arXiv subjects

Vania Conan

Publications and source records attributed to Vania Conan.

17 recordsLinked to original sources

arXiv2018

Distributed Function Chaining with Anycast Routing

Current networks more and more rely on virtualized middleboxes to flexibly provide security, protocol optimization, and policy compliance functionalities. As such, delivering these services requires that the traffic be steered through the desired sequence of virtual appliances. Current solutions introduce a new logically centralized entity, often called orchestrator, needing to build its own holistic view of the whole network so to decide where to direct the traffic. We advocate that such a centralized orchestration is not necessary and that, on the contrary, the same objectives can be achieved by augmenting the network layer routing so to include the notion of service and its chaining. In this paper, we support our claim by designing such a system. We also present an implementation and an early evaluation, showing that we can easily steer traffic through available resources. This approach also offers promising features such as incremental deployability, multi-domain service chaining, failure resiliency, and easy maintenance.

cs.NI↗

arXiv2016

MUREN: delivering edge services in joint SDN-SDR multi-radio nodes

To meet the growing local and distributed computing needs, the cloud is now descending to the network edge and sometimes to user equipments. This approach aims at distributing computing, data processing, and networking services closer to the end users. Instead of concentrating data and computation in a small number of large clouds, many edge systems are envisioned to be deployed close to the end users or where computing and intelligent networking can best meet user needs. In this paper, we go further converging such massively distributed computing systems with multiple radio accesses. We propose an architecture called MUREN (Multi-Radio Edge Node) for managing traffic in future mobile edge networks. Our solution is based on the Mobile Edge Cloud (MEC) architecture and its close interaction with Software Defined Networking (SDN), the whole jointly interacting with Software-Defined Radios (SDR). We have implemented our architecture in a proof of concept and tested it with two edge scenarios. Our experiments show that centralizing the intelligence in the MEC allows to guarantee the requirements of the edge services either by adapting the waveform parameters, or through changing the radio interface or even by reconfiguring the applications. More generally, the best decision can be seen as the optimal reaction to the wireless links variations

cs.NI↗

arXiv2012

Temporal Reachability Graphs

While a natural fit for modeling and understanding mobile networks, time-varying graphs remain poorly understood. Indeed, many of the usual concepts of static graphs have no obvious counterpart in time-varying ones. In this paper, we introduce the notion of temporal reachability graphs. A (tau,delta)-reachability graph} is a time-varying directed graph derived from an existing connectivity graph. An edge exists from one node to another in the reachability graph at time t if there exists a journey (i.e., a spatiotemporal path) in the connectivity graph from the first node to the second, leaving after t, with a positive edge traversal time tau, and arriving within a maximum delay delta. We make three contributions. First, we develop the theoretical framework around temporal reachability graphs. Second, we harness our theoretical findings to propose an algorithm for their efficient computation. Finally, we demonstrate the analytic power of the temporal reachability graph concept by applying it to synthetic and real-life datasets. On top of defining clear upper bounds on communication capabilities, reachability graphs highlight asymmetric communication opportunities and offloading potential.

cs.NI↗

arXiv2012

Push-and-Track: Saving Infrastructure Bandwidth Through Opportunistic Forwarding

Major wireless operators are nowadays facing network capacity issues in striving to meet the growing demands of mobile users. At the same time, 3G-enabled devices increasingly benefit from ad hoc radio connectivity (e.g., Wi-Fi). In this context of hybrid connectivity, we propose Push-and-track, a content dissemina- tion framework that harnesses ad hoc communication opportunities to minimize the load on the wireless infrastructure while guaranteeing tight delivery delays. It achieves this through a control loop that collects user-sent acknowledgements to determine if new copies need to be reinjected into the network through the 3G interface. Push-and-Track is flexible and can be applied to a variety of scenarios, including periodic message flooding and floating data. For the former, this paper examines multiple strategies to determine how many copies of the content should be injected, when, and to whom; for the latter, it examines the achievable offload ratio depending on the freshness constraints. The short delay-tolerance of common content, such as news or road traffic updates, make them suitable for such a system. Use cases with a long delay-tolerance, such as software updates, are an even better fit. Based on a realistic large-scale vehicular dataset from the city of Bologna composed of more than 10,000 vehicles, we demonstrate that Push-and-Track consistently meets its delivery objectives while reducing the use of the 3G network by about 90%.

cs.NI↗

arXiv2011

Using Neighborhood Beyond One Hop in Disruption-Tolerant Networks

Most disruption-tolerant networking (DTN) protocols available in the literature have focused on mere contact and intercontact characteristics to make forwarding decisions. Nevertheless, there is a world behind contacts: just because one node is not in contact with some potential destination, it does not mean that this node is alone. There may be interesting end-to-end transmission opportunities through other nearby nodes. Existing protocols miss such possibilities by maintaining a simple contact-based view of the network. In this paper, we investigate how the vicinity of a node evolves through time and whether such information can be useful when routing data. We observe a clear tradeoff between routing performance and the cost for monitoring the neighborhood. Our analyses suggest that limiting a node's neighborhood view to three or four hops is more than enough to significantly improve forwarding efficiency without incurring prohibitive overhead.

