The HPCG Benchmark: Getting and Interpreting Performance from a New Metric for HPC Systems
The High Performance Conjugate Gradients (HPCG) Benchmark is emerging as a new community metric for ranking high performance computing systems. The first list of results was released at ISC'14, including optimized results for systems built upon Fujitsu, Intel, Nvidia technologies.
In this BOF we first present the organization of HPCG and opportunities for optimizing performance and follow with presentations from vendors and leadership computing facilities who have participated in HPCG optimization efforts. We spend the remaining time in open discussion about the future of HPCG design and implementation strategies for further improvements.
Presentations:
- Jack Dongarra, Michael Heroux, Piotr Luszczek, Toward a New (or Another) Metric for Ranking High Performance Computing Systems
- Jack Dongarra, Michael Heroux, Piotr Luszczek, HPCG Winners and Results Analysis
- Yutong Lu, HPCG on Tianhe2
- Jongsoo Park, HPCG Results on IA
- Kiyoshi Kumahata, HPCG Performance Improvement on the K computer
- Massimiliano Fatica, A CUDA Implementation of the High Performance Conjugate Gradient (HPCG) Benchmark
Link to the official HPCG BOF at SC14
