Bradley Huffaker

Topology discovery by active probing

As the Internet has grown, so has the challenge of accurate measurement and modeling of its topology. Commonly used but coarse methods of measuring topology, e.g., BGP tables, suffer from several limitations. To pursue more accurate empirically-based topology modeling, in 1998 CAIDA began its Macroscopic Topology Project, which focuses on actively measuring topology and round trip time (RTT) information across a large cross-section of the commodity Internet. We describe CAIDAs topology measurement architecture and our analysis and visualization tools. We describe differences between IP and AS (BGP-based) granularities of topology modeling, including advantages and limitations of both, as well as how correlation between both types of data can yield more relevant insights. We introduce four new visualization metaphors for handling macroscopic topology data, as well as a tool for aggregating multiple IP addresses into the same physical router. We highlight results of our analyses, in particular relationships between RTT and topology data, and how source and destination selection and geopolitical boundaries affect those relationships.

Orbis: rescaling degree correlations to generate annotated internet topologies

Researchers involved in designing network services and protocols rely on results from simulation and emulation environments to understand their application performance and scalability. To better understand the behavior of these applications and predict their performance when deployed on the actual Internet, the generated topologies must closely match real network characteristics, not just in terms of graph structure (node interconnectivity) but also with respect to various node and link annotations. Relevant annotations include link latencies, AS membership and whether a router is a peering or internal router. Finally, it should be possible to rescale a given topology to a variety of sizes while still maintaining its essential characteristics. In this paper, we propose techniques to generate annotated, Internet router graphs of different sizes based on existing observations of Internet characteristics. We find that our generated graphs match a variety of graph properties of observed topologies for a range of target graph sizes. While the best available data of Internet topology currently remains imperfect, the quality of our generated topologies will improve with the fidelity of available measurement techniques or next generation architectures that make Internet structure more transparent.

AS relationships, customer cones, and validation

Business relationships between ASes in the Internet are typically confidential, yet knowledge of them is essential to understand many aspects of Internet structure, performance, dynamics, and evolution. We present a new algorithm to infer these relationships using BGP paths. Unlike previous approaches, our algorithm does not assume the presence (or seek to maximize the number) of valley-free paths, instead relying on three assumptions about the Internets inter-domain structure: (1) an AS enters into a provider relationship to become globally reachable; and (2) there exists a peering clique of ASes at the top of the hierarchy, and (3) there is no cycle of p2c links. We assemble the largest source of validation data for AS-relationship inferences to date, validating 34.6% of our 126,082 c2p and p2p inferences to be 99.6% and 98.7% accurate, respectively. Using these inferred relationships, we evaluate three algorithms for inferring each ASs customer cone, defined as the set of ASes an AS can reach using customer links. We demonstrate the utility of our algorithms for studying the rise and fall of large transit providers over the last fifteen years, including recent claims about the flattening of the AS-level topology and the decreasing influence of tier-1 ASes on the global Internet.

On routing asymmetry in the Internet

Routing asymmetry in the Internet can significantly affect the manner in which we model and simulate its behavior. In this paper, we study routing asymmetry in the Internet and present quantitative evaluations on the extent of such asymmetry today. Our quantitative evaluations provide a measure of the difference between the forward and reverse paths between two end points. Routing asymmetry has not been studied extensively before; this is primarily due to the lack of a systematic approach for quantifying asymmetry except for simply computing the difference between the forward and reverse path lengths. By applying our framework for representing asymmetry, we quantify routing asymmetry for both US higher education academic networks and general commercial networks at two different levels: the autonomous system (AS) level and the router (or link) level. We take into consideration, not only the difference in the forward and reverse path lengths, but also the AS and link identities and the sequence in which these entities appear on the paths. We measure the AS level routing asymmetry, and provide upper lower bounds on link level routing asymmetry. Our studies show that academic networks appear to be more symmetric than general commercially deployed networks. Furthermore, our studies demonstrate that routing asymmetry exhibits a skewed distribution i.e., a few end-points seem to display a higher extent of participation on asymmetric routes.

Traceroute probe method and forward IP path inference

Several traceroute probe methods exist, each designed to perform better in a scenario where another fails. This paper examines the effects that the choice of probe method has on the inferred forward IP path by comparing the paths inferred with UDP, ICMP, and TCP-based traceroute methods to (1) a list of routable IP addresses, (2) a list of known routers, and (3) a list of well-known websites. We further compare methods by examining seven months of macroscopic Internet topology data collected by CAIDAs Archipelago infrastructure. We found significant differences in the topology observed using different probe methods. In particular, we found that ICMP-based traceroute methods tend to successfully reach more destinations, as well as collect evidence of a greater number of AS links. UDP-based methods infer the greatest number of IP links, despite reaching the fewest destinations. We hypothesise that some per-flow load balancers implement different forwarding policies for TCP and UDP, and run a specific experiment to confirm this hypothesis.

