IMDEA Networks Institute Digital Repository
Not a member yet
1915 research outputs found
Sort by
Video over mobile networks [Guest editorial]
No full textVideo is a major challenge for the future mobile Internet as it is foreseen to account for close to 70 percent of consumer mobile traffic by 2016. Already, mobile network operators are suffering from the exploding number of subscribers accessing high-volume mobile data services, and this trend is expected to intensify even further in the near future. The enormous proliferation of multimedia-capable portable devices ¿ such as smart phones, tablets, and laptops equipped with 3G and WLAN interfaces ¿ has increased the volume and variety of multimedia digital content and related services.pu
A Game Theoretic Approach to Distributed Opportunistic Scheduling
Get PDFDistributed Opportunistic Scheduling (DOS) is inherently harder than conventional opportunistic scheduling due to the absence of a central entity that has knowledge of all the channel states. With DOS, stations contend for the channel using random access; after a successful contention, they measure the channel conditions and only transmit in case of a good channel, while giving up the transmission opportunity when the channel conditions are poor. The distributed nature of DOS systems makes them vulnerable to selfish users: by deviating from the protocol and using more transmission opportunities, a selfish user can gain a greater share of the wireless resources at the expense of the well-behaved users. In this paper, we address the selfishness problem in DOS from a game theoretic standpoint. We propose an algorithm that satisfies the following properties: (i) when all stations implement the algorithm, the wireless network is driven to the optimal point of operation, and (ii) one or more selfish stations cannot gain any profit by deviating from the algorithm. The key idea of the algorithm is to react to a selfish station by using a more aggressive configuration that (indirectly) punishes this station. We build on multivariable control theory to design a mechanism for punishment that on the one hand is sufficiently severe to prevent selfish behavior while on the other hand is light enough to guarantee that, in the absence of selfish behavior, the system is stable and converges to the optimum point of operation. We conduct a game theoretic analysis based on repeated games to show the algorithm's effectiveness against selfish stations. These results are confirmed by extensive simulations.pu
Experimental analysis of the socio-economic phenomena in the BitTorrent ecosystem
Get PDFBitTorrent is the most successful Peer-to-Peer (P2P) application and is responsible for a major portion of Internet traffic. It has been largely studied using simulations, models and real measurements. Although simulations and modelling are easier to perform, they typically simplify analysed problems and in case of BitTorrent they are likely to miss some of the effects which occur in real swarms. Thus, in this thesis we rely on real measurements. In the first part of the thesis we present the summary of measurement techniques used so far and we use it as a base to design our tools that allow us to perform different types of analysis at different resolution level.
Using these tools we collect several large-scale datasets to study different aspects of BitTorrent with a special focus on socio-economic aspects. Using our datasets, we first investigate the topology of real BitTorrent swarms and how the traffic is actually exchanged among peers. Our analysis shows that the resilience of BitTorrent swarms is lower than corresponding random graphs. We also observe that ISP policies, locality-aware clients and network events (e.g., network congestion) lead to locality-biased composition of neighbourhood in the swarms. This means that
the peer contains more neighbours from local provider than expected from purely random neighbours selection process. Those results are of interest to the companies which use BitTorrent for daily operations as well as for ISPs which carry BitTorrent traffic. In the next part of the thesis we look at the BitTorrent from the perspective of the content and content publishers in a major BitTorrent portals. We focus on the factors that seem to drive the popularity of the BitTorrent and, as a result, could affect its associated traffic in the Internet. We show that a small fraction of publishers (around 100 users) is responsible for more than two-thirds of the published content. Those publishers can be divided into two groups: (i) profit driven and (ii) fake publishers. The former group leverages the published copyrighted content (typically very popular) on BitTorrent portals to attract content consumers to their web sites for financial gain. Removing this group may have a significant impact on the popularity of BitTorrent portals and, as a result, may affect a big portion of the Internet traffic associated to BitTorrent. The latter group is responsible for fake content, which is mostly linked to malicious activity and creates a serious threat for the Bit-
Torrent ecosystem and for the Internet in general. To mitigate this threat, in the last part of the thesis we present a new tool named TorrentGuard for the early detec-
tion of fake content that could help to significantly reduce the number of computer infections and scams suffered by BitTorrent users. This tool is available through
web portal and as a plugin to Vuze, a popular BitTorrent client. Finally, we present MYPROBE, the web portal that allows to query our database and to gather different
pieces of information regarding BitTorrent content publishers.Telematics EngineeringUniversidad Carlos III de Madrid, Spainpu
Automatic failure recovery for software-defined networks
Get PDFTolerating and recovering from link and switch failures are fundamental requirements of most networks, including Software-Defined Networks (SDNs). However, instead of traditional behaviors such as network-wide routing re-convergence, failure recovery in an SDN is determined by the specific software logic running at the controller. While this admits more freedom to respond to a failure event, it ultimately means that each controller application must include its own recovery logic, which makes the code more difficult to write and potentially more error-prone.
