An Energy-Aware Online Learning Framework for Resource Management in Heterogeneous Platforms
This addresses the problem of efficient resource management for mobile device developers, offering an incremental improvement over offline methods.
The paper tackles the challenge of optimizing energy consumption and response time in mobile platforms by proposing an online imitation learning framework that adapts control policies to new applications at runtime, achieving adaptation after executing less than 25% of instructions on a commercial platform with 16 benchmarks.
Mobile platforms must satisfy the contradictory requirements of fast response time and minimum energy consumption as a function of dynamically changing applications. To address this need, system-on-chips (SoC) that are at the heart of these devices provide a variety of control knobs, such as the number of active cores and their voltage/frequency levels. Controlling these knobs optimally at runtime is challenging for two reasons. First, the large configuration space prohibits exhaustive solutions. Second, control policies designed offline are at best sub-optimal since many potential new applications are unknown at design-time. We address these challenges by proposing an online imitation learning approach. Our key idea is to construct an offline policy and adapt it online to new applications to optimize a given metric (e.g., energy). The proposed methodology leverages the supervision enabled by power-performance models learned at runtime. We demonstrate its effectiveness on a commercial mobile platform with 16 diverse benchmarks. Our approach successfully adapts the control policy to an unknown application after executing less than 25% of its instructions.