Constantinos Dovrolis

Bandwidth estimation: metrics, measurement techniques, and tools

In a packet network, the terms bandwidth and throughput often characterize the amount of data that the network can transfer per unit of time. Bandwidth estimation is of interest to users wishing to optimize end-to-end transport performance, overlay network routing, and peer-to-peer file distribution. Techniques for accurate bandwidth estimation are also important for traffic engineering and capacity planning support. Existing bandwidth estimation tools measure one or more of three related metrics: capacity, available bandwidth, and bulk transfer capacity. Currently available bandwidth estimation tools employ a variety of strategies to measure these metrics. In this survey we review the recent bandwidth estimation literature focusing on underlying techniques and methodologies as well as open source bandwidth measurement tools.

What do packet dispersion techniques measure

The packet pair technique estimates the capacity of a path (bottleneck bandwidth) from the dispersion (spacing) experienced by two back-to-back packets. We demonstrate that the dispersion of packet pairs in loaded paths follows a multimodal distribution, and discuss the queueing effects that cause the multiple modes. We show that the path capacity is often not the global mode, and so it cannot be estimated using standard statistical procedures. The effect of the size of the probing packets is also investigated, showing that the conventional wisdom of using maximum sized packet pairs is not optimal. We then study the dispersion of long packet trains. Increasing the length of the packet train reduces the measurement variance, but the estimates converge to a value, referred to as the asymptotic dispersion rate (ADR), that is lower than the capacity. We derive the effect of the cross traffic in the dispersion of long packet trains, showing that the ADR is not the available bandwidth in the path, as was assumed in previous work. Putting all the pieces together, we present a capacity estimation methodology that has been implemented in a tool called pathrate.

Proportional differentiated services: delay differentiation and packet scheduling

Internet applications and users have very diverse service expectations, making the current same-service-to-all model inadequate and limiting. In the relative differentiated services approach, the network traffic is grouped in a small number of service classes which are ordered based on their packet forwarding quality, in terms of per-hop metrics for the queueing delays and packet losses. The users and applications, in this context, can adaptivelychoose the class that best meets their quality and pricing constraints, based on the assurance that higher classes will be better, or at least no worse, than lower classes. In this work, we propose the proportional differentiation model as a way to refine and quantify this basic premise of relative differentiated services. The proportional differentiation model aims to provide the network operator with the tuning knobs for adjusting the quality spacing between classes, independent of the class loads; this cannot be achieved with other relative differentiation models, such as strict prioritization or capacity differentiation. We apply the proportional model on queueing-delay differentiation only, leaving the problem of coupled delay and loss differentiation for future work. We discuss the dynamics of the proportional delay differentiation model and state the conditions under which it is feasible. Then, we identify and evaluate (using simulations) two packet schedulers that approximate the proportional differentiation model in heavy-load conditions, even in short timescales. Finally, we demonstrate that such per-hop and class-based mechanisms can provide consistent end-to-end differentiation to individual flows from different classes, independently of the network path and flow characteristics.

End-to-end available bandwidth: measurement methodology, dynamics, and relation with TCP throughput

The available bandwidth (avail-bw) in a network path is of major importance in congestion control, streaming applications, QoS verification, server selection, and overlay networks. We describe an end-to-end methodology, called Self-Loading Periodic Streams (SLoPS), for measuring avail-bw. The basic idea in SLoPS is that the one-way delays of a periodic packet stream show an increasing trend when the streams rate is higher than the avail-bw. We implemented SLoPS in a tool called pathload. The accuracy of the tool has been evaluated with both simulations and experiments over real-world Internet paths. Pathload is non-intrusive, meaning that it does not cause significant increases in the network utilization, delays, or losses. We used pathload to evaluate the variability (dynamics) of the avail-bw in some paths that cross USA and Europe. The avail-bw becomes significantly more variable in heavily utilized paths, as well as in paths with limited capacity (probably due to a lower degree of statistical multiplexing). We finally examine the relation between avail-bw and TCP throughput. A persistent TCP connection can be used to roughly measure the avail-bw in a path, but TCP saturates the path, and increases significantly the path delays and jitter.

Packet-dispersion techniques and a capacity-estimation methodology

The packet-pair technique aims to estimate the capacity of a path (bottleneck bandwidth) from the dispersion of two equal-sized probing packets sent back to back. It has been also argued that the dispersion of longer packet bursts (packet trains) can estimate the available bandwidth of a path. This paper examines such packet-pair and packet-train dispersion techniques in depth. We first demonstrate that, in general, packet-pair bandwidth measurements follow a multimodal distribution and explain the causes of multiple local modes. The path capacity is a local mode, often different than the global mode of this distribution. We illustrate the effects of network load, cross-traffic packet-size variability, and probing packet size on the bandwidth distribution of packet pairs. We then switch to the dispersion of long packet trains. The mean of the packet-train dispersion distribution corresponds to a bandwidth metric that we refer to as average dispersion rate (ADR). We show that the ADR is a lower bound of the capacity and an upper bound of the available bandwidth of a path. Putting all of the pieces together, we present a capacity-estimation methodology that has been implemented in a tool called pathrate. We report on our experiences with pathrate after having measured hundreds of Internet paths over the last three years.