In this paper, we propose a runtime system that automates failure recovery and enables network developers to write simpler, failure-agnostic code. To this end, upon detecting a failure, our approach first spawns a new controller instance that runs in an emulated environment consisting of the network topology excluding the failed elements. Then, it quickly replays inputs observed by the controller before the failure occurred, leading the emulated network into the forwarding state that accounts for the failed elements. Finally, it recovers the network by installing the difference ruleset between emulated and current forwarding states.TRUEpu
Simple Approximate Analysis of Floating Content for Context-Aware Applications
Get PDFContext-awareness is a peculiar characteristic of an ever
expanding set of applications that make use of a combination
of restricted spatio-temporal locality and mobile communications, to deliver a variety of services to the end user. Communication requirements for context-aware applications significantly differ from those of ordinary applications; opportunistic communications are extremely well-suited to them, because they naturally incorporate context. Recently, an opportunistic communication paradigm called "Floating Content" was proposed, which is conceived to support serverless, distributed content sharing. In this work, we present a simple (in that it uses few primitive system parameters), approximate analytical model for the performance analysis of context-aware applications that use floating content. From a system design perspective, our analysis can be used to tune key system parameters so as to achieve the desired application performance. In particular, we apply our analysis to estimate the "success probability" for two representative categories of context-aware applications, and show how the system can be configured to achieve the application’s target. In order to complement our analytical study, we validate our model using extensive simulations under different settings and mobility patterns. Our simulation results show that our model-based predictions are indeed highly accurate under a wide range of conditions.pu
Incorporating Rate Adaptation into Green Networking for Future Data Centers
Get PDFDespite some proposals for energy-efficient topologies, most of the studies for saving energy in data center networks are focused on traffic engineering, i.e., consolidating flows and switching off unnecessary network devices. The major weakness of this approach is network oscillation brought by the frequent change of network topology when traffic fluctuates very fast. In this paper, we propose to incorporate rate adaptation into green data center networks. With rate adaptive network devices, we aim at approaching network-wide energy proportionality by routing optimization. We formalize the problem with an integer program and propose an efficient approximation algorithm –TSRR, solving the problem quickly while guaranteeing a constant performance ratio. Extensive range of simulations confirm that more than 40% of the energy can be saved while introducing very slight stretch on network delay.TRUEpu
Mobility Management in Next Generation Mobile Networks
Get PDFIn this demo we propose a novel architecture suitable for deployment in future mobile networks based on the distributed mobility management (DMM) paradigm. In DMM, the IP anchoring point gets closer to users, with the aim to flatten the architecture and to truly enable the fixed/mobile convergence. DMM allows operators to tackle the explosion of data traffic without burdening the core part of the network, offering a common IP framework for mobility and heterogeneous access. As a use case scenario, this demo has shown a possible deployment for a content delivery network's nodes, in order to exploit the DMM benefits.TRUEpu
A Hop-by-hop Energy Efficient Distributed Routing Scheme
Get PDFThis publication was presented at the Greenmetrics 2013 Workshop (17 June 2013), hosted in Pittsburgh, Pennsylvania, USA.Energy inefficiencies in current networks provide both challenges and opportunities for energy saving. Recently, there are many works focusing on minimizing energy cost from the routing perspective. However, most existing work view them as optimization problems and solve them in a centralized manner such as with a solver or using approximations.
In this paper, we focus on a network-wide bi-objective optimization problem, which simultaneously minimizes the total energy consumption and the total traffic delay using speed scaling. We propose a hop-by-hop dynamic distributed routing scheme for the implementation of this network-wide optimization problem. Our scheme is more practical to realize in large distributed networks compared with current centralized energy minimization methods. We can also achieve near global optimal in a distributed manner, while mostly used shortest path routing protocols such as OSPF cannot make it. Our routing scheme is totally distributed and maintains loop-free during every instant of routing. Simulations conducted in real world data sets show that the distributed loop-free routing scheme converges to the near Pareto optimal values. Also, our method outperforms the widely applied shortest path routing strategy with 30% energy saving.TRUEpu
Insights of YouTube View Check System
Get PDFYouTube is the most popular website for videos nowadays. Its business model is based on advertisement: ads are introduced in the videos and uploaders are paid based on the number of views of their videos. For this reason, it is imporant for YouTube to be able to check that views are real, and discard those that are faked.
The main purpose of this paper is to understand the mech-
anisms implemented by YouTube to detect and discard faked
views. To this aim, we have developed a robot that performs
faked views; by building on a thorough study of the patterns and the most common traffic sources of videos, and emulating this patterns, our robot shows a very similar behavior to real users. From the results obtained from our experiments, we observe that: (i) we are able to overcome the 301 limit, which corresponds to the first thorough analysis YouTube performs to detect faked views, (ii) we are capable of pushing the total number of views to an unlimited large number, but at a limited growth rate, (iii)
YouTube appears to be insensitive to many behaviors that show that views are faked, except for the IP address used by the client, and (iv) regardless of how sophisticated the robot is, YouTube only counts a fraction of the views realized (around one half).
In order to compare the behavior observed by our robot to
other behaviours, we have considered (i) an experiment with real people watching the video, and (ii) services offered in the Internet that sell views for YouTube videos. We have observed from the first experiment that even with real views, YouTube only seems to be counting about half of them as real views. Furthermore, the second experiment confirms that even some basic tools are able to increase the number of views (e.g., in some cases we found that the average view duration was 0 but still for those cases YouTube counted a large number of views).
The experiments conducted seem to show that YouTube
implements a very basic scheme to detect fake views that is
easily tricked by simple tools, and that YouTube discards a
subsantial number of the views performed even for real views. Such results seem to point to the need for revisting the algorithms implemented by YouTube to detect fake views.Telematics EngineeringUniversidad Carlos III de Madrid, Spainpu