Contribution to a conference proceedings/Contribution to a book

FZJ-2016-07827

http://join2-wiki.gsi.de/foswiki/pub/Main/Artwork/join2_logo100x88.png The Mont-Blanc Prototype: An Alternative Approach for HPC Systems

Rajovic, N. ; Rico, A. ; Mantovani, F. ; Ruiz, D. ; Vilarrubi, J. O. ; Gomez, C. ; Backes, L. ; Nieto, D. ; Servat, H. ; Martorell, X. ; Labarta, J. ; Ayguade, E. ; Adeniyi-Jones, C. ; Derradji, S. ; Gloaguen, H. ; Lanucara, P. ; Sanna, N. ; Mehaut, J.-F. