Enhanced Phase-Driven $Q$ -Learning-Based DRM for Multicore Processors

Authors:

Abstract
In this paper, we propose a new dynamic reliability management (DRM) technique for multi-core processors using phase-driven Q-learning based method. Our technique considers a wide range of long-term reliability issues and maximizes the throughput of the processor subject to the reliability constraint. We employ ON/OFF switching actions and dynamic voltage and frequency scaling (DVFS) as control knobs (i.e. working modes) to tune the state of cores of the processor. In order to achieve this, our technique detects program phases and adaptively determines the optimal working modes for each phase using the Q-learning based method. By integrating the phase detection into the Q-learning based management, our technique can provide efficient management for the programs with highly diverse phases. We also propose three additional modules to improve the management efficiency of our technique. In order to evaluate our technique, we use it to manage a 3D CPU with high-diver programs. Several failure mechanisms are considered in this case study. Our proposed technique is compared with two existing Q-learning based techniques Das:learning, Kim:learning. The experimental results demonstrate that when the number of phases is smaller than the number of working modes, our technique can achieve more than 1.36x improvement in performance with 60% memory space savings.

This publication has 36 references indexed in Scilit: