IMDEA Networks Institute Digital Repository
Not a member yet
1915 research outputs found
Sort by
ABridges: Scalable, self-configuring Ethernet campus networks
Get PDFThis article describes a scalable, self-configuring architecture for campus networks, the ABridges architecture. It is a two-tiered hierarchy of layer two switches in which network islands running independent rapid spanning tree protocols communicate through a core formed by island root bridges (ABridges). ABridges use AMSTP, a simplified and self configuring version of MSTP protocol, to establish shortest paths in the core using multiple spanning tree instances, one instance rooted at each core edge ABridge. The architecture is very efficient in terms of network usage and path length due to the ability of AMSTP to provide optimum paths in the core mesh, while RSTP is used to aggregate efficiently the traffic at islands networks, where sparsely connected, tree-like topologies are frequent and recommended. Convergence speed is as fast as existing Rapid Spanning Tree and Multiple Spanning Tree Protocols.TRUEpu
Enabling Global Multimedia Distributed Services based on Hierarchical DHT Overlay Networks
Get PDFThe provision of innovating multimedia services is a high
priority for service providers. Due to the the high traffic
volume characteristics of multimedia content, decentralised
services are a desired solution for multimedia scenarios.
Based on the ongoing work of the IETF P2PSIP WG this paper defines a simple way to interconnect different domains using on peer-to-peer technology. The objective is to provide connectivity between users of different domains and allow decentralised multimedia services to be placed where they can be provided most effectively. Furthermore, the routing performance and routing state is analysed for the proposed hierarchical DHT overlay architecture.TRUEpu
Hierarchical Up/Down Routing Architecture for Ethernet backbones and campus networks
Get PDFWe describe a new layer two distributed and
scalable routing architecture. It uses an automatic hierarchical
node identifier assignment mechanism associated to the rapid
spanning tree protocol. Enhanced Up/Down mechanisms are
used to prohibit some turns at nodes to break cycles, instead of
blocking links like the spannning tree protocol does. The
protocol performance is similar or better than other Turn
Prohibition algorithms recently proposed with lower
complexity O (Nd) and better scalability. Simulations show
that the fraction of prohibited turns over random networks is
less than 0.2. The effect of root bridge election on the
performance of the protocol is limited both in the random and
regular networks studied.
The use of hierarchical, tree-descriptive addresses simplifies
the routing. and avoids the need of all nodes having a global
knowleddge of the network topology. Routing frames through
the hierarchical tree at very high speed is possible by
progressive decoding of frame destination address, without
routing tables or port address learning. Coexistence with
standard bridges is achieved using combined devices: bridges
that forward the frames having global destination MAC
addresses as standard bridges and frames with local MAC
frames with the proposed protocol.TRUEpu
Enabling Global Multimedia Distributed Services based on Hierarchical DHT Overlay Networks
Get PDFThe provision of innovating multimedia services is a high
priority for service providers. Due to the the high traffic
volume characteristics of multimedia content, decentralised
services are a desired solution for multimedia scenarios.
Based on the ongoing work of the IETF P2PSIP WG
this paper defines a simple way to interconnect different domains using on peer-to-peer technology. The objective is
to provide connectivity between users of different domains
and allow decentralised multimedia services to be placed
where they can be provided most effectively. Furthermore,
the routing performance and routing state is analysed for
the proposed hierarchical DHT overlay architecture.TRUEpu
Loop-freeness in multipath BGP through propagating the longest path
Get PDFThe concurrent use of multiple paths through a
communications network has the potential to provide many benefits, including better utilization of the network and increased robustness. A key part of a multipath network architecture is the ability for routing protocols to install multiple routes over multiple paths in the routing table. In this paper we propose changes to local BGP processing that allow a BGP router to use multiple paths concurrently without
compromising loop-freeness.Telematics EngineeringUniversidad Carlos III de Madrid, Spainpu
A Control Theoretic Approach for Throughput Optimisation in IEEE 802.11e EDCA WLANs
Get PDFThe MAC layer of the 802.11 standard, based on the CSMA/CA mechanism, specifies a set of parameters to control the aggressiveness of stations when trying to access the channel. However, these parameters are statically set independently of the conditions of the WLAN(e.g. the number of contending stations), leading to poor performance for most scenarios. To overcome this limitation previous work proposes to adapt the value of one of those parameters, namely the CW, based on an estimation of the conditions of the WLAN. However, these approaches suffer from two major drawbacks: i) they require extending the capabilities of
standard devices or ii) are based on heuristics. This thesis proposes a control theoretic approach to adapt the CW to the conditions of the WLAN, based on an analytical model of its operation, that is fully compliant with the
