Downtown Dubai
IEEE site
IEEE Computer Society

Monday, November 7, Ruby Room

08:45 - 09:00

WNM: Welcome and Opening Remark

Chair: Aniket Mahanti (University of Auckland, New Zealand)

9:00 AM - 10:30 AM

WNM Session 1:

Chair: Saad B. Qaisar (School of Electrical Engineering and Computer Science (SEECS), NUST & National University of Sciences & Technology, Pakistan)

9:00 On the Analysis of Internet Paths with DisNETPerf, a Distributed Paths Performance Analyzer

Sarah Wassermann (Universite de Liege, Belgium); Pedro Casas (Austrian Institute of Technology (AIT), Austria); Benoit Donnet (Universite de Liege (ULg), Belgium); Guy Leduc (University of Liege, Belgium); Marco Mellia (Politecnico di Torino, Italy) traceroute is the most widely used Internet path analysis tool today, largely used by both researchers and operators to study the topology of the Internet and to diagnose routing failures and poor performance events. A major limitation of traceroute when the destination is not controllable by the user is its inability to measure \emph{reverse} paths, i.e., the path from any given destination back to the source. This is a major drawback for ISPs, who need to understand the performance of the Internet paths connecting popular services (e.g., YouTube, Facebook, etc.) to their customers. Even if multiple public servers distributed across the Internet and distributed measurement platforms can ease this issue by providing partial reverse path visibility through adhoc measurements, there is still a real need for a structured approach capable of analyzing the performance of Internet paths connecting any pair of nodes (i.e., servers, routers, hosts, etc.). While the problem of \emph{reverse} traceroute has been addressed in the past, proposed techniques rely on IP address spoofing -- which might lead to security concerns, and assume the availability of certain route-tracking options -- which might not be available. In this paper we introduce and evaluate DisNETPerf, a new tool which provides exactly the same type of information as traceroute, but for paths connecting arbitrarily selected nodes. DisNETPerf works by firstly locating probes (i.e., measurement points) that are the closest to a given target node, using then these probes to perform traceroute measurements from the target point-of-view to a given destination for path performance monitoring and troubleshooting purposes. We propose two techniques for probe location, and demonstrate that the reverse path (from server to users) can be measured with very high accuracy in certain scenarios. Our paper additionally analyzes relevant characteristics of Internet paths and distributed measurement platforms, which reinforce the applicability and relevance of DisNETPerf in current Internet.

9:30 Packet-Pair Dispersion Signatures in Multihop Networks

Negar Mosharraf and Anura P Jayasumana (Colorado State University, USA) Packet-pair technique is a widely used method for characterizing end-to-end network paths. A new analytical model is presented for the packet-pair based signature that accurately describes the behavior of packet-pairs in multihop network paths with multiple tight links. The relationship between the input and output gaps of packet-pairs and the corresponding distribution of end-to-end packet-pair dispersions are derived. Then it used to derive the signature characterizing the path. The model is verified via OPNETĀ® based simulations. We explore how the signature is shaped by factors such as number of the hubs, initial dispersion, and cross traffic. Path signature is an essential tool for numerous applications that need to distinguish between different network paths. The analytical model provides deeper insights to path signatures in the presence of multiple tight links, thus enabling accurate network monitoring, problem diagnosis, and estimation of link and path parameters such as network bandwidths and link capacities.

10:00 Efficient Network Topology Measurement Based on Ingress to Subnet Reachability

Ibrahim Coskun, Muhammed Abdullah Canbaz and Mehmet Hadi Gunes (University of Nevada, Reno, USA) Internet topology measurement studies rely on a few vantage points to trace towards a large number of destinations. Such traces contain significant overlaps, which increase the measurement traffic and duration. As there is considerable redundancy in the generated measurement probes, researchers have been exploring probe reduction approaches with minimal loss of the topological information. To decrease the probing overhead, in this paper, we propose InSub probing technique which relies on ingress points of the Autonomous System (AS) towards the target subnetworks. To identify ingress points for each subnetwork of the target AS, we first probe the subnet prefixes announced via BGP. After identifying the ingress points of subnetworks, the subsequent probing is conducted based on the vantage point reachability of the ingresses. As shown in an experimental study with 10 ASes, InSub approach considerably reduces the probing overhead while minimizing the loss of topological information.

10:30 PM - 11:00 AM

Coffee Break

11:00 AM - 12:30 PM

WNM Session 2:

Chair: Anura P Jayasumana (Colorado State University, USA)

11:00 Stability and Consistency of the LISP Pull Routing Architecture

Yue Li (Telecom ParisTech, France); Damien Saucez (INRIA, France); Luigi Iannone (Telecom ParisTech, France); Benoit Donnet (Universite de Liege (ULg), Belgium) Future Internet has been a hot topic for the last decade. One of the approaches put forward in order to revise the Internet architecture is LISP - Locator/ID Separation Protocol, which leverages on the separation of the identifier and the locator roles of IP addresses. Contrary to the classical push model used by the BGP-based routing architecture, LISP relies on a pull model. In particular, routing information is pulled from a new network element, the Mapping System, to provide the association between the identifier (i.e., the address used to identify a host inside a domain) and a list of locators (i.e., the addresses to locate an attachment point) upon an explicit query. In this paper, we evaluate a LISP Mapping System deployment in the public LISP Beta Network from two standpoints: Stability and Consistency. Our measurements show that the mapping information is stable over time and consistent between the different mapping entities and the vantage points. Our analysis shows that there are cases where the Mapping System is unstable and/or inconsistent, hence, beside proposing a taxonomy in order to classify them, we carry out an in-depth investigation of such cases so to provide hints on how to improve the performance of LISP.

