IMDEA Networks Institute Digital Repository
Not a member yet
1915 research outputs found
Sort by
Learning from Experience: Efficient Decentralized Scheduling for 60GHz Mesh Networks
Get PDFDue to the directionality of transmissions in millimeter
wave (mm-wave) networks, wireless stations are usually
unable to overhear when other stations access the channel. This makes it hard to design efficient distributed beam coordination and scheduling mechanisms. At the same time, centralized schemes only perform well in relatively simple, static scenarios. In practical settings where links exhibit different channel qualities and in the context of relaying or in-band backhauling, centrally coordinating all stations becomes difficult. In this paper, we propose a low complexity, decentralized, learning-based scheduling algorithm for mm-wave networks that handles heterogeneous link rates and packet sizes efficiently. Compared to state-of-the art slotted channel access for mm-wave networks, the proposed mechanism achieves throughput gains of up to a factor of 8 in single-hop scenarios and end-to-end throughput improvements of up to a factor of 1.6 in multi-hop topologies.TRUEpu
Distributed Slicing in Dynamic Systems
Get PDFPeer to peer (P2P) systems have moved from application specific architectures to a generic service oriented design philosophy. This raised interesting problems in connection with providing useful P2P middleware services capable of dealing with resource assignment and management in a large-scale, heterogeneous and unreliable environment. The slicing problem consists of partitioning a P2P network into k groups (slices) of a given portion of the network nodes that share similar resource values. As the network is large and dynamic this partitioning is continuously updated without any node knowing the network size. In this paper, we propose the first algorithm to solve the slicing problem. We introduce the metric of slice disorder and show that the existing ordering algorithm cannot nullify this disorder. We propose a new algorithm that speeds up the existing ordering algorithm but that suffers from the same inaccuracy. Then, we propose another algorithm based on ranking that is provably convergent under reasonable assumptions. In particular, we notice experimentally that ordering algorithms suffer from resource-correlated churn while the ranking algorithm can cope with it. These algorithms are proved viable theoretically and experimentally.pu
Pushing the Frontier: Exploring the African Web Ecosystem
No full textThe video of the presentation is available at https://www.internetsociety.org/afpif-2016/day1-presentations-and-livestream (Session 4: Measurements Panel)It is well known that Africa’s mobile and fixed Internet infrastructure is progressing at a rapid pace. A flurry of recent research has quantified this, highlighting the expansion of its underlying connectivity network. However, improving the infrastructure is not useful without appropriately provisioned services to utilize it. This paper measures the availability of web content infrastructure in Africa. Whereas others have explored web infrastructure in developed regions, we shed light on practices in developing regions. To achieve this, we apply a comprehensive measurement methodology to collect data from a variety of sources. We focus on a large content delivery network to reveal that Africa’s content infrastructure is, indeed, expanding. However, we find much web content is still served from the US and Europe. We discover that many of the problems faced are actually caused by significant inter-AS delays in Africa, which contribute to local ISPs not sharing their cache capacity. We discover that a related problem is the poor DNS configuration used by some ISPs, which confounds the attempts of providers to optimize their delivery. We then explore a number of other websites to show that large web infrastructure deployments are a rarity in Africa and that even regional websites host their services abroad. We conclude by making suggestions for improvements.TRUEpu
Single and Multiple Buffer Processing
Get PDFBuffer management policies are online algorithms that control a limited buffer of packets with homogeneous or heterogeneous characteristics, deciding whether to accept new packets when they arrive, which packets to process and transmit, and possibly whether to push out packets already residing in the buffer. Although settings differ, the problem is always to achieve the best possible competitive ratio, i.e., find a policy with good worst-case guarantees in comparison with an optimal offline clairvoyant algorithm. The policies themselves are often simple, simplicity being an important advantage for implementation in switches; the hard problem is to find proofs of lower
and especially upper bounds for their competitive ratios. Thus, this problem is more theoretical in nature, although the resulting throughput guarantees are important tools in the design of network elements. Comprehensive surveys of this field have been given in the past by Goldwasser and
Epstein and van Stee.TRUEpu
Measurement-Based Coalescing Control for 802.3az
Get PDFIEEE 802.3az standard (EEE), is the energy-aware
