Bottleneck Link Research Articles

Implementation and Evaluation of Activity-Based Congestion Management Using P4 (P4-ABC)

Activity-Based Congestion management (ABC) is a novel domain-based QoS mechanism providing more fairness among customers on bottleneck links. It avoids per-flow or per-customer states in the core network and is suitable for application in future 5G networks. However, ABC cannot be configured on standard devices. P4 is a novel programmable data plane specification which allows defining new headers and forwarding behavior. In this work, we implement an ABC prototype using P4 and point out challenges experienced during implementation. Experimental validation of ABC using the P4-based prototype reveals the desired fairness results.

Enhanced Quality of Adaptive Video Streaming in HTTP-Based Protocol Using FESTIVE

For altering the network prerequisites most of the video participants fully relies upon logic i;e data rate adjustment which will modify data rate. In this paper there are three key things to do when in contrast to previous measurements i;e efficiency, fairness, stability. There is a problem known as bottleneck link which is identified when multiple data rate adaptive players watch a video. we have also mentioned about bitrate adaptation approach and three essential factors speed of transmission range, data rate choice and the process of chunk distribution. By this, everyone can discover the troubles of various countless negative interactions that occur as an effect of protecting the bitrate adaptation over HTTP. Based on appreciation of these techniques, we guide how to balance between fairness, stability and effectivity that will lead to robust video adaptation.

Machine learning for measurement-based bandwidth estimation

The dispersion that arises when packets traverse a network carries information that can reveal relevant network characteristics. Using a fluid-flow model of a bottleneck link with first-in first-out multiplexing, accepted probing tools measure the packet dispersion to estimate the available bandwidth, i.e., the residual capacity that is left over by other traffic. Difficulties arise, however, if the dispersion is distorted compared to the model, e.g., by non-fluid traffic, multiple bottlenecks, clustering of packets due to interrupt coalescing, and inaccurate time-stamping in general. It is recognized that modeling these effects is cumbersome if not intractable. This motivates us to explore the use of machine learning in bandwidth estimation. We train a neural network using vectors of the packet dispersion that is characteristic of the available bandwidth. Our testing results reveal that even a shallow neural network identifies the available bandwidth with high precision. We also apply the neural network under a variety of notoriously difficult conditions that have not been included in the training, such as randomly generated networks with the multiple bottleneck links and heavy cross traffic burstiness. Compared to two state-of-the-art model-based techniques as well as a recent machine learning-based technique (Yin et al., 2016), our neural network approach shows improved performance. Further, our neural network can effectively control the estimation procedure in an iterative implementation. We also evaluate our method with other supervised machine learning techniques.

Measuring bandwidth and buffer occupancy to improve the QoE of HTTP adaptive streaming

Live and on-demand video streaming service systems consume significant portion of Internet traffic all over the world. HTTP adaptive streaming is becoming the de-facto standard for adaptive video streaming solutions. Conventional estimation scheme for bandwidth estimation is not appropriate to estimate the bandwidth when multiple clients compete for a common bottleneck link, due to the ON–OFF traffic pattern. They overestimate the network bandwidth which leads to degradation of quality of experience by unnecessary changes in video quality, average video quality and unfairness of video quality. In this paper, we proposed receiver-side bandwidth-measured method to achieve a better quality of experience in multiple-client scenario. The proposed method estimates the obtainable network bandwidth based on the buffer status and segment throughput. The video buffer model is associated with three thresholds (i.e., one for initial start-up and two for operating thresholds). NS-3 network simulator has been deployed to measure the performance of HTTP adaptive streaming. Simulation outputs reflect that the proposed method enhances the quality of experience than conventional methods.

