Extending fixed task-priority schedulability by interference limitation

While the earliest deadline first algorithm is known to be optimal as a uniprocessor scheduling policy, the implementation comes at a cost in terms of complexity. Fixed taskpriority algorithms on the other hand have lower complexity but higher likelihood of task sets being declared unschedulable, when compared to earliest deadline first (EDF). Various attempts have been undertaken to increase the chances of proving a task set schedulable with similar low complexity. In some cases, this was achieved by modifying applications to limit preemptions, at the cost of flexibility. In this work, we explore several variants of a concept to limit interference by locking down the ready queue at certain instances. The aim is to increase the prospects of schedulability of a given task system, without compromising on complexity or flexibility, when compared to the regular fixed task-priority algorithm. As a final contribution, a new preemption threshold assignment algorithm is provided which is less complex and more straightforward than the previous method available in the literature.

JamLab: augmenting sensornet testbeds with realistic and controlled interference generation

Radio interference drastically affects the performance of sensor-net communications, leading to packet loss and reduced energy-efficiency. As an increasing number of wireless devices operates on the same ISM frequencies, there is a strong need for understanding and debugging the performance of existing sensornet protocols under interference. Doing so requires a low-cost flexible testbed infrastructure that allows the repeatable generation of a wide range of interference patterns. Unfortunately, to date, existing sensornet testbeds lack such capabilities, and do not permit to study easily the coexistence problems between devices sharing the same frequencies. This paper addresses the current lack of such an infrastructure by using off-the-shelf sensor motes to record and playback interference patterns as well as to generate customizable and repeat-able interference in real-time. We propose and develop JamLab: a low-cost infrastructure to augment existing sensornet testbeds with accurate interference generation while limiting the overhead to a simple upload of the appropriate software. We explain how we tackle the hardware limitations and get an accurate measurement and regeneration of interference, and we experimentally evaluate the accuracy of JamLab with respect to time, space, and intensity. We further use JamLab to characterize the impact of interference on sensornet MAC protocols.

Quantifying the channel quality for interference-aware wireless sensor networks

Reliability of communications is key to expand application domains for sensor networks. SinceWireless Sensor Networks (WSN) operate in the license-free Industrial Scientific and Medical (ISM) bands and hence share the spectrum with other wireless technologies, addressing interference is an important challenge. In order to minimize its effect, nodes can dynamically adapt radio resources provided information about current spectrum usage is available. We present a new channel quality metric, based on availability of the channel over time, which meaningfully quantifies spectrum usage. We discuss the optimum scanning time for capturing the channel condition while maintaining energy-efficiency. Using data collected from a number of Wi-Fi networks operating in a library building, we show that our metric has strong correlation with the Packet Reception Rate (PRR). This suggests that quantifying interference in the channel can help in adapting resources for better reliability. We present a discussion of the usage of our metric for various resource allocation and adaptation strategies.

A Random Access Protocol incorporating Multi-Packet Reception, Retransmission Diversity and Successive Interference Cancellation

8th International Workshop on Multiple Access Communications (MACOM2015), Helsinki, Finland.

Analysis of self-interference within DAG tasks

6th Real-Time Scheduling Open Problems Seminar (RTSOPS 2015), Lund, Sweden.

On the Scalability of Constructive Interference in Low-Power Wireless Networks

12th European Conference on Wireless Sensor Networks (EWSN 2015). 9 to 11, Feb, 2015. Porto, Portugal

PIASA: A power and interference aware resource management strategy for heterogeneous workloads in cloud data centers

Cloud data centers have been progressively adopted in different scenarios, as reflected in the execution of heterogeneous applications with diverse workloads and diverse quality of service (QoS) requirements. Virtual machine (VM) technology eases resource management in physical servers and helps cloud providers achieve goals such as optimization of energy consumption. However, the performance of an application running inside a VM is not guaranteed due to the interference among co-hosted workloads sharing the same physical resources. Moreover, the different types of co-hosted applications with diverse QoS requirements as well as the dynamic behavior of the cloud makes efficient provisioning of resources even more difficult and a challenging problem in cloud data centers. In this paper, we address the problem of resource allocation within a data center that runs different types of application workloads, particularly CPU- and network-intensive applications. To address these challenges, we propose an interference- and power-aware management mechanism that combines a performance deviation estimator and a scheduling algorithm to guide the resource allocation in virtualized environments. We conduct simulations by injecting synthetic workloads whose characteristics follow the last version of the Google Cloud tracelogs. The results indicate that our performance-enforcing strategy is able to fulfill contracted SLAs of real-world environments while reducing energy costs by as much as 21%.