cs.NI↗

arXiv2010

From Encounters to Plausible Mobility

Inferring plausible node mobility based only on information from wireless contact traces is a difficult problem. Working with mobility information allows richer protocol simulations, particularly in dense networks, but requires complex set-ups to measure. On the other hand, contact information is easier to measure but only allows for simplistic simulation models. In a contact trace a lot of node movement information is irretrievably lost so the original positions and velocities are in general out of reach. In this paper, we propose a fast heuristic algorithm, inspired by dynamic force-based graph drawing, capable of inferring a plausible movement from any contact trace, and evaluate it on both synthetic and real-life contact traces. Our results reveal that (i) the quality of the inferred mobility is directly linked to the precision of the measured contact trace, and (ii) the simple addition of appropriate anticipation forces between nodes leads to an accurate inferred mobility.

cs.NI↗

arXiv2010

Relieving the Wireless Infrastructure: When Opportunistic Networks Meet Guaranteed Delays

Major wireless operators are nowadays facing network capacity issues in striving to meet the growing demands of mobile users. At the same time, 3G-enabled devices increasingly benefit from ad hoc radio connectivity (e.g., Wi-Fi). In this context of hybrid connectivity, we propose Push-and-track, a content dissemination framework that harnesses ad hoc communication opportunities to minimize the load on the wireless infrastructure while guaranteeing tight delivery delays. It achieves this through a control loop that collects user-sent acknowledgements to determine if new copies need to be reinjected into the network through the 3G interface. Push-and-Track includes multiple strategies to determine how many copies of the content should be injected, when, and to whom. The short delay-tolerance of common content, such as news or road traffic updates, make them suitable for such a system. Based on a realistic large-scale vehicular dataset from the city of Bologna composed of more than 10,000 vehicles, we demonstrate that Push-and-Track consistently meets its delivery objectives while reducing the use of the 3G network by over 90%.

cs.NI↗

arXiv2010

HYMAD: Hybrid DTN-MANET Routing for Dense and Highly Dynamic Wireless Networks

Delay/Disruption-Tolerant Network (DTN) protocols typically address sparse intermittently connected networks whereas Mobile Ad-hoc Network (MANET) protocols address the fairly stable and fully connected ones. But many intermediate situations may occur on mobility dynamics or radio link instability. In such cases, where the network frequently splits into evolving connected groups, none of the conventional routing paradigms (DTN or MANET) are fully satisfactory. In this paper we propose HYMAD, a Hybrid DTN-MANET routing protocol which uses DTN between disjoint groups of nodes while using MANET routing within these groups. HYMAD is fully decentralized and only makes use of topological information exchanges between the nodes. The strength of HYMAD lies in its ability to adapt to the changing connectivity patterns of the network. We evaluate the scheme in simulation by replaying synthetic and real life mobility traces which exhibit a broad range of connectivity dynamics. The results show that HYMAD introduces limited overhead and outperforms the multi-copy Spray-and-Wait DTN routing protocol it extends, both in terms of delivery ratio and delay. This hybrid DTN-MANET approach offers a promising venue for the delivery of elastic data in mobile ad-hoc networks as it retains the resilience of a \textit{pure} DTN protocol while significantly improving performance.

cs.NI↗

arXiv2010

Tuning Message Size in Opportunistic Mobile Networks

We describe a new model for studying intermittently connected mobile networks, based on Markovian random temporal graphs, that captures the influence of message size, maximum tolerated delay and link stability on the delivery ratio.

cs.NI↗

arXiv2010

Dimensionnement des messages dans un reseau mobile opportuniste

Understanding transport capacity in intermittently connected mobile networks (ICMN) is crucial since different applications have different interactivity and bandwidth requirements. One practical issue is how to transform an application's messages into packets suitable for transport over an ICMN. In this paper, we propose a new Markovian model for random temporal graphs and show, both analytically and by replaying a real life trace obtained in a rollerblading tour, that the size of the messages sent over an ICMN has a decisive impact on their delivery ratio. A given message could therefore be broken down into smaller packets to increase reliability. However, we also show that this gain in reliability only appears under tight constraints on the maximum delay tolerated. Mobile application designers should therefore balance message size against both application requirements and network topology dynamics to improve performance.

