PRACE

PRACE: Partnership for Advanced Computing in Europe

Our lab has participated actively in all the different phases of the PRACE project (PRACE-PP, PRACE-1IP, PRACE-2IP, PRACE-3IP). PRACE’s mission is to enable high impact scientific discovery and engineering research and development across all disciplines to enhance European competitiveness for the benefit of society. PRACE seeks to realize this mission by offering world class computing and data management resources and services through a peer review process. Furthermore, PRACE seeks to strengthen the European users of HPC in industry through various initiatives and has a strong interest in improving energy efficiency of computing systems and reducing their environmental impact.

Our work in the context of PRACE ranges from delivering top quality user guides for HPC systems to porting, studying and optimizing HPC applications.

PRACE Whitepapers

  • D. Siakavaras, K. Nikas, N. Anastopoulos, G. Goumas, “NUMA-CIC: Issues and Challenges for Scaling Scientific Applications on a Large Scale ccNuma Prototype.” (pdf)
  • I. E. Venetis, G. Goumas, M. Geveler, D. Ribbrock, “Porting FEASTFLOW to the Intel Xeon Phi: Lessons Learned.” (pdf)
  • K. Nikas, D. Siakavaras, V. Karakasis, J. C. Meyer, L. Natvig, “An Energy-centric Study of Conjugate Gradient Method.” (pdf)
  • J. C. Meyer, L. Natvig, V. Karakasis, D. Siakavaras, K. Nikas, “Energy-efficient Sparse Matrix Auto-tuning with CSX.” (pdf)
  • V. Karakasis, G. Goumas, K. Nikas, N. Koziris, J. Ruokolainen, P. Råback, “Using State-Of-The-Art Sparse Matrix Optimizations for Accelerating the Performance of Elmer.” (pdf)

Best Practice Guides

  • Best Practice Guide – Curie  (November 2013 — pdf)
  • Best Practice Guide – SuperMUC (May 2013 — pdf)
  • Best Practice Guide – Generic x86 (May 2013 — pdf)