Measuring sustainability and performance in hybrid computing systems
Measuring Sustainability in Multi-Scale High-Performance Computing
Distributed, Parallel, and Cluster Computing
Summary
Modern computing uses different kinds of computers working together, which can be tricky to manage efficiently. This paper by the authors introduces a new way to measure how well these systems perform while saving energy and keeping accuracy. They tested their method on a hybrid system and found interesting trade-offs between accuracy and energy use. Their approach helps find the best balance for running demanding tasks like artificial intelligence or quantum computing. This can guide future software that decides where to run different tasks to save energy and improve speed.
What this means in practice
- •For cloud infrastructure teams: Optimize scheduling to assign demanding workloads to the best nodes for energy-efficient performance.
- •For data center operators: Identify operating points that balance power consumption and computational accuracy to reduce overall energy costs.
Authors
Carlos J Barrios, Frédéric Le Mouël, Yves Denneulin
Abstract
The transition from traditional High Performance Computing (HPC) to the Computing Continuum emphasizes efficient resource management and sustainable practices across Multi-Scale hybrid architectures. This paper introduces a multidimensional metric framework to characterize these systems and guide deployment strategies for modern workloads. The framework combines Architectural Performance metrics (such as Throughput, Latency, Scalability), System Utilization, and key Sustainability and Accuracy indicators (such as Energy Efficiency and Power Consumption). Using a modular hybrid testbed, experiments reveal complex relationships among metrics, especially the trade-offs between accuracy and energy, and the efficiency of hybrid nodes. The guidelines help identify optimal operating points and lay the groundwork for improving orchestrators and schedulers (e.g., Kubernetes) to assign demanding applications, including AI and Quantum Computing, to suitable system modules, ensuring high performance and sustainability.