arXiv subjects

Miguel Rio

Publications and source records attributed to Miguel Rio.

13 recordsLinked to original sources

arXiv2024

ReStorEdge: An edge computing system with reuse semantics

This paper investigates an edge computing system where requests are processed by a set of replicated edge servers. We investigate a class of applications where similar queries produce identical results. To reduce processing overhead on the edge servers we store the results of previous computations and return them when new queries are sufficiently similar to earlier ones that produced the results, avoiding the necessity of processing every new query. We implement a similarity-based data classification system, which we evaluate based on real-world datasets of images and voice queries. We evaluate a range of orchestration strategies to distribute queries and cached results between edge nodes and show that the throughput of queries over a system of distributed edge nodes can be increased by 25-33%, increasing its capacity for higher workloads.

cs.ET↗

arXiv2021

Skeleton-Split Framework using Spatial Temporal Graph Convolutional Networks for Action Recogntion

There has been a dramatic increase in the volume of videos and their related content uploaded to the internet. Accordingly, the need for efficient algorithms to analyse this vast amount of data has attracted significant research interest. An action recognition system based upon human body motions has been proven to interpret videos contents accurately. This work aims to recognize activities of daily living using the ST-GCN model, providing a comparison between four different partitioning strategies: spatial configuration partitioning, full distance split, connection split, and index split. To achieve this aim, we present the first implementation of the ST-GCN framework upon the HMDB-51 dataset. We have achieved 48.88 % top-1 accuracy by using the connection split partitioning approach. Through experimental simulation, we show that our proposals have achieved the highest accuracy performance on the UCF-101 dataset using the ST-GCN framework than the state-of-the-art approach. Finally, accuracy of 73.25 % top-1 is achieved by using the index split partitioning strategy.

cs.CV↗

arXiv2021

Skeleton Split Strategies for Spatial Temporal Graph Convolution Networks

A skeleton representation of the human body has been proven to be effective for this task. The skeletons are presented in graphs form-like. However, the topology of a graph is not structured like Euclidean-based data. Therefore, a new set of methods to perform the convolution operation upon the skeleton graph is presented. Our proposal is based upon the ST-GCN framework proposed by Yan et al. [1]. In this study, we present an improved set of label mapping methods for the ST-GCN framework. We introduce three split processes (full distance split, connection split, and index split) as an alternative approach for the convolution operation. To evaluate the performance, the experiments presented in this study have been trained using two benchmark datasets: NTU-RGB+D and Kinetics. Our results indicate that all of our split processes outperform the previous partition strategies and are more stable during training without using the edge importance weighting additional training parameter. Therefore, our proposal can provide a more realistic solution for real-time applications centred on daily living recognition systems activities for indoor environments.

cs.CV↗

arXiv2017

Bringing SDN to the Mobile Edge

Nowadays, Software Defined Network (SDN) architectures and applications are revolutionizing the way wired networks are built and operate. However, little is known about the potential of this disruptive technology in wireless mobile networks. In fact, SDN is based on a centralized network control principle, while existing mobile network protocols give emphasis on the distribution of network resources and their management. Therefore, it is challenging to apply SDN ideas in the context of mobile networks. In this paper, we propose methods to overcome these challenges and make SDN more suitable for the mobile environment. Our main idea is to combine centralized SDN and distributed control in a hybrid design that takes the best of the two paradigms; (i) global network view and control programmability of SDN and (ii) robustness of distributed protocols. We discuss the pros and cons of each method and highlight them in an SDN prototype implementation built using off-the-shelf mobile devices.

cs.NI↗

arXiv2014

TARDIS: Stably shifting traffic in space and time (extended version)

This paper describes TARDIS (Traffic Assignment and Retiming Dynamics with Inherent Stability) which is an algorithmic procedure designed to reallocate traffic within Internet Service Provider (ISP) networks. Recent work has investigated the idea of shifting traffic in time (from peak to off-peak) or in space (by using different links). This work gives a unified scheme for both time and space shifting to reduce costs. Particular attention is given to the commonly used 95th percentile pricing scheme. The work has three main innovations: firstly, introducing the Shapley Gradient, a way of comparing traffic pricing between different links at different times of day; secondly, a unified way of reallocating traffic in time and/or in space; thirdly, a continuous approximation to this system is proved to be stable. A trace-driven investigation using data from two service providers shows that the algorithm can create large savings in transit costs even when only small proportions of the traffic can be shifted.

cs.NI↗

arXiv2013

Distributed Overlay Anycast Table using Space filling curves

In this paper we present the \emph{Distributed Overlay Anycast Table}, a structured overlay that implements application-layer anycast, allowing the discovery of the closest host that is a member of a given group. One application is in locality-aware peer-to-peer networks, where peers need to discover low-latency peers participating in the distribution of a particular file or stream. The DOAT makes use of network delay coordinates and a space filling curve to achieve locality-aware routing across the overlay, and Bloom filters to aggregate group identifiers. The solution is designed to optimise both accuracy and query time, which are essential for real-time applications. We simulated DOAT using both random and realistic node distributions. The results show that accuracy is high and query time is low.