Increasing multicast transmission rate with localized multipath in software-defined networks

Network layer multicast is a highly efficient one-to-many transmission mode. Data rates supported by different group members may differ if these members are located in different network environments. Currently there are roughly two types of methods solving the problem, one is limiting the data rate so that every group member can sustain transmissions, and the other is building multiple trees to increase the provision of network bandwidth. The former is inefficient in bandwidth usage, and the latter adds too many states in the network, which is a serious problem in Software-Defined Networks. In this paper, we propose to build localized extra path(s) for each bottleneck link in the tree. By providing extra bandwidth to reinforce the bottleneck links, the overall data rate is increased. As extra paths are only built in small areas around the bottleneck links, the number of states added in the network is restrained to be as small as possible. Experiments on Mininet verify the effectiveness of our solution.

RecFlow: SDN-based receiver-driven flow scheduling in datacenters

Datacenter applications (e.g., web search, recommendation systems, and social networking) are designed to have a high fanout for the purpose of achieving scalable performance. Frequent fabric congestion (e.g., due to incast, imperfect hashing) is a corollary of such a design. This is true even when the network utilization is low. Such fabric congestion exhibits both temporal as well as spatial (intra-rack and inter-rack) variations. There exist two basic design paradigms which are used to address this issue. Current solutions lie somewhere between the two. On one hand we have arbiter based approaches where senders poll a centralized arbiter and collectively obey global scheduling decisions. On the other end of the spectrum, we have self adjusting end point based approaches where senders independently adjust transmission rate based on network congestion. The former incurs greater overhead, compared to the latter which trades off complexity for sub-optimality. Our work seeks a middle ground - optimality of arbiter based approaches with the simplicity of self adjusting end point based approaches. Our key design principle is that since the receiver has complete information regarding the flows destined for it, rather than having a centralized arbiter schedule flows or the senders making independent scheduling decisions, the receiver can orchestrate the various flows destined for it. Since multiple receivers may be using a bottleneck link, datapath visibility should be used to ensure fair sharing of the bottleneck capacity between receivers with minimum overhead. We propose RecFlow, which is a receiver-based proactive congestion control scheme. RecFlow employs OpenFlow provided path visibility to track changing bottlenecks on the fly. It spaces TCP acknowledgements to prevent traffic bursts and ensure that no receiver exceeds its fair share of the bottleneck capacity. The goal is to reduce buffer overflows while maintaining fairness among flows and high link utilization. Using extensive simulation results and real testbed evaluation, we show that compared to the state-of-the-art, RecFlow achieves up to 6× improvement in the inter-rack scenario and 1.5× in the intra-rack scenario while sharing the link capacity fairly between all flows.

Minstrel PIE: Curtailing queue delay in unresponsive traffic environments

Active Queue Management (AQM) algorithms aim to maintain a proper trade-off between queue delay and bottleneck link utilization. However, it is often noticed that this trade-off is not achieved convincingly when unresponsive UDP flows coexist with responsive TCP flows. This paper proposes an extension to Proportional Integral controller Enhanced (PIE) algorithm called Minstrel PIE, which adapts the reference queue delay to improve the trade-off between queue delay and link utilization when unresponsive flows share the same bottleneck queue as responsive flows. Extensive evaluations through simulations and real time experiments demonstrate that Minstrel PIE improves the performance of PIE in the presence of unresponsive flows, and delivers similar performance otherwise. Moreover, the Minstrel PIE algorithm does not introduce new knobs to improve the performance of PIE and hence, can be easily deployed without any additional complexity.

The Research about Emergency Logistics Capacity Bottleneck Based on Emergency Logistics Network

This paper focuses on improving the overall capacity of the physical elements of the emergency logistics network. On the basis of emergency logistics network, this paper quantifies the logistics ability of emergency materials passing through each node and line in the network, and introduces the constraint theory to find out the bottleneck link of emergency logistics network, then improve the bottleneck link pertinently and improve the ability of the whole emergency logistics network. Finally, a concrete example is given to verify the method. It will help to better solve the logistics problems in emergencies such as disasters.

