IMDEA Networks Institute Digital Repository
Not a member yet
1915 research outputs found
Sort by
Stochastic Networks with Multipath Flow Control: Impact of Resource Pools on Flow-Level Performance and Network Congestion
Get PDFMultipath flow control has been proposed as a key way to
improve the Internet’s performance, reliability, and flexibility in supporting changing loads. Yet, at this point, there are
very few tools to quantify the performance benefits; particularly in the context of a stochastic network supporting best
effort flows, e.g., file transfers and web browsing sessions,
where the metric of interest is transfer delay. This paper’s
focus is on developing analysis tools to evaluate flow-level
performance and to support network design when multipath
bandwidth allocation is based on proportional fairness. To
overcome the analytical intractability of such systems we
study closely related multipath approximations based on insensitive allocations such as balanced fairness. We obtain
flow-level performance bounds on the mean per bit delay,
exhibiting the role of resource pooling in the network, and
use these to explore scenarios where increased path diversity need not result in high gains. While insightful these
results are difficult to use to drive network design and capacity allocation. To that end, we study the large deviations for congestion events, i.e., accumulation of flows, in
networks supporting multipath flow control. We show that
such asymptotics are determined by certain critical resource
pools, and study the sensitivity of congestion asymptotics
to the pool’s capacity and traffic loads. This suggests a disciplined approach to a capacity allocation problem in multipath networks based on a linear optimization problem.TRUEpu
B-Neck: a distributed and quiescent max-min fair algorithm
Get PDFThe problem of fairly distributing a network capacity among a set of sessions has been widely studied. In this problem, each session connects via a single path a source and a destination, and its objective is to maximize its as- signed transmission rate (i.e., its throughput). Since the links of the network have limited bandwidth, some form of criterion has to be defined to fairly distribute them among the sessions. A popular criterion is max-min fairness that, in short, guarantees that each session i gets a rate λi such that no session s can increase λs without causing another session s′ to end up with a rate λs′< λs. Many max-min fair algorithms have been proposed, both centralized and distributed. However, to our knowledge, all proposed distributed algorithms require control packets being continuously transmitted to recompute the max-min fair rates when needed.
In this paper we propose B-Neck, a max-min fair distributed algorithm that is also quiescent. This means that, in absence of changes (i.e., session arrivals or departures), once the max-min rates have been computed B-Neck stops generating network traffic. As far as we know, B-Neck is the first max-min fair distributed algorithm that does not require a continuous injection of control traffic to compute the rates. When changes occur, affected sessions are asynchronously informed of their new rate (i.e., sessions do not need to poll the network for changes). The correctness of B-Neck is formally proved, and extensive simulations are conducted. In them it is shown that B-Neck converges relatively fast and behaves nicely in presence of sessions arriving and departing.Institute IMDEA Networkspu
A Quantize-and-Forward Scheme for Future Wireless Relay Networks
Get PDFThe orthogonal multiple-access relay channel with two sources is considered. The goal of this paper is to show the applicability and effectiveness of a previously introduced
quantize-and-forward scheme to a more realistic channel and
system model, including orthogonal frequency division multiple
access and multipath fading channels. Simulation results are provided to demonstrate the gain of quantize-and-forward relayed communication as opposed to the point-to-point links without the relay.TRUEpu
Graceful BGP session shutdown
Get PDFIn the Operations and Management area, a «graceful shutdown» mechanism for BGP peering links is being standardized in collaboration with France Telecom -Orange, Internet Initiative Japan, and Cisco Systems. The draft capturing this work («Graceful BGP session shutdown», draft-ietf-grow-bgp-gshut-03), working group document of the GROW Working Group, describes operational procedures aimed at reducing the amount of traffic lost during planned maintenances of routers or links, involving the shutdown of BGP peering sessions. It also provides recommendations to router vendors for the support of a graceful shutdown mechanism that eases operational aspects of the solution. The authors are: Pierre Francois, Bruno Decraene, Cristel Pelsser, Keyur Patel and Clarence Filsfils.This draft describes operational procedures aimed at reducing the amount of traffic lost during planned maintenances of routers or links, involving the shutdown of BGP peering sessions.Internet Engineering Task Force (IETF)pu
Quality of service implications of power control and multiuser detection based cross-layer design
Get PDFIn order to allow for dense spatial reuse in wireless
