Leveraging Application Characteristics for Energy Savings
Abstract
This chapter provides insights and studies into power draw and energy consumption of a variety of scientific applications. Multiple aspects are considered, including power monitoring, capping, and predicting. This allowed us to achieve improved scheduling in the studied use cases. From the standpoint of reducing the energy consumption, it was possible to regulate frequency and/or power in order to minimize the effects of performance degradation. This involved combining the essential application characteristics and fully leveraging them for controlled influence on performance/energy trade-offs. Some of the observed effects were counter-intuitive due to complex interactions between modern scientific applications and the advanced hardware platforms.
Fields of Study & Subjects
Parallel Computing and Optimization Techniques
Energy Efficiency in Computing
Distributed and Parallel Computing Systems
Publication details
- Journal / source
- CRC Press eBooks
- Pages
- 129–154
- Publication date
- 6 February 2026
- Language
- EN
- Document type
- Book chapter