Abstract

The evolution of high-performance computing is associated with the growth of energy consumption. Performance of cluster computes (CC) is increased via rise in performance and the number of used processors, GPUs and coprocessors. Increment in the number of computing elements results significant growth of energy consumption. Power grids limits for supercomputer centers (SCC) are driving the transition to more energy efficient solutions. Often upgrade of computing resources is done step-by-step, i.e. parts of older supercomputers are removed from service and replaced with newer ones. A single SCC at any time can operate several computing systems with different performance and power consumption. That’s why the problem of scheduling parallel programs execution on SCC resources to optimize energy consumption and minimize the increase in execution time (energy efficient scheduling) is important. An algorithm for energy efficient scheduling for a set of CCs was implemented. Experiments on NPB benchmarks allowed to achieve significant reduction in energy consumption with minor increase of runtime.

中文翻译：

---

一种用于共享设施超级计算机中心的节能调度算法

摘要

高性能计算的发展与能源消耗的增长有关。集群计算 (CC) 的性能通过提高性能以及使用的处理器、GPU 和协处理器的数量而提高。计算单元数量的增加导致能耗的显着增长。超级计算机中心 (SCC) 的电网限制正在推动向更节能的解决方案过渡。计算资源的升级通常是逐步完成的，即旧的超级计算机的一部分将停止使用并替换为新的。单个 SCC 在任何时候都可以运行多个性能和功耗不同的计算系统。这就是为什么在 SCC 资源上调度并行程序执行以优化能源消耗和最小化执行时间增加的问题（节能调度）很重要。实现了一组 CC 的节能调度算法。NPB 基准测试允许在运行时间略有增加的情况下显着降低能耗。