Learning-Augmented Dynamic Power Management with Multiple States via New Ski Rental Bounds

TL;DR

This paper introduces a learning-augmented online algorithm for dynamic power management with multiple states, achieving near-optimal performance by leveraging predictions of idle periods and providing robust guarantees against prediction errors.

Contribution

It presents a novel learning-augmented ski rental algorithm with tight error dependence, improving power management decisions in systems with multiple states.

Findings

Algorithm performs near-optimally with accurate predictions.

Performance degrades gracefully with prediction errors.

Experimental results support theoretical guarantees.

Abstract

We study the online problem of minimizing power consumption in systems with multiple power-saving states. During idle periods of unknown lengths, an algorithm has to choose between power-saving states of different energy consumption and wake-up costs. We develop a learning-augmented online algorithm that makes decisions based on (potentially inaccurate) predicted lengths of the idle periods. The algorithm's performance is near-optimal when predictions are accurate and degrades gracefully with increasing prediction error, with a worst-case guarantee almost identical to the optimal classical online algorithm for the problem. A key ingredient in our approach is a new algorithm for the online ski rental problem in the learning augmented setting with tight dependence on the prediction error. We support our theoretical findings with experiments.