cs.NI↗

arXiv2013

Scalable peer-to-peer streaming for live entertainment content

We present a system for streaming live entertainment content over the Internet originating from a single source to a scalable number of consumers without resorting to centralised or provider- provisioned resources. The system creates a peer-to-peer overlay network, which attempts to optimise use of existing capacity to ensure quality of service, delivering low start-up delay and lag in playout of the live content. There are three main aspects of our solution. Firstly, a swarming mechanism that constructs an overlay topology for minimising propagation delays from the source to end consumers. Secondly, a distributed overlay anycast system that uses a location-based search algorithm for peers to quickly find the closest peers in a given stream. Finally, a novel incentives mechanism that encourages peers to donate capacity even when the user is not actively consuming content.

cs.NI↗

arXiv2009

Two Algorithms for Network Size Estimation for Master/Slave Ad Hoc Networks

This paper proposes an adaptation of two network size estimation methods: random tour and gossip-based aggregation to suit master/slave mobile ad hoc networks. We show that it is feasible to accurately estimate the size of ad hoc networks when topology changes due to mobility using both methods. The algorithms were modified to account for the specific constraints of master/slave ad hoc networks and the results show that the proposed modifications perform better on these networks than the original protocols. Each of the two algorithms presents strengths and weaknesses and these are outlined in this paper.

cs.NI↗

arXiv2009

Criticisms of modelling packet traffic using long-range dependence

This paper criticises the notion that long-range dependence is an important contributor to the queuing behaviour of real Internet traffic. The idea is questioned in two different ways. Firstly, a class of models used to simulate Internet traffic is shown to have important theoretical flaws. It is shown that this behaviour is inconsistent with the behaviour of real traffic traces. Secondly, the notion that long-range correlations significantly affects the queuing performance of traffic is investigated by destroying those correlations in real traffic traces (by reordering). It is shown that the longer ranges of correlations are not important except in one case with an extremely high load.

cs.NI↗

arXiv2009

Evaluating and Optimising Models of Network Growth

This paper presents a statistically sound method for measuring the accuracy with which a probabilistic model reflects the growth of a network, and a method for optimising parameters in such a model. The technique is data-driven, and can be used for the modeling and simulation of any kind of evolving network. The overall framework, a Framework for Evolving Topology Analysis (FETA), is tested on data sets collected from the Internet AS-level topology, social networking websites and a co-authorship network. Statistical models of the growth of these networks are produced and tested using a likelihood-based method. The models are then used to generate artificial topologies with the same statistical properties as the originals. This work can be used to predict future growth patterns for a known network, or to generate artificial models of graph topology evolution for simulation purposes. Particular application examples include strategic network planning, user profiling in social networks or infrastructure deployment in managed overlay-based services.

cs.NI↗

arXiv2008

Beyond Node Degree: Evaluating AS Topology Models

Many models have been proposed to generate Internet Autonomous System (AS) topologies, most of which make structural assumptions about the AS graph. In this paper we compare AS topology generation models with several observed AS topologies. In contrast to most previous works, we avoid making assumptions about which topological properties are important to characterize the AS topology. Our analysis shows that, although matching degree-based properties, the existing AS topology generation models fail to capture the complexity of the local interconnection structure between ASs. Furthermore, we use BGP data from multiple vantage points to show that additional measurement locations significantly affect local structure properties, such as clustering and node centrality. Degree-based properties, however, are not notably affected by additional measurements locations. These observations are particularly valid in the core. The shortcomings of AS topology generation models stems from an underestimation of the complexity of the connectivity in the core caused by inappropriate use of BGP data.

cs.NI↗

arXiv2007

Towards Informative Statistical Flow Inversion

A problem which has recently attracted research attention is that of estimating the distribution of flow sizes in internet traffic. On high traffic links it is sometimes impossible to record every packet. Researchers have approached the problem of estimating flow lengths from sampled packet data in two separate ways. Firstly, different sampling methodologies can be tried to more accurately measure the desired system parameters. One such method is the sample-and-hold method where, if a packet is sampled, all subsequent packets in that flow are sampled. Secondly, statistical methods can be used to ``invert'' the sampled data and produce an estimate of flow lengths from a sample. In this paper we propose, implement and test two variants on the sample-and-hold method. In addition we show how the sample-and-hold method can be inverted to get an estimation of the genuine distribution of flow sizes. Experiments are carried out on real network traces to compare standard packet sampling with three variants of sample-and-hold. The methods are compared for their ability to reconstruct the genuine distribution of flow sizes in the traffic.

cs.NI↗

arXiv2007

Revisiting the Issues On Netflow Sample and Export Performance

The high volume of packets and packet rates of traffic on some router links makes it exceedingly difficult for routers to examine every packet in order to keep detailed statistics about the traffic which is traversing the router. Sampling is commonly applied on routers in order to limit the load incurred by the collection of information that the router has to undertake when evaluating flow information for monitoring purposes. The sampling process in nearly all cases is a deterministic process of choosing 1 in every N packets on a per-interface basis, and then forming the flow statistics based on the collected sampled statistics. Even though this sampling may not be significant for some statistics, such as packet rate, others can be severely distorted. However, it is important to consider the sampling techniques and their relative accuracy when applied to different traffic patterns. The main disadvantage of sampling is the loss of accuracy in the collected trace when compared to the original traffic stream. To date there has not been a detailed analysis of the impact of sampling at a router in various traffic profiles and flow criteria. In this paper, we assess the performance of the sampling process as used in NetFlow in detail, and we discuss some techniques for the compensation of loss of monitoring detail.

cs.PF↗