802.11e standard. We use a Proportional Integrator controller in order to drive the WLAN to its optimal point of operation and perform a theoretic analysis to determine its configuration. We show by means of an exhaustive performance evaluation that our algorithm maximises the total throughput of the WLAN and substantially outperforms previous standardcompliant proposals.Telematics EngineeringUniversidad Carlos III de Madrid, Spainpu
Applying a Macro Model of Ad Hoc Networks to Access Control
No full textOur previous work proposes a macro model that controls the flows across naturally occurring bottlenecks in multi-hop wireless networks. The macro model resembles a conventional wired network, and allows us to apply extant research on flow control directly to ad hoc networks. In this paper, we further justify the feasibility of applying our macro model of ad hoc networks to access control and the validity of only controlling the flows across bottlenecks. We describe a simple rule that partitions the network into super-nodes. We then introduce a probabilistic geographic routing protocol that routes packets between bottlenecks to avoid congestion within super-nodes. Simulation results show that not only is the average flow in non- bottlenecked areas less than the average flow across bottlenecks, but the flows in each of the transmission areas that is not in a bottleneck are less than the flows in the transmission areas within the bottleneck.TRUEpu
Arquitectura de Pasarela Residencial Orientada a la Autoconfiguración
Get PDFLos dispositivos de comunicación de datos mejoran sus funcionalidades día tras día. Con cada nuevo equipo, el
usuario debe aprender a configurarlos, administrarlos, cargar nuevas actualizaciones y cuando se requiere una nueva funcionalidad no soportada por dicho equipo, cambiarlo por uno nuevo. Hoy en día, configurar un modem-router xDSL es una tarea complicada para un usuario inexperto y varios proveedores ya optan por la configuración remota. Conforme estos equipos se vuelvan más complejos, esta ́ultima funcionalidad será más
y más demandada, aunque los usuarios con más experiencia
querrán tener la posibilidad de realizar una configuración local. Este artículo propone una arquitectura general de una pasarela residencial con la flexibilidad suficiente como para cargar y descargar módulos individuales que desempeñen funcionalidades muy dispares. Esta arquitectura está especialmente pensada para implantarse en una red de siguiente generación, aunque puede emplearse en cualquier tipo de red.TRUEpu
Nozzilla: A Novel Peer to Peer Architecture for Video Streaming
Get PDFMany peer-to-peer video streaming systems use application-level multicast where the peers are responsible of forwarding the video packets between them, usually following the path of one or more tree overlays. Since the performance of such systems depends on the uplink capacity available at each peer (measured as bandwidth, delay and jitter), previous work focuses on sharing the load responsibility between all peers by dividing the video
traffic into several stripes and requiring a peer to be an interior node in one multicast tree. However, although these techniques consider the uplink capacity limit, they do not adapt easily for situations where the resources are either very abundant or very scarce. This thesis proposes Nozzilla, a new approach based on structured peer-to-peer that dynamically takes into account the available capacity resources. Increased efficiency and reliability is achieved in several ways. First, the peers in the overlay are grouped according to their current resources, improving the search for a suitable parent inside the multicast tree.
Second, the system uses path diversity by splitting the video traffic in several stripes and distributes each stripe along a different multicast tree. Third, the protocol can be used in a quality-of-service enabled network, which can assign different classes of service for each video stripe. This makes possible the usage of techniques such as multi-description coding and scalable video coding and is more robust to failures since the loss of a single stripe does not cause the loss of video reception. Finally, the algorithm improves the performance of the multicast tree by increasing the load distribution between peers and reducing the resource demand on the root.Telematic EngineeringUniversidad Carlos III de Madrid, Spainpu
Hierarchical Up/Down Routing Architecture for Ethernet backbones and campus networks
Get PDFWe describe a new layer two distributed and
scalable routing architecture. It uses an automatic hierarchical
node identifier assignment mechanism associated to the rapid
spanning tree protocol. Enhanced Up/Down mechanisms are
used to prohibit some turns at nodes to break cycles, instead of
blocking links like the spannning tree protocol does. The
protocol performance is similar or better than other Turn
Prohibition algorithms recently proposed with lower
complexity O (Nd) and better scalability. Simulations show
that the fraction of prohibited turns over random networks is
less than 0.2. The effect of root bridge election on the
performance of the protocol is limited both in the random and
regular networks studied.
The use of hierarchical, tree-descriptive addresses simplifies
the routing. and avoids the need of all nodes having a global
knowleddge of the network topology. Routing frames through
the hierarchical tree at very high speed is possible by
progressive decoding of frame destination address, without
routing tables or port address learning. Coexistence with
standard bridges is achieved using combined devices: bridges
that forward the frames having global destination MAC
addresses as standard bridges and frames with local MAC
frames with the proposed protocol.TRUEpu