Scalable and interference aware wi-fi mesh networks using cots devices

A crescente tendencia no acesso móvel tem sido potenciada pela tecnologia IEEE 802.11. Contudo, estas redes têm alcance rádio limitado. Para a extensão da sua cobertura é possível recorrer a redes emalhadas sem fios baseadas na tecnologia IEEE 802.11, com vantagem do ponto de vista do custo e da flexibilidade de instalação, face a soluções cabladas. Redes emalhadas sem fios constituídas por nós com apenas uma interface têm escalabilidade reduzida. A principal razão dessa limitação deve-se ao uso do mecanismo de acesso ao meio partilhado Carrier Sense Multiple Access with Collision Avoidance (CSMA/CA) em topologias multi-hop. Especificamente, o CSMA/CA não evita o problema do nó escondido levando ao aumento do número de colisões e correspondente degradação de desempenho com impacto direto no throughput e na latência. Com a redução da tecnologia rádio torna-se viável a utilização de múltiplos rádios por nó, sem com isso aumentar significativamente o custo da solução final de comunicações. A utilização de mais do que um rádio por nó de comuniações permite superar os problemas de desempenho inerentes ás redes formadas por nós com apenas um rádio. O objetivo desta tese, passa por desenvolver uma nova solução para redes emalhadas multi-cana, duar-radio, utilizando para isso novos mecanismos que complementam os mecanismos definidos no IEEE 802.11 para o estabelecimento de um Basic Service Set (BSS). A solução é baseada na solução WiFIX, um protocolo de routing para redes emalhadas de interface única e reutiliza os mecanismos já implementados nas redes IEEE 802.11 para difundir métricas que permitam à rede escalar de forma eficaz minimizando o impacto na performance. A rede multi-hop é formada por nós equipados com duas interfaces, organizados numa topologia hierárquica sobre múltiplas relações Access Point (AP) – Station (STA). Os resultados experimentais obtidos mostram a eficácia e o bom desempenho da solução proposta face à solução WiFIX original.

Caracterização do papel da adrenalina na maturação dos adrenorreceptores B2

As respostas pós-juncionais mediadas por adrenorreceptores β2 (ARβ2), responsáveis pelo relaxamento do músculo liso, na veia safena do cão, estão ausentes à nascença. Pelo contrário, no rato recém-nascido já se verifica a estimulação da adenilil ciclase pela activação dos ARβ2. Não existem ainda estudos no coelho recém-nascido. O principal objectivo deste trabalho é avaliar as respostas pós-juncionais mediadas pelos ARβ2 em coelhos recém-nascidos e jovens e relacionar essas respostas com a adrenalina produzida nas glândulas supra-renais. Traçaram-se curvas de dose-resposta à isoprenalina (agonista β) utilizando-se anéis de aorta montados em banho de órgãos isolados ligado a um transdutor de força isométrica. As catecolaminas das supra-renais foram quantificadas por RP-HPLC-ED. Em aortas pré-contraídas com fenilefrina (agonista α1), a isoprenalina causou relaxamento total apenas em coelhos recém-nascidos (n=10). O relaxamento máximo nos coelhos jovens foi de 21±4% (n=23). A potência da isoprenalina foi maior nos recém-nascidos (EC50=1.15×10-8±7.2×10-10 M, n=10) do que nos coelhos jovens (EC50=1.29×10-7 ±4.7×10-9 M, n=23). O relaxamento máximo com isoprenalina, em aortas pré-contraídas com prostaglandina F2α (PGF2α), no grupo de coelhos recém-nascidos foi de 95±3.6% (n=16). O relaxamento máximo nos coelhos jovens foi de 43.7±8.6% (n=9). Na pré-contracção com PGF2α a potência da isoprenalina registou-se maior nos recémnascidos (EC50=9.59×10-9±4.0×10-10 M, n=16) do que nos coelhos jovens (EC50=2.13×10- 8±3.8×10-9 M, n=9), estando concordante com os resultados da pré-contracção com fenilefrina. Nas supra-renais dos recém-nascidos, o conteúdo de noradrenalina foi de 586±128 nmol/mg e da adrenalina foi de 1915±356 nmol/mg (n=4) e nos coelhos jovens foi de 112±12 nmol/mg e de 3644±403 nmol/mg (n=6), respectivamente. As respostas mediadas por ARβ2 no coelho desenvolvem-se mais cedo do que no cão, pois já estão presentes no nascimento. Tal como no rato, no coelho a adrenalina é já a catecolamina em maior quantidade à nascença, enquanto no cão é vestigial. Há uma relação temporal entre a síntese da adrenalina, a única catecolamina biogénica com alta afinidade para os ARβ2 e a maturação das respostas pós-juncionais mediadas por esses receptores. Um protocolo para experiências futuras destinadas a testar esta hipótese, com base no knockdown da Feniletanolamina-N-metiltransferase por RNAi foi elaborado e incluído neste documento.