cs.NI↗

arXiv2010

Plausible Mobility: Inferring Movement from Contacts

We address the difficult question of inferring plausible node mobility based only on information from wireless contact traces. Working with mobility information allows richer protocol simulations, particularly in dense networks, but requires complex set-ups to measure, whereas contact information is easier to measure but only allows for simplistic simulation models. In a contact trace a lot of node movement information is irretrievably lost so the original positions and velocities are in general out of reach. We propose a fast heuristic algorithm, inspired by dynamic force-based graph drawing, capable of inferring a plausible movement from any contact trace, and evaluate it on both synthetic and real-life contact traces. Our results reveal that (i) the quality of the inferred mobility is directly linked to the precision of the measured contact trace, and (ii) the simple addition of appropriate anticipation forces between nodes leads to an accurate inferred mobility.

cs.NI↗

arXiv2010

HYMAD: Hybrid DTN-MANET Routing for Dense and Highly Dynamic Wireless Networks

In this paper we propose HYMAD, a Hybrid DTN-MANET routing protocol which uses DTN between disjoint groups of nodes while using MANET routing within these groups. HYMAD is fully decentralized and only makes use of topological information exchanges between the nodes. We evaluate the scheme in simulation by replaying real life traces which exhibit this highly dynamic connectivity. The results show that HYMAD outperforms the multi-copy Spray-and-Wait DTN routing protocol it extends, both in terms of delivery ratio and delay, for any number of message copies. Our conclusion is that such a Hybrid DTN-MANET approach offers a promising venue for the delivery of elastic data in mobile ad-hoc networks as it retains the resilience of a pure DTN protocol while significantly improving performance.

cs.NI↗

arXiv2009

Performance of Opportunistic Epidemic Routing on Edge-Markovian Dynamic Graphs

Connectivity patterns in intermittently-connected mobile networks (ICMN) can be modeled as edge-Markovian dynamic graphs. We propose a new model for epidemic propagation on such graphs and calculate a closed-form expression that links the best achievable delivery ratio to common ICMN parameters such as message size, maximum tolerated delay, and link lifetime. These theoretical results are compared to those obtained by replaying a real-life contact trace.

cs.NI↗

arXiv2009

Maximizing the number of accepted flows in TDMA-based wireless ad hoc networks is APX-complete

Full exploitation of the bandwidth resources of Wireless Networks is challenging because of the sharing of the radio medium among neighboring nodes. Practical algorithms and distributed schemes that tries to optimising the use of the network radio resources. In this technical report we present the proof that maximising the network capacity is is an APX Complete problem (not approximable within 1/(1 - 2^(-k)) - eps for eps > 0).

cs.NI↗

arXiv2007

The heterogeneity of inter-contact time distributions: its importance for routing in delay tolerant networks

Prior work on routing in delay tolerant networks (DTNs) has commonly made the assumption that each pair of nodes shares the same inter-contact time distribution as every other pair. The main argument in this paper is that researchers should also be looking at heterogeneous inter-contact time distributions. We demonstrate the presence of such heterogeneity in the often-used Dartmouth Wi-Fi data set. We also show that DTN routing can benefit from knowing these distributions. We first introduce a new stochastic model focusing on the inter-contact time distributions between all pairs of nodes, which we validate on real connectivity patterns. We then analytically derive the mean delivery time for a bundle of information traversing the network for simple single copy routing schemes. The purpose is to examine the theoretic impact of heterogeneous inter-contact time distributions. Finally, we show that we can exploit this user diversity to improve routing performance.

cs.NI↗

arXiv2005

Evaluating Mobility Pattern Space Routing for DTNs

Because a delay tolerant network (DTN) can often be partitioned, the problem of routing is very challenging. However, routing benefits considerably if one can take advantage of knowledge concerning node mobility. This paper addresses this problem with a generic algorithm based on the use of a high-dimensional Euclidean space, that we call MobySpace, constructed upon nodes' mobility patterns. We provide here an analysis and the large scale evaluation of this routing scheme in the context of ambient networking by replaying real mobility traces. The specific MobySpace evaluated is based on the frequency of visit of nodes for each possible location. We show that the MobySpace can achieve good performance compared to that of the other algorithms we implemented, especially when we perform routing on the nodes that have a high connection time. We determine that the degree of homogeneity of mobility patterns of nodes has a high impact on routing. And finally, we study the ability of nodes to learn their own mobility patterns.

cs.NI↗

arXiv2005

DTN Routing in a Mobility Pattern Space

Routing in Delay Tolerant Networks (DTNs) benefits considerably if one can take advantage of knowledge concerning node mobility. The main contribution of this paper is the definition of a generic routing scheme for DTNs using a high-dimensional Euclidean space constructed upon nodes' mobility patterns. For example, nodes are represented as points having as coordinates their probability of being found in each possible location. We present simulation results indicating that such a scheme can be beneficial in a scenario inspired by studies done on real mobility traces. This work should open the way to further use of the virtual space formalism in DTN routing.

cs.NI↗