XMAS: An Efficient Mobile Adaptive Streaming Scheme Based on Traffic Shaping

Adaptive video streaming can provide adequate Quality of Experience by dynamically adjusting video rates in responding to fluctuating service conditions. However, adaptive video streaming shows dismal performances in wireless networks where several players share a wireless bottleneck link. This paper proposes a novel video rate selection scheme called XMAS for efficient video streaming in wireless networks. XMAS consists of two components: an available bandwidth estimation part and a video rate selection part. To redress the problem of inaccurate bandwidth estimation due to peculiar on-off transmission patterns of mobile video streaming, we devised a novel client-based traffic shaping scheme that effectively throttles server's packet transmission. Equipped with accurate bandwidth estimates, XMAS determines target transmission rates considering playback buffer levels. We implemented XMAS on Linux and performed rigorous experiments to analyze its behavior and performance. Our performance results showed that XMAS achieves up to 20% increase in average video rates while reducing rebuffer rates significantly.

An Optimal Peer Selection Algorithm for Mesh-based Peer-to-Peer Networks

In order to achieve faster content distribution speed and stronger fault tolerance, a P2P peer can connect to multiple peers in parallel and receive chunks of the data simultaneously. A critical issue in this environment is selecting a set of nodes participating in swarming sessions. Previous related researches only focus on performance metrics, such as downloading time or the round-trip time, but in this paper, we consider a new performance metric which is closely related to the network and propose a peer selection algorithm that produces the set of peers generating optimal worst link stress. We prove that the optimal algorithm is practicable and has the advantages with the experiments on PlanetLab. The algorithm optimizes the congestion level of the bottleneck link. It means the algorithm can maximize the affordable throughput. Second, the network load is well balanced. A balanced network improves the utilization of resources and leads to the fast content distribution. We also notice that if every client follows our algorithm in selecting peers, the probability is high that all sessions could benefit. We expect that the algorithm in this paper can be used complementary to existing methods to derive new and valuable insights in peer-to-peer networking.

Congestion pricing in a real-world oriented agent-based simulation context

This paper investigates optimal congestion pricing strategies using a real-world oriented agent-based simulation framework which allows for complex user behavior. The applied simulation approach accounts for iteratively learning transport users, stochastic demand, and only approximates the user equilibrium, which may be considered as closer to real-world than a model where transport users behave completely rational, have a perfect knowledge about all travel alternatives, and travel behavior strictly follows the user equilibrium. Two congestion pricing rules are developed and investigated. The first one directly builds on the Pigouvian taxation principle and computes marginal external congestion costs based on the queuing dynamics at the bottleneck links; resulting toll payments differ from agent to agent depending on the position in the queue (QCP approach). The second one uses control-theoretical elements to adjust toll levels depending on the congestion level in order to reduce or eliminate traffic congestion; resulting toll payments are the same for all travelers per time bin and road segment (LP approach). The pricing rules are applied to Vickrey's bottleneck model and the case study of the Greater Berlin area. The simulation experiments reveal that with and without mode and departure time choice, the rather simple LP rule results in a higher system welfare compared to the more complex QCP approach. The LP rule appears to better take into account the system's dynamics and the agents' learning behavior. The results also reveal that pricing significantly reduces traffic congestion, however, there is still a remaining delay, even with departure time choice. Overall, this paper points out further need for research and contributes to the exploration of optimization heuristics for real-world oriented simulation approaches.

Characterizing the Rate-Memory Tradeoff in Cache Networks Within a Factor of 2

We consider a basic caching system, where a single server with a database of $N$ files (e.g., movies) is connected to a set of $K$ users through a shared bottleneck link. Each user has a local cache memory with a size of $M$ files. The system operates in two phases: a placement phase, where each cache memory is populated up to its size from the database, and a following delivery phase, where each user requests a file from the database, and the server is responsible for delivering the requested contents. The objective is to design the two phases to minimize the load (peak or average) of the bottleneck link. We characterize the rate-memory tradeoff of the above caching system within a factor of 2.00884 for both the peak rate and the average rate (under uniform file popularity), improving the state of the arts that are within a factor of 4 and 4.7, respectively. Moreover, in a practically important case where the number of files ( $N$ ) is large, we exactly characterize the tradeoff for systems with no more than five users and characterize the tradeoff within a factor of 2 otherwise. To establish these results, we develop two new converse bounds that improve over the state of the art.

