This dataset supports the article entitled "Workload-Aware Runtime Energy Management for HPC Systems" accepted for publication 
in International Workshop on Optimization of Energy Efficient HPC & Distributed Systems (OPTIM), July 16-20, 2018, Orleans, France.
============================================================================

Person responsible for collecting the data: Karunakar Reddy Basireddy, krb1g15@ecs.soton.ac.uk

============================================================================

Date of data collection: 20/02/2018 to 21/03/2018

============================================================================

Dataset DOI: 10.5258/SOTON/D0517

============================================================================

Licenses/restrictions placed on the data: CC-BY/Public - No restriction

============================================================================

Links to publications that cite or use the data: xxxxx

============================================================================

Data Supporting Figures:

Fig. 3 - Comparison of the proposed technique with reported approaches for single application scenario in terms of normalized energy consumption, executing on the Xeon E5-2630.

Fig. 4 - Normalized energy consumption of various approaches for double application scenario, executing on the Xeon E5-2630.

Fig. 5 - Comparison of the proposed approach with reported approaches for triple application scenario in terms of normalized energy consumption (evaluated on the Xeon E5-2630).

Fig. 6 - Mean performance difference between prop-NUMA and other approaches, with standard deviation error bars (evaluated on the Xeon E5-2630).

Fig. 7 - Comparison of proposed approach (prop-NNUMA) with reported approaches in terms of normalized energy consumption for various application scenarios, executing on the Xeon Phi.

Fig. 8 - Mean performance difference between prop-NUMA and other approaches, with standard deviation error bars (evaluated on the Xeon Phi).

============================================================================