Saulo O. D. Luiz

Optimization of timeout-based power management policies for network interfaces

Dynamic power management policies are essential for battery-powered consumer electronics devices, which are mostly equipped with power consuming network interfaces. Timeout-based power management policies are implemented on most operating systems, but the selection of the policy parameters generally resort to heuristic approaches and there is lack of support for network interfaces dynamic power management. Therefore, this paper introduces an innovative solution for timeout-based power management policies for network interfaces. The presented policy exploits statistical data analysis and the optimization of both power savings and performance. The effectiveness of the introduced policy is demonstrated by experimental results1.

Multisize Sliding Window in Workload Estimation for Dynamic Power Management

Energy efficiency has become one of the key challenges for a large class of electronic systems. Longer time between battery recharges is highly desirable for battery-powered devices such as mobile phones, digital cameras, Internet tablets, and electronic organizers. Energy efficiency is also important for electronic systems powered from the electric grid since it may reduce power consumption and the cooling requirements. Power savings are possible because electronic systems generally have an idle state, when, for example, the processor can run in a low-power state. Thus, the correct estimation of the workload model plays an essential role in the decision of which and when a power state transition should be performed by the electronic system. This paper introduces a multisize sliding window workload estimation technique for dynamic power management (DPM) in nonstationary environments. This technique reduces both the effects of identification delay and sampling error present in the previous fixed-size sliding window approach. The system is modeled by discrete-time Markov chains and the model offers a rigorous mathematical formulation of the problem and allows one to obtain an excellent trade-off between performance and power consumption.

System identification and energy-aware processor utilization control

Minimization of power consumption is essential for battery-powered computing systems. This paper proposes an innovative solution for the processor power management problem. The proposed technique does not require accurate power consumption and performance model of the processor. The proposed solution exploits feedback control principles and the controller design is optimized for a given workload. Experimental results obtained from a case study are presented demonstrating, usefulness and the feasibility of the proposed solution.

Formal specification of DSP gateway for data transmission between processor cores of OMAP platform

In this article, formal models for the data transmission mechanism between ARM and DSP cores of OMAP platform 161x via DSP Gateway are presented. These models are represented as timed automata. The automata for the behavior of tasks running in the DSP when receiving requests of read or write operations from ARM are described. Data transmission between ARM and DSP is modelled according to the mechanism offered by tokliBIOS, the system kernel at DSP side. The formal model presented in this work helps developers to understand the communication mechanisms of DSP Gateway and facilitates its usage and future development.

Computer systems power model estimation

The estimation of power models for computer systems is important for the development process of dynamic power managers. In this work, a power model is estimated based on a constrained linear least-squares technique.

Identification and control for processor power management

Minimization of power consumption is essential for battery-powered computing systems. In this paper, System identification and Feedback control are applied to the processor power management problem when the workload is known a priori.

Workload estimation f or power management effects on battery-powered embedded systems

The effects of accuracy and delay of the identification of a nonstationary workload of a battery powered device using a dynamic power management technique are analyzed in this paper. The system is modeled by discrete-time Markov chains coupled to the battery model. Such model allows a rigorous mathematical formulation of the problem and to trade-off between performance and battery lifetime.

Dynamic power management for network interfaces

Minimizing the power consumption in mobile devices is critical to increase the battery life. The present work simulated and implemented, at the user space level, four dynamic power management policies for network interfaces. In addition, one mechanism was improved and enabled the identification of three specific workloads for network interfaces.

Dynamic timeout power policy for network interfaces

Network interfaces are present in a variety of battery powered computer systems, such as notebooks, ultrabooks, tablets and mobile phones. Considering that the battery life is still a constraint, there are opportunities for power management by reducing the power consumption of idle network interfaces. In this work, a dynamic timeout power policy for network interfaces is presented, incorporating workload estimation and a problem optimization at runtime, thus providing power savings and suitable performance.

Using location to infer user contextual information: A case study on pervasive advertising

Smartphones are available anytime and anywhere, enabling applications to determine the user location. In this work we investigate the hypothesis that it is possible to determine contextual information of the users based on the places they usually go. In this sense, we developed a model to infer such information and carried out an empirical experiment to validate the model. We also use the contextual information obtained from the model to increase the relevance of delivered advertisements in a scenario of pervasive advertising.