11:30 Performance Evaluation of Mobile Broadband Cellular Networks in Pakistan

Tayyab Arshad (National University of Sciences and Technology, Pakistan); M Faheem Awan and Tahir Ahmad (National University of Science and Technology, Pakistan); Saad B. Qaisar (School of Electrical Engineering and Computer Science (SEECS), NUST & National University of Sciences & Technology, Pakistan) Pakistan is a rapidly growing cellular market with a growth rate of 185 % in number of mobile broadband users during 2014-16 and over 15.75 million mobile broadband users.. The exponential increase in broadband penetration demands investigation of mobile broadband performance from an end user perspective. To the best of our knowledge, no extensive study exists to document mobile broadband services experience in Pakistan except a recent Pakistan Telecommunication Authority (PTA) QoS report [8]. This deficiency of evidence is distressing because benchmarking Mobile Wireless broadband performance is primarily important for customers in Pakistan as well as for mobile operators. This will help both mobile operators and regulators to improve and maintain an acceptable level of quality of service. Thus, we conducted a pilot study to benchmark and analyze quality of user experience of mobile wireless users by deployment of an Android application for mobile customers to measure performance of each cellular broadband provider (Ufone, Telenor, Zong, Warid and Mobilink) in Pakistan. We analyzed performance attained at the end user device, quality of service that users are getting and identified performance bottlenecks and possible reasons. On basis of these results, we provide insights to improve mobile broadband internet services (3G and LTE) in Pakistan. Among our list of mobile operators, performance of Zong is the best in 3G networks followed by Mobilink. We consider throughput and latency as core characterizing parameters. Zong 3G has higher sustained throughput, better latency to international servers but in other factors like average signal strength, jitter and last mile latency, Mobilink 3G is superior than Zong 3G. Zong 4G performs better in overall results whereas Warid performs better in last mile latency and jitter. Our results reinforce Pakistan Telecommunication Authority(PTA), QoS survey of 3G/4G for the year 2015.

12:00 Operating System Classification Performance of TCP/IP Protocol Headers

Ahmet Aksoy and Mehmet Hadi Gunes (University of Nevada, Reno, USA) Identification of operating systems in a local network is an issue for both network management and security. Network practitioners rely on some classifier tools, but those tools rules are generated by an expert. Hence existing approaches need to be manually updated for each new operating system. In this paper, we analyze the TCP/IP packet headers to automate operating system classification. To this end, we measure the classification performance of each protocol and determine the unique features between operating systems. We utilize a genetic algorithm to determine the relevant packet header features. Then, we use several machine learning algorithms to generate set of rules that can differentiate operating systems. Overall, with IP, ICMP, TCP, UDP, HTTP, DNS, SSL, SSH, and FTP, protocol header information, on average, operating system classification can be performed at a rate of 68.0%, 51.6%, 98.4%, 71.1%, 78.7%, 29.2%, 25.0%, 22.5%, and 14.0%, respectively. In general, feature extraction with genetic algorithm further improves the results, e.g. to an average of 99.1% for TCP.

12:30 PM - 1:30 PM

Lunch Break

Keynote Speeches

1:30 PM - 2:30 PM

Keynote Speaker: Niklas Carlsson, Linköping University, Sweden

Title: "YouTube Popularity Dynamics and Third-party Authentication"

Abstract: This talk will present some of our measurement-based work on video popularity dynamics and third-party authentication. Video dissemination through sites such as YouTube can have widespread impacts on opinions, thoughts, and cultures. Not all videos will reach the same popularity and have the same impact. In this talk, we first present some of our work on modeling video popularity dynamics and assess, both qualitatively and quantitatively, the impacts of various content-agnostic factors on video popularity. Second, we present an analysis of the third-party authentication landscape, including a comparison with the third-party content delivery landscape, a longitudinal study of current trends, as well as an evaluation of cross-site information leakages and privacy risks. If time, we may also discuss current personalization trends, and briefly introduce our novel design and implementation of an interactive branched video player solution that allows users to select their own paths through a branched multi-path video, while ensuring seamless playback even when the users defer their branch path choices to the last possible moment.

Niklas Carlsson

Speaker Bio: Niklas Carlsson is an Associate Professor at Linköping University, Sweden. He received his M.Sc. degree in Engineering Physics from Umeå University, Sweden, and his Ph.D. in Computer Science from the University of Saskatchewan, Canada. He has previously worked as a Postdoctoral Fellow at the University of Saskatchewan, Canada, and as a Research Associate at the University of Calgary, Canada. His research interests are in the areas of design, modeling, characterization, and performance evaluation of distributed systems and networks. He actively serves on international program committees and publishes research papers in leading conferences. Current community involvement includes organizing ACM GreenMetrics (since 2009), being TCP co-chair of IEEE MASCOTS 2015, being an associate editor for IEEE Transactions on Sustainable Computing, being the current chair of IEEE STC on Sustainable Computing, and the acting secretary-treasurer of ACM SIGMETRICS.