Unreliable failure detectors have been an important abstraction to build dependable distributed applications over asynchronous systems subject faults. Their implementations are commonly based on timeouts ensure algorithm termination. However, for built the Internet, it is hard estimate this time value due traffic variations. Thus, different types of predictors used model behavior and make predictions delays. In order increase quality service (QoS), self-tuned dynamically adapt their...

Resumo A evasão no ensino superior representa uma preocupação na gestão e um campo fértil para estudos acadêmicos. No Brasil, o Ministério de Educação realiza anualmente Censo da Superior. As diferentes metodologias cálculo análises indicadores são muitas vezes consideradas imprecisas inconsistentes, fragilizando sua importância nas estratégias políticas públicas controle melhorias dos serviços educacionais âmbito das instituições superior. Este trabalho exploratório com abordagem...

Distributed Denial of Service (DDoS) attacks become increasingly sophisticated and massive in traffic volume. These can be mainly classified as IP Spoofing or Real Source IP. In special, are characterized by the use malware-infected hosts to simulate real network traffic. Those constantly evolving, new infection methods always being employed attackers. To deal with such constant change, research community is searching for advanced approaches mitigate, even eliminate, those threats. One these...

Traditional perimeter defense is typically being performed through dedicated firewall-based devices. However, it becomes necessary to decrease the surface and exposure cyber attacks by hiding infrastructure, applications access controls, as well increasing security levels. Software Defined Perimeter (SDP) brings new functionality Single Packet Authorization (SPA) first step. Through analysis of SDP protocol there were issues that need be improved or addressed. This work proposes adaptations...

Intelligent computing techniques have a paramount importance to the treatment of cybersecurity incidents. In such Artificial Intelligence (AI) context, while most algorithms explored in domain aim present solutions intrusion detection problems, these seldom approach correction procedures that are resolution incident problems already took place. practice, knowledge regarding data and is being under-used development intelligent systems, sometimes even lost not used at all. this work proposes...

The design of dependable distributed applications is a hard task, mainly because the indefinable statistic behavior communication delays. Despite this feature, in practice, most system monitors make use timeouts (a maximum waiting time) to ensure some termination properties their protocols. To have better results, dynamically predict new timeout values based on observed delays improve performance and accuracy In last years, time series theory emerged computer science area as good tool...

Summary Intrusion detection systems (IDSs) are a fundamental component of defense solutions. In particular, IDSs aim to detect malicious activities on computer and networks by relying data classification models built from training dataset. However, classifiers' performance can vary for each attack pattern. A common technique overcome this issue is use ensemble methods, where multiple classifiers employed final decision taken combining their outputs. Despite the potential advantages such an...

To manage cybersecurity incidents, organizations maintain an incident response team to mitigate the incidents and avoid unavailability of services. Despite need generate fast correct responses these problems, only a few companies have ways learning from past problem-solving experiences. This work proposes knowledge-based framework for this problem. The paper specifies main tasks method case-based management resolution procedures. It shows that domain expert knowledge about problems routines...

Delay or Disruption Tolerant Networks (DTN) are characterized by long delays and intermittent connectivity, requiring efficient energy consumption for increasing the mobile nodes lifetime. The movements of modify network topology, changing number connection opportunities between nodes. This paper proposes a new technique saving on DTN using trajectory inference model powered machine learning techniques. objective this work is to reduce mobility prediction method. Experimental results...

The widespread use of the Internet comes accompanied with severe threats for web applications security. Intrusion Detection Systems (IDS) have been considered to deal diversity and complexity attacks. In this context, work proposes an algorithm attack detection, exploring ananomaly-based technique: wavelet transform. proposed analyzes anomalies within variations on characters frequencies in requests. Experimental results show high rates detection without false positive occurrences.Keywords:...

Os provedores de serviço Internet (Internet Service Provider – ISPs) oferecem uma infraestrutura comunicação essencial para a realização tarefas cotidianas da sociedade através e o advento novas tecnologias em rede. Todavia, um desafio dos ISPs consiste reduzir taxa churn, qual engloba cancelamentos planos clientes. Apesar existirem esforços na literatura, os ISP permanecem com carência ferramentas identificar churn. Esse trabalho propõe ChurnSense, processo perfis clientes base técnicas...

Distributed simulations aim to replicate real-world behavior viacomputer networks. Developing source code for such simulations, especiallywhile following standards like High-Level Architecture (HLA), is often error-prone due its complexity, prompting generators requiring abstractmodels. This work proposes a model-driven methodology that explores theObject-Process Methodology (OPM) modeling language in conjunction withUML generate accurate and readable high-level models DSEEP andtransform...

Intelligent computing techniques have a paramount importance to the treatment of cybersecurity incidents.In such Artificial Intelligence (AI) context, while most algorithms explored in domain aim present solutions intrusion detection problems, these seldom approach correction procedures that are resolution incident problems already took place.In practice, knowledge regarding data and is being under-used development intelligent systems, sometimes even lost not used at all.In this work...

O presente trabalho insere-se no debate sobre a gestao da seguranca informacao, caracterizando este processo nas empresas incubadas e graduadas Incubadora Tecnologica de Santa Maria (ITSM), salientando percepcao dos gestores das que tange importância utilizacao mecanismos para gerir forma segura suas informacoes. Assim, pesquisa foi dividida em duas etapas, sendo primeira exploratoria qualitativa segunda descritiva quantitativa. A amostra utilizada nesse estudo denominada probabilistica o...

Proteger os recursos de informação tornou-se um grande desafio às organizações devido ao avanço das tecnologias informação, a integração entre essas e o fluxo constante informações que trafegam através redes comunicação. O estabelecimento uma Política Segurança da Informação – PSI pode resolver parte dos problemas relacionados à segurança, mas não resolvê-los integralmente, pois humanos, presentes no ambiente interno organizações, podem comprometer efetividade PSI. Dada relevância aspectos...

The analysis of network traffic is a key area for the management fault-tolerant systems, since anomalies in can affect availability and quality service (QoS). This work proposes an intrusion detection tool based on two-dimensional wavelet transform to quickly effectively detect computer networks generated by denial (DoS). Experiments were performed using two databases: synthetic (DARPA) another one from data collected at Federal University Santa Maria (UFSM), allowing under different...

A Failure Detector (FD) is an essential building block to develop reliable applications on Mobile Ad-hoc NETworks (MANETs). To detect faulty nodes correctly the node mobility and disconnections are challenges that must be circumvented by FD. This paper presents FD algorithm detects efficiently in MANETs. approach explores signal power intensity receiving messages improve failure detector knowledge. The experimental results have shown improvements Quality of Service (QoS) when compared with...

Abstract Due to their fundamental role in the design of fault-tolerant distributed systems, consensus protocols have been widely studied. Most research this area has focused on providing ways for circumventing impossibility reaching a purely asynchronous system subject failures. Of particular interest are indulgent based upon weak failure detection oracles. Following first works that were more concerned with correctness such protocols, performance issues related them now topic gained...