Measuring the deployment of IPv6: topology, routing and performance

We use historical BGP data and recent active measurements to analyze trends in the growth, structure, dynamics and performance of the evolving IPv6 Internet, and compare them to the evolution of IPv4. We find that the IPv6 network is maturing, albeit slowly. While most core Internet transit providers have deployed IPv6, edge networks are lagging. Early IPv6 network deployment was stronger in Europe and the Asia-Pacific region, than in North America. Current IPv6 network deployment still shows the same pattern. The IPv6 topology is characterized by a single dominant player -- Hurricane Electric -- which appears in a large fraction of IPv6 AS paths, and is more dominant in IPv6 than the most dominant player in IPv4. Routing dynamics in the IPv6 topology are largely similar to those in IPv4, and churn in both networks grows at the same rate as the underlying topologies. Our measurements suggest that performance over IPv6 paths is comparable to that over IPv4 paths if the AS-level paths are the same, but can be much worse than IPv4 if the AS-level paths differ.

Inferring AS relationships: dead end or lively beginning?

Recent techniques for inferring business relationships between ASs [1,2] have yielded maps that have extremely few invalid BGP paths in the terminology of Gao[3]. However, some relationships inferred by these newer algorithms are incorrect, leading to the deduction of unrealistic AS hierarchies. We investigate this problem and discover what causes it. Having obtained such insight, we generalize the problem of AS relationship inference as a multiobjective optimization problem with node-degree-based corrections to the original objective function of minimizing the number of invalid paths. We solve the generalized version of the problem using the semidefinite programming relaxation of the MAX2SAT problem. Keeping the number of invalid paths small, we obtain a more veracious solution than that yielded by recent heuristics.

Identifying IPv6 network problems in the dual-stack world

One of the major hurdles limiting IPv6 adoption is the existence of poorly managed experimental IPv6 sites that negatively affect the perceived quality of the IPv6 Internet. To assist network operators in improving IPv6 networks, we are exploring methods to identify wide-area IPv6 network problems. Our approach makes use of parallel IPv4 and IPv6 connectivity to dual-stacked nodes.We identify the existence of an IPv6 path problem by comparing IPv6 delay measurements to IPv4 delay measurements. Our test results indicate that the majority of IPv6 paths have delay characteristics comparable to those of IPv4, although a small number of paths exhibit a much larger delay with IPv6. Thus, we hope to improve the quality of the IPv6 Internet by identifying the worst set of problems.Our methodology is simple. We create a list of systems with IPv6 and IPv4 addresses in actual use by monitoring DNS messages. We then measure delay to each address in order to select a few systems per site based on their IPv6:IPv4 response-time ratios. Finally, we run traceroute with Path MTU discovery to the selected systems and then visualize the results for comparative path analysis. This paper presents the tools used to support this study, and the results of our measurements conducted from two locations in Japan and one in Spain.

The internet measurement data catalog

Internet data remains one of the basic components of computer science network research. Despite its necessity, available data is limited by legal, social, and technical constraints on its collection and distribution. Thus, optimal distribution of knowledge about available data is a valuable service to the research community. To this end, CAIDA has developed the Internet Measurement Data Catalog to:provide a searchable index of available dataenhance documentation of datasets via a public annotation systemadvance network science by promoting reproducible researchThis paper describes the impetus, design, and planned deployment of the Internet Measurement Data Catalog.

Inferring Complex AS Relationships

The traditional approach of modeling relationships between ASes abstracts relationship types into three broad categories: transit, peering, and sibling. More complicated configurations exist, and understanding them may advance our knowledge of Internet economics and improve models of routing. We use BGP, traceroute, and geolocation data to extend CAIDAs AS relationship inference algorithm to infer two types of complex relationships: hybrid relationships, where two ASes have different relationships at different interconnection points, and partial transit relationships, which restrict the scope of a customer relationship to the providers peers and customers. Using this new algorithm, we find 4.5% of the 90,272 provider-customer relationships observed in March 2014 were complex, including 1,071 hybrid relationships and 2,955 partial-transit relationships. Because most peering relationships are invisible, we believe these numbers are lower bounds. We used feedback from operators, and relationships encoded in BGP communities and RPSL, to validate 20% and 6.9% of our partial transit and hybrid inferences, respectively, and found our inferences have 92.9% and 97.0% positive predictive values. Hybrid relationships are not only established betweenlarge transit providers; in 57% of the inferred hybrid transit/peering relationships the customer had a customer cone of fewer than 5 ASes.