Intelligent Bio-Inspired system for manufacturing scheduling under uncertainties

A novel approach to scheduling resolution by combining Autonomic Computing (AC), Multi-Agent Systems (MAS), Case-based Reasoning (CBR), and Bio-Inspired Optimization Techniques (BIT) will be described. AC has emerged as a paradigm aiming at incorporating applications with a management structure similar to the central nervous system. The main intentions are to improve resource utilization and service quality. In this paper we envisage the use of MAS paradigm for supporting dynamic and distributed scheduling in Manufacturing Systems with AC properties, in order to reduce the complexity of managing manufacturing systems and human interference. The proposed CBR based Intelligent Scheduling System was evaluated under different dynamic manufacturing scenarios.

Meta-heuristics self-configuration for scheduling

Scheduling resolution requires the intervention of highly skilled human problemsolvers. This is a very hard and challenging domain because current systems are becoming more and more complex, distributed, interconnected and subject to rapidly changing. A natural Autonomic Computing evolution in relation to Current Computing is to provide systems with Self-Managing ability with a minimum human interference. This paper addresses the resolution of complex scheduling problems using cooperative negotiation. A Multi-Agent Autonomic and Meta-heuristics based framework with self-configuring capabilities is proposed.

Self-optimizing through CBR learning

In this paper, we foresee the use of Multi-Agent Systems for supporting dynamic and distributed scheduling in Manufacturing Systems. We also envisage the use of Autonomic properties in order to reduce the complexity of managing systems and human interference. By combining Multi-Agent Systems, Autonomic Computing, and Nature Inspired Techniques we propose an approach for the resolution of dynamic scheduling problem, with Case-based Reasoning Learning capabilities. The objective is to permit a system to be able to automatically adopt/select a Meta-heuristic and respective parameterization considering scheduling characteristics. From the comparison of the obtained results with previous results, we conclude about the benefits of its use.

Autonomic computing for scheduling in manufacturing systems

We describe a novel approach to scheduling resolution by combining Autonomic Computing (AC), Multi-Agent Systems (MAS) and Nature Inspired Optimization Techniques (NIT). Autonomic Computing has emerged as paradigm aiming at embedding applications with a management structure similar to a central nervous system. A natural Autonomic Computing evolution in relation to Current Computing is to provide systems with Self-Managing ability with a minimum human interference. In this paper we envisage the use of Multi-Agent Systems paradigm for supporting dynamic and distributed scheduling in Manufacturing Systems with Autonomic properties, in order to reduce the complexity of managing systems and human interference. Additionally, we consider the resolution of realistic problems. The scheduling of a Cutting and Treatment Stainless Steel Sheet Line will be evaluated. Results show that proposed approach has advantages when compared with other scheduling systems.

Self-managing agents for dynamic scheduling in manufacturing

The main purpose of this paper is to propose a Multi-Agent Autonomic and Bio-Inspired based framework with selfmanaging capabilities to solve complex scheduling problems using cooperative negotiation. Scheduling resolution requires the intervention of highly skilled human problem-solvers. This is a very hard and challenging domain because current systems are becoming more and more complex, distributed, interconnected and subject to rapidly changing. A natural Autonomic Computing (AC) evolution in relation to Current Computing is to provide systems with Self-Managing ability with a minimum human interference.

A avaliação de impacte ambiental em Portugal: análise exploratória às pedreiras da região Norte

Mestrado em Engenharia Geotécnica e Geoambiente