IEEE Transactions on Very Large Scale Integration (VLSI) Systems
A predictive system shutdown method for energy saving of event-driven computation
ICCAD '97 Proceedings of the 1997 IEEE/ACM international conference on Computer-aided design
Dynamic power management using adaptive learning tree
ICCAD '99 Proceedings of the 1999 IEEE/ACM international conference on Computer-aided design
Adaptive disk spin—down for mobile computers
Mobile Networks and Applications
Numerical Recipes in C: The Art of Scientific Computing
Numerical Recipes in C: The Art of Scientific Computing
Dynamic management of power consumption
Power aware computing
Dynamic power management using machine learning
Proceedings of the 2006 IEEE/ACM international conference on Computer-aided design
Policy optimization for dynamic power management
IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems
Online learning with expert advice and finite-horizon constraints
AAAI'08 Proceedings of the 23rd national conference on Artificial intelligence - Volume 1
Hi-index | 0.00 |
Power management techniques for mobile appliances put the components of the systems into low power states to maximize battery life while minimizing the impact on the perceived performance of the devices. Static timeout policies are the state-of-the-art approach for solving power management problems. In this work, we propose adaptive timeout policies as a simple and efficient solution for fine-grained power management. As discussed in the paper, the policies reduce the latency of static timeout policies by nearly one half at the same power savings. This result can be also viewed as increasing the power savings of static timeout policies at the same latency target. The main objective of our work is to propose practical adaptive policies. Therefore, our adaptive solution is fast enough to be executed within less than one millisecond, and sufficiently simple to be deployed directly on a microcontroller. We validate our ideas on two recorded CPU activity traces, which involve more than 10 million entries each.