Scholarlink

In this article we demonstrate performance-energy optimization of multi-GPU genetic algorithm execution using power capping. Firstly, we outline elements concerning the design and implementation of a multi-GPU framework for the execution of a genetic algorithm, allowing the application of a solution to a variety of problems. Secondly, the implementation of three algorithms is proposed and discussed: Traveling Salesman, Knapsack, and Partition. Then, we present a testbed environment with a high-performance computing node with 2 multi-core Intel Xeon CPUs and 8 NVIDIA Quadro RTX 6000 GPUs as well as a Yokogawa WT-310E power meter. Finally, we describe and discuss the optimization results of the implementations using 1, 2, 4, and 8 GPUs under different power caps imposed by NVIDIA NVML. We show the scalability of the solution in terms of fitness versus the number of GPUs used and analyze execution times and energy consumption of various cases under various power caps. We demonstrate that for 8 GPUs, using the power cap of 140 W per GPU, we can obtain considerable energy savings of over 17.93% for Traveling Salesman, 15.88% for Knapsack, 21.97% for Partition, with small increases of execution time: 0.89% for Traveling Salesman, 1.41% for Knapsack, and 14.64% for Partition, versus the results for the default power cap of 260 W per GPU.

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Performance-Energy Investigation of Selected Applications Using a Parallel Multi-GPU Genetic Algorithm Under Power Capping

  • 会议论文

  • 发表日期：

    2025年

查看全文
  • Filip Magdziak,
  • Paweł Czarnul

摘要

In this article we demonstrate performance-energy optimization of multi-GPU genetic algorithm execution using power capping. Firstly, we outline elements concerning the design and implementation of a multi-GPU framework for the execution of a genetic algorithm, allowing the application of a solution to a variety of problems. Secondly, the implementation of three algorithms is proposed and discussed: Traveling Salesman, Knapsack, and Partition. Then, we present a testbed environment with a high-performance computing node with 2 multi-core Intel Xeon CPUs and 8 NVIDIA Quadro RTX 6000 GPUs as well as a Yokogawa WT-310E power meter. Finally, we describe and discuss the optimization results of the implementations using 1, 2, 4, and 8 GPUs under different power caps imposed by NVIDIA NVML. We show the scalability of the solution in terms of fitness versus the number of GPUs used and analyze execution times and energy consumption of various cases under various power caps. We demonstrate that for 8 GPUs, using the power cap of 140 W per GPU, we can obtain considerable energy savings of over 17.93% for Traveling Salesman, 15.88% for Knapsack, 21.97% for Partition, with small increases of execution time: 0.89% for Traveling Salesman, 1.41% for Knapsack, and 14.64% for Partition, versus the results for the default power cap of 260 W per GPU.