A case for relative differentiated services and the proportional differentiation model

Internet applications and users have very diverse quality of service expectations, making the same-service-to-all model of the current Internet inadequate and limiting. There is a widespread consensus today that the Internet architecture has to extended with service differentiation mechanisms so that certain users and applications can get better service than others at a higher cost. One approach, referred to as absolute differentiated services, is based on sophisticated admission control and resource reservation mechanisms in order to provide guarantees or statistical assurances for absolute performance measures, such as a minimum service rate or maximum end-to-end delay. Another approach, which is simpler in terms of implementation, deployment, and network manageability, is to offer relative differentiated services between a small number of service classes. These classes are ordered based on their packet forwarding quality, in terms of per-hop metrics for the queuing delays and packet losses, giving the assurance that higher classes are better than lower classes. The applications and users, in this context, can dynamically select the class that best meets their quality and pricing constraints, without a priori guarantees for the actual performance level of each class. The relative differentiation approach can be further refined and quantified using the proportional differentiation model. This model aims to provide the network operator with the tuning knobs for adjusting the quality spacing between classes, independent of the class loads. When this spacing is feasible in short timescales, it can lead to predictable and controllable class differentiation, which ore two important features for any relative differentiation model. The proportional differentiation model can be approximated in practice with simple forwarding mechanisms (packet scheduling and buffer management) that we describe.

Passive estimation of TCP round-trip times

We propose and evaluate a passive measurement methodology that estimates the distribution of Round-Trip Times (RTTs) for the TCP connections that flow through a network link. Such an RTT distribution is important in buffer provisioning, configuration of active queue management, and detection of congestion unresponsive traffic. The proposed methodology is based on two techniques. The first technique is applicable to TCP caller-to-callee flows, and it is based on the 3-way handshake messages. The second technique is applicable to callee-to-caller flows, when the callee transfers a number of MSS segments to the caller, and it is based on the slow-start phase of TCP. The complete estimation algorithm reports an RTT for 55-85% of the TCP workload, in terms of bytes, in the traces that we examined. Verification experiments show that about 90% of the passive measurements are within 10% or 5ms, whichever is larger, of the RTT that ping would measure. Also, measurements on several NLANR traces show that the two estimation techniques agree within 25ms for 70-80% of the processed TCP connections. We also apply the estimation methodology on a number of NLANR traces and examine the variability of the measured RTT distributions in both short and long timescales.

Beware of BGP attacks

This note attempts to raise awareness within the network research community about the security of the interdomain routing infrastructure. We identify several attack objectives and mechanisms, assuming that one or more BGP routers have been compromised. Then, we review the existing and proposed countermeasures, showing that they are either generally ineffective (route filtering), or probably too heavyweight to deploy (S-BGP). We also review several recent proposals, and conclude by arguing that a significant research effort is urgently needed in the area of routing security.

Ten years in the evolution of the internet ecosystem

Our goal is to understand the evolution of the Autonomous System (AS) ecosystem over the last decade. Instead of focusing on abstract topological properties, we classify ASes into a number of species depending on their function and business type. Further, we consider the semantics of inter-AS links, in terms of customer-provider versus peering relations. We find that the available historic datasets from RouteViews and RIPE are not sufficient to infer the evolution of peering links, and so we restrict our focus to customer-provider links. Our findings highlight some important trends in the evolution of the Internet over the last decade, and hint at what the Internet is heading towards. After an exponential increase phase until 2001, the Internet now grows linearly in terms of both ASes and inter-AS links. The growth is mostly due to enterprise networks and content/access providers at the periphery of the Internet. The average path length remains almost constant mostly due to the increasing multihoming degree of transit and content/access providers. In recent years, enterprise networks prefer to connect to small transit providers, while content/access providers connect equally to both large and small transit providers. The AS species differ significantly from each other with respect to their rewiring activity; content/access providers are the most active. A few large transit providers act as attractors or repellers of customers. For many providers, strong attractiveness precedes strong repulsiveness by 3-9 months. Finally, in terms of regional growth, we find that the AS ecosystem is now larger and more dynamic in Europe than in North America.

Proportional differentiated services, part II: loss rate differentiation and packet dropping

The proportional differentiation model was proposed in Dovrolis et al. (1999), as a target for controllable and predictable relative differentiated services. Only the delay differentiation aspect of the model was considered, and focused on packet scheduling mechanisms. In this paper, we extend the proportional differentiation model in the direction of loss rate differentiation. Several previous mechanisms for buffer management and packet dropping, such as complete buffer partitioning, partial buffer sharing, or multi-class RED, are not suitable for relative differentiated services. We propose and evaluate two dropping mechanisms that closely approximate the proportional loss rate differentiation model. The two droppers, PLR(/spl infin/) and PLR(M), differ in the time interval over which the loss rates are measured and proportionally adjusted. This difference results in several trade-offs, in terms of implementation complexity, accuracy, and ability to deal with nonstationary traffic loads. We also re-evaluate the delay differentiation that the waiting time priorities (WTP) scheduler, of Dovrolis et al., can achieve in the presence of finite buffers and packet losses; this study extends their results providing further insight into the behavior of WTP in heavy load conditions. Finally, we examine the coupled effect of delay and loss rate proportional differentiation on the throughput of bulk-transfer TCP connections.