<span class="var-sub_title">RGB (Redfish Green500 Benchmarker): A Green500 Benchmarking Tool Using Redfish</span> SC18 Proceedings

The International Conference for High Performance Computing, Networking, Storage, and Analysis

RGB (Redfish Green500 Benchmarker): A Green500 Benchmarking Tool Using Redfish

Authors: Elham Hojati (Texas Tech University), Yong Chen (Texas Tech University), Alan Sill (Texas Tech University), Jon Hass (Dell Inc)

Abstract: Performance and energy are important factors for supercomputers and data-centers with a trade-off between them. Energy efficiency metric considers both of these properties. The Green500 is a branch of Top500 project which provides a list of supercomputers based on energy efficiency. It has a manual methodology for this calculation.

Redfish is a new generation of management technologies for the hardware layer of data-centers. Our project focuses on designing and developing an automated Green500 benchmark tool using Redfish, called Redfish Green500 Benchmarker, or RGB in short. It offers the integration of Redfish and Green500, and automates Green500 benchmarking process with leveraging the internal capability of Redfish enabled equipment. It also enhances the Redfish standard to make sure it addresses the requirements of Green500 calculations. This research will also conduct validation and evaluation of RGB on real-world clusters for small-scale to medium-scale tests, and on the data-center simulator we have developed.

Best Poster Finalist (BP): no

Poster: pdf
Poster summary: PDF
Reproducibility Description Appendix: PDF

Back to Poster Archive Listing