ad hoc networks, multiple access interference must be dealt
with. This calls for advanced physical layer techniques, such as multiuser detection or power control. However, these techniques can only be efficiently applied to ad hoc networks when they are part of a joint PHY/MAC cross-layer design. In order to better understand both, the potential but also the limits of handling interference by means of multiuser detection and power control, respectively, in this paper we provide a comprehensive comparison between multiuser detection based and power control based cross layer designs. We study the behavior of both approaches in
terms of throughput, delay, as well as fairness in scenarios with high and low user densities, respectively. To provide more detailed insight in the interaction between MAC and PHY, we separate for each approach the throughput results into gains achieved solely by the MAC layer and by the PHY layer, respectively. These results highlight, among other aspects, some fundamental disadvantage of power control in distributed environments. We conclude that multiuser based approaches are significantly more beneficial in ad hoc scenarios than power control based schemes.TRUEpu
Overhearing-Assisted Optimization of Address AutoConfiguration in Position Aware-VANETs
Get PDFVehicular networks allow vehicles to exchange information
that can be used to improve traffic efficiency and drivers’
safety. In addition to these new applications, Internet connectivity is also expected to be available in the cars of the near future, speeding up the global adoption of vehicular communication systems. One of the requirements for connecting vehicles to the Internet is their ability to auto-configure IP addresses. In this article, we propose an optimization to Geographically Scoped stateless Address Configuration (GeoSAC), which is an IP address
auto-configuration mechanism for geographically aware location vehicles. The benefits of this optimization are two-fold: it can reduce the IP address configuration time, and it can be used to reduce the signaling overhead of GeoSAC. The optimization does not require any changes to the GeoSAC operation, being fully compatible with the original solution. We derive an analytical model for the probability of our optimization being effective in realistic scenarios and for the IP address configuration time. We also provide a thorough evaluation of the performance improvements of the optimization, including simulations with a realistic model for wireless technology, real vehicular traces, and experiments with a real prototype, which provide strong support for our analytical model.TRUEpu
No full textInternet traffic has increased steeply in recent years, mainly due to the fruition of video and other streaming contents, social platforms and peer-to-peer networks. In addition, the quick penetration of hand-held devices equipped with multiple radios (e.g., 3G and WiFi), sees to it that wireless access represents an ever-growing portion of current and future demand, thus encouraging operators to investigate and deploy different combinations of wireless access technologies with the purpose of educing their operational costs (the so-called “4G” architecture).It is hence necessary to adopt an efficient mobility management technique to meet users’ expectation of an “anywhere, anytime” connectivity. Nevertheless, the use of centralized mobility management approaches – such as Mobile IPv6 and Proxy Mobile IPv6 – is foreseen to bring some difficulties to operators, due to the expected large number of mobile users and their exigent demands. All this has triggered the need for Distributed Mobility Management (DMM) alternatives,
focused on moving the mobility anchors from the core network to the edge, pushing them closer to the users. The purpose of such new research direction is to overcome the
limitations imposed by a centralized approach, alleviating operators’ costs by deploying a more efficient network, envisioning also the heterogeneity of the underlying technology. This work first explores two protocols for mobility support, Mobile IPv6 and Proxy Mobile IPv6, taken as main referents for, respectively, the host-based mobility approach and the network based one. We next elaborate the extensions and the changes to transform them
according to the distributed mobility management paradigm, proposing several complete solutions. Finally, we analytically compare the distributed solutions to their centralized counterparts, in order to derive which are the most suitable scenarios for their applicability.Telematics EngineeringUniversidad Carlos III de Madrid, Spainpu
Scalable max–min fairness in wireless ad hoc networks
No full textOur previous work proposes a macro model to perform flow and access control in wireless ad hoc networks. In this paper, we demonstrate specifically how to apply the model to achieve max-min fair rate allocation. Our proposed scheme is simple and scalable when comparing to other techniques in the literature. Moreover, it has the ability to provide stability in mobile environment. Simulation results show that our new method provides a good max-min fair flow assignment, and with that assignment, quality of service guarantees can be achieved for real-time applications.TRUEpu