摘要
To improve the power consumption of parallel applications at the runtime, modern processors provide frequency scaling and power limiting capabilities. In this work, a runtime strategy is proposed to distribute a given power allocation among the cluster nodes assigned to the application while balancing their performance change. The strategy operates in a timeslice-based manner to estimate the current application performance and power usage per node followed by power redistribution across the nodes. Experiments, performed on four nodes (112 cores) of a modern computing platform interconnected with Infiniband showed that even a significant power budget reduction of 20% may result in a performance degradation of as low as 1% under the proposed strategy compared with the execution in the unlimited power case.
To improve the power consumption of parallel applications at the runtime, modern processors provide frequency scaling and power limiting capabilities. In this work, a runtime strategy is proposed to distribute a given power allocation among the cluster nodes assigned to the application while balancing their performance change. The strategy operates in a timeslice-based manner to estimate the current application performance and power usage per node followed by power redistribution across the nodes. Experiments, performed on four nodes (112 cores) of a modern computing platform interconnected with Infiniband showed that even a significant power budget reduction of 20% may result in a performance degradation of as low as 1% under the proposed strategy compared with the execution in the unlimited power case.
作者
Vaibhav Sundriyal
Masha Sosonkina
Vaibhav Sundriyal;Masha Sosonkina(Department of Computational Modeling and Simulation Engineering, Old Dominion University, Norfolk, Virginia, USA)
