resources, this should be as efficient and light as possible. The GSE (Generic Stream Encapsulation) protocol enables efficient transport of any protocol, originally designed for DVB-S2, that is to say on Quasi Error Free (QEF) forward links of satellite systems. This paper addresses limitations in GSE for the return link context and examines the requirements for an encapsulation mechanism for satellite systems using multiple access. This includes the return link but also mesh satellite systems and any satellite systems that uses multiple access. This paper also analyses approaches to derive requirements and provides a first analysis of these requirements.

Digital video broadcast return channel via satellite system (RCS) is a key European standard that specifies a broadband satellite system using dynamic capacity assignment and multi-frequency time-division multiple access. This study introduces new architectural components for an RCS system to support advanced Internet protocol (IP)-oriented multimedia. This will allow closer integration of satellite networks with the next generation of multimedia-enabled networks enabling operators to offer triple-play services that simultaneously support voice over IP, television-over IP and other services. These types of services are particularly suited to next generation satellite systems that combine transparent and multi-spot beam satellites with on-board regenerative signal processing, and can evolve to use adaptive coding and modulation.

Multicast research has explored the security challenges faced in group communications. Multicast transport and multicast security need to work in close collaboration to realise a multicast service. However, there has been comparatively little work to combine the two technologies. In this paper the authors, presenting an example of partially integrating Timed Efficient Stream Loss-Tolerant Authentication (TESLA) protocol and the File Delivery over Unidirectional Transport (FLUTE) protocol. TESLA provides source authorization and data integrity for multicast groups. This paper proposes a loose synchronization mechanism for a unidirectional transmission environment, suited to satellite networks. It also analyses the effects on Forward Error Control (FEC) and the congestion control building blocks of FLUTE when used with TESLA.

The datagram congestion control protocol (DCCP) is a new IETF-defined transport protocol for Internet multimedia. The DCCP congestion control identifier 4 (CCID4) was specifically designed to support the stream of small packets generated by applications such as voice over IP (VoIP). This uses equation based congestion control. The long round trip delay when the path includes a satellite link could therefore degrade the performance of VoIP. In this paper, we analyze using simulation the performance of VoIP using CCID4 over a DVB-RCS network. We also analyze the different variants of CCID4 that have been proposed. The paper concludes that the current algorithm of CCID4 suffers for paths with appreciable delay when used to support voice calls with a bursty nature. The results show that these effects are only partially mitigated by the currently proposed variants of CCID4.