Learn More
—One of the major problems in managing large-scale distributed systems is the prediction of the application performance. The complexity of the systems and the availability of monitored data have motivated the applicability of machine learning and other statistical techniques to induce performance models and forecast performance degradation problems.(More)
As distributed component-based applications increase in size and complexity, on-line application monitoring becomes a crucial issue for assuring quality of service. In this paper we propose a monitoring architecture built into the component system itself that allows a fine-grained observation of the resources being in use by an application. Then, we provide(More)
In this work, we present an approach to performance problem determination in SCS, a CORBA-based component system. Particularly, our approach is built on a monitoring infrastructure provided by the middleware itself and an analysis mechanism based on a Bayesian network model.
Scientists often rely on analytical modeling for predicting performance because it provides a flexible and fast estimate without the need for low-level details. With traditional performance modeling tools, users first develop a performance model and then repeatedly evaluate and analyze the model manually. As we observe a strong trend towards system designs(More)
This paper presents an application developed in Matlab for relative sag source location in medium voltage distribution systems. The main objective is the identification of potential users or industrial customers causing sags in a circuit. Application uses topological information of the circuit and works with S-Transform to produce distortion power patterns(More)
Large-scale data centers allow organizations to gain access to computer resources without incurring high costs in purchasing and maintaining IT infrastructure. In these environments, due to the large number of hardware and software involved, anomaly detection is difficult but essential for service provisioning. Computer systems hosted in data centers(More)
Spectrum selection is a key issue in Dynamic Spectrum Access (DSA). The purpose of the selection is to minimize interference with legacy devices and maximize the discovery of opportunities or white spaces. There are several solutions to this issue, and Reinforcement Learning algorithms are among the most successful. Through simulation, we compare the(More)
Dynamic circuits can reduce queuing delays and provide alternative routes to network congestions. In general, dynamic circuits are available to users who need to transmit large volumes of data and the circuits are explicitly generated by the users or by custom application systems. In this paper, we present a proposal for automatic generation of dynamic(More)