Modelling and analysis of TCP congestion control mechanisms using stochastic reward nets

Modelling of congestion control mechanisms of the transmission control protocol (TCP) helps to obtain parametric results that help to better understand the TCP behaviour under different operating conditions. Several analytical models were proposed to analyse the behaviour of TCP congestion control mechanisms. However, most of these models were designed for a few TCP sessions with non-persistent connections. A few analytical studies were presented for modelling persistent TCP connections. Nevertheless, these studies have many limiting assumptions and are not scalable. In this work, a stochastic reward nets model is introduced for the TCP Reno with the persistent TCP connections, which share two bottleneck links in a wide area network. A microscopic approach was used for constructing the proposed model that captures most features of congestion control mechanisms used by the TCP Reno, and interactions between different TCP sessions. The proposed model relaxed several limiting assumptions adopted by other related models introduced in the literature. To validate the proposed model, analytical results are extensively compared with simulation results.

TFDASH: A Fairness, Stability, and Efficiency Aware Rate Control Approach for Multiple Clients Over DASH

Dynamic adaptive streaming over HTTP (DASH) has recently been widely deployed in the Internet and adopted in the industry. It, however, does not impose any adaptation logic for selecting the quality of video fragments requested by clients and suffers from lackluster performance with respect to a number of desirable properties: efficiency, stability, and fairness when multiple players compete for a bottleneck link. In this paper, we propose a throughput-friendly DASH (TFDASH) rate control scheme for video streaming with multiple clients over DASH to well balance the trade-offs among efficiency, stability, and fairness. The core idea behind guaranteeing fairness and high efficiency (bandwidth utilization) is to avoid OFF periods during the downloading process for all clients, i.e., the bandwidth is in perfect-subscription or over-subscription with bandwidth utilization approach to 100\%. We also propose a dual-threshold buffer model to solve the instability problem caused by the above idea. As a result, by integrating these novel components, we also propose a probability-driven rate adaption logic taking into account several key factors that most influence visual quality, including buffer occupancy, video playback quality, video bit-rate switching frequency and amplitude, to guarantee high-quality video streaming. Our experiments evidently demonstrate the superior performance of the proposed method.

CCMA: A Dynamical Concurrent-Connection Management Agent to Mitigate TCP Incast in Datacenters

While aggregator applications with many-to-one TCP traffic pattern prevail in datacenter networks, they often suffer from catastrophic goodput degradation due to their highly bursty traffic overwhelming a single bottleneck link, which is known as the TCP Incast issue. In this paper, we first derive three analytical results that mitigate Incast via managing connections, i.e., equalizing the in-flight packet amounts of concurrent connections, closing all the connections with higher timeout possibilities, and closing some of the connections with identical timeout possibilities. Afterward, we design an Incast mitigation application, i.e., concurrent-connection management agent (CCMA). Unlike previous Incast solutions, CCMA is an application-layer agent on the receiver, which manages concurrent TCP connections on behalf of aggregator applications. CCMA requires no modification to existing communication hardware or software. For multiple coexisting aggregators, CCMA sequentially schedules them to avoid traffic collision. For each aggregator, CCMA dynamically adjusts the advertised window sizes and the concurrency number of each aggregator following our analytical results to achieve low Incast probability and high goodput. The extensive real-world experiments and NS3 simulations show that CCMA maintains zero Incast probability while greatly improving the goodput of TCP by 7.8–13.5 times.

What happens when stochastic adaptive video streaming players share a bottleneck link?

What happens when stochastic adaptive video streaming players share a bottleneck link?

