Runtime Energy Minimization of Distributed Many-Core Systems using Transfer Learning

PROCEEDINGS OF THE 2022 DESIGN, AUTOMATION & TEST IN EUROPE CONFERENCE & EXHIBITION (DATE 2022)(2022)

引用 1|浏览9
暂无评分
摘要
The heterogeneity of computing resources continues to permeate into many-core systems making energy-efficiency a challenging objective. Existing rule-based and model-driven methods return sub-optimal energy-efficiency and limited scalability as system complexity increases to the domain of distributed systems. This is exacerbated further by dynamic variations of workloads and quality-of-service (QoS) demands. This work presents a QoS-aware runtime management method for energy minimization using a transfer learning (TL) driven exploration strategy. It enhances standard Q-learning to improve both learning speed and operational optimality (i.e., QoS and energy). The core to our approach is a multi-dimensional knowledge transfer across a task's state-action space. It accelerates the learning of dynamic voltage/frequency scaling (DVFS) control actions for tuning power/performance trade-offs. Firstly, the method identifies and transfers already learned policies between explored and behaviorally similar states referred to as Intra-Task Learning Transfer (ITLT). Secondly, if no similar “expert” states are available, it accelerates exploration at a local state's level through what's known as Intra-State Learning Transfer (ISLT). A comparative evaluation of the approach indicates faster and more balanced exploration. This is shown through energy savings ranging from 7.30% to 18.06%, and improved QoS from 10.43% to 14.3%, when compared to existing exploration strategies. This method is demonstrated under WordPress and TensorFlow workloads on a server cluster.
更多
查看译文
关键词
runtime energy minimization,distributed many-core systems,challenging objective,rule-based methods,return sub-optimal energy-efficiency,system complexity,distributed systems,quality-of-service demands,QoS-aware runtime management method,transfer learning driven exploration strategy,standard Q-learning,learning speed,operational optimality,multidimensional knowledge transfer,state-action space,learned policies,explored states,behaviorally similar states,similar expert states,local state,balanced exploration,energy savings,exploration strategies,intra-state learning transfer,intra-task learning transfer,transfer learning,sub-optimal energy-efficiency,quality-of-service,dynamic voltage-frequency scaling,DVFS control actions,WordPress,TensorFlow workloads,server cluster
AI 理解论文
溯源树
样例
生成溯源树,研究论文发展脉络
Chat Paper
正在生成论文摘要