alternative to legacy Ethernet. To save energy by extending the sojourn in the Low Power Idle state of EEE, packet coalescing has been proposed. While coalescing improves by far the energy efficiency of EEE, it is still far from achieving energy consumption proportional to traffic. Moreover, coalescing can introduce high delays. In this work, we use sensitivity analysis to evaluate the impact of coalescing timers and buffer sizes, and to shed
light on the delay incurred by adopting coalescing schemes. Accordingly, we design and study measurement-based coalescing control solutions that tune the coalescing parameters on-the-fly, according to the instantaneous load and the coalescing delay experienced by the packets. Our results show that, by relying on run-time delay measurements, simple and practical adaptive coalescing schemes outperform traditional static and dynamic coalescing. Notably, our schemes double the energy saving benefit of legacy EEE coalescing and allow to control the coalescing delay.TRUEpu
A TDMA-based MAC Protocol Exploiting the Near-Far Effect in Underwater Acoustic Networks
Get PDFA prime source of collisions in underwater acoustic communication networks (UWANs) is the so called near-far effect, where a node located farther from the receiver is jammed by a closer node. While common practice considers such situation as a challenge, in this paper we consider it as a resource, and use it to increase network throughput of spatial reuse time-division multiple access. We propose a transmission allocation algorithm that opportunistically utilizes information on occurrences of near-far scenarios in UWANs to maximize channel utilization. Numerical results show that, at a slight cost in terms of fairness, our scheduling solutions achieve higher throughput and lower transmission delay than benchmark spatial-reuse scheduling protocols. To allow the reproducibility of our results, we publish the implementation of our proposed algorithm.TRUEpu
Demo: MONROE, a distributed platform to measure and assess mobile broadband networks
Get PDFMBB; large scale experiments; Demo.This demo presents the MONROE distributed platform and how it can be used to implement measurement and assessment experiments with operational mobile broadband networks (MBBs). MONROE provides registered experimenters with open access to hundreds of nodes, distributed over several European countries and equipped with multiple MBB connections, and a backend system that collects the measurement results. Experiments are scheduled through a user-friendly web client, with no need to directly access the nodes. The platform further embeds tools for real-time traffic flow analysis and a powerful visualization tool.TRUEpu
The Fail-Safe Operation of Collaborative Driving Systems
No full textPublished online December 2014.We consider the problem of collaborative driving systems in which the vehicles share sensor information and make joint decisions on vehicle control. Such system involves vehicular control, communications links and environment sensing. We propose a multiple stack architecture that is
based on the nature of collaborative driving systems. The architecture facilitates the design, implementation, verification and testing of collaborative driving protocol. The advantage of using absolute time in a distributed system is also demonstrated. Specifically, we propose a
merge protocol that assists a driver in merging with other vehicles based on the proposed architecture. We create a lock protocol and specify it as an extended finite state machine as a subroutine of the merge protocol. Absolute time simplifies a deadlock-free lock protocol. The
lock protocol is verified using probabilistic verification.pu
Network Interface Power Management and TCP Congestion Control: a Troubled Marriage
Get PDFOptimizing the tradeoff between power saving and Quality of Service (QoS) in the current Internet is a challenging research objective, whose difficulty stems also from the dominant presence of TCP traffic, and its elastic nature. In a previous work we have shown that an intertwining exists between capacity scaling approaches and TCP congestion control. In this paper we investigate the reasons of such intertwining, and we evaluate how and how much the dynamics of the two algorithms affect each other’s performance. More specifically, we will show that such an interaction is essentially due to the overlap of the two closed loop controls, with different time constants.pu
A Handshake-based Protocol Exploiting the Near-Far Effect in Underwater Acoustic Networks
Get PDFWe propose a scheme that opportunistically exploits received power diversity across different packets to both favor spatial reuse in underwater acoustic networks and limit the exposed terminal problem. Since the power attenuation in the underwater acoustic channel is large, such power differences are common. This situation, also referred to as the near-far effect and often considered a problem, is converted into a resource by means of multipacket reception (MPR). Yet, even without MPR capability our scheme offers performance benefits. Our scheme is simple, lightweight and distributed, and can be easily implemented over any protocol based on carrier-sense multiple access with collision avoidance. Our results show that higher throughput and lower transmission delay are achieved compared to a benchmark channel access protocol. Our results are validated and demonstrated in a lake experiment. To allow reproducibility, the implementation of our scheme is publicly available.pu