B-DASH: broadcast-based dynamic adaptive streaming over HTTP

Multiple video players competing at a bottleneck link give rise to overlapping ON-OFF traffic patterns. These patterns may introduce the problem of overlapping oscillatory effects amongst players resulting in poor video quality, frequent flickering and video freezes. We propose a distributed heuristic hybrid-based approach called B-DASH to address this problem. B-DASH implements two mechanisms: 1) message exchange; 2) adaptation. During message exchange, a broadcast mechanism is used to send information to neighbouring players. Based on the information present in this exchange, players can adapt to different network conditions. In this adaptation approach, B-DASH players with the higher estimated bandwidth are forced to reduce their bitrate request, while players with the lowest estimated bandwidth increase their bitrate request. This action reduces the oscillatory effects of the ON-OFF traffic patterns. Adaptive video streaming players using the B-DASH approach are compared to other players using conventional, PANDA and ELASTIC approaches. Experimental evaluation of B-DASH indicates good quality of experience (QoE) characteristics for players. Possible trade-offs for players using the B-DASH approach is investigated under various network conditions. The results show improvements of up to 5% bandwidth utilisation, 79% fairness, 27% re-buffering ratios, 57% stability and 9% quality.

A Proactive Flow Admission and Re-Routing Scheme for Load Balancing and Mitigation of Congestion Propagation in SDN Data Plane

The centralized architecture in software-defined network (SDN) provides a global view of the underlying network, paving the way for enormous research in the area of SDN traffic engineering (SDN TE). This research focuses on the load balancing aspects of SDN TE, given that the existing reactive methods for data-plane load balancing eventually result in packet loss and proactive schemes for data plane load balancing do not address congestion propagation. In the proposed work, the SDN controller periodically monitors flow level statistics and utilization on each link in the network and over-utilized links that cause network congestion and packet loss are identified as bottleneck links. For load balancing the identified largest flow and further traffic through these bottleneck links are rerouted through the lightly-loaded alternate path. The proposed scheme models a Bayesian Network using the observed port utilization and residual bandwidth to decide whether the newly computed alternate path can handle the new flow load before flow admission which in turn reduces congestion propagation. The simulation results show that when the network traffic increases the proposed method efficiently re-routes the flows and balance the network load which substantially improves the network efficiency and the quality of service (QoS) parameters.

A self-tuning controller for queuing delay regulation in TCP/AQM networks

AQM router aims primarily to control the network congestion through marking/dropping packets which are used as congestion feedback in traffic sources to balance their flow rate. However, stabilizing queuing delay and maximizing link utilization have been considered as the main control objectives, especially in media dominated networks. Usually, most of the AQM algorithms are designed for a nominal operating point. However, time-varying nature of network parameters frequently violates their robustness bounds. In this paper, a self-tuning compensated PID controller is proposed to address the time-varying nature of network conditions caused by parameter variations and unresponsive connections. The proposed scheme consists of network parameter estimation and a self-tuning AQM. Traffic load, network delay, and bottleneck link capacity are the time-varying network parameters whose variation effects should be compensated by the controller gains adaptation. As the controller gains are simply and directly obtained from the dynamic model, the obtained self-tuning controller can reasonably adapt itself to different operating conditions, while preserving the simplicity of the PI controllers. Packet-level simulations using ns2 show the outperformance of the developed controller for both latency regulation and resource utilization.

Dynamic ECN marking threshold algorithm for TCP congestion control in data center networks

Recent years have witnessed the rapid development in data center, which hosts a wide variety of applications and services with diverse requirements for network performance. Data center networks (DCNs) have the characteristics of high bandwidth and low latency and TCP incast usually happens when multiple senders simultaneously communicate with a single receiver. In this paper, we first investigate the DCTCP, which is widely deployed in DCNs and uses Explicit Congestion Notification (ECN) as a congestion signal when queue length is larger than a fixed threshold. We demonstrate that the queuing delay may be high or the bottleneck link utilization can be insufficient due to inappropriate fixed threshold. Then we propose a dynamic ECN marking threshold algorithm based on the number of concurrent flows, referred to as DEMT, to improve the network performance. After that, an enhanced DEMT algorithm, referred to as EDEMT, is proposed by using the ratio of the queue length over the marking threshold to perceive the degree of network congestion. We also integrate DEMT with DCTCP to observe its behavior and theoretically analyze the throughput. Finally, simulation results demonstrate that the DEMT and EDEMT algorithms can effectively improve the network throughput while they can maintain a small queue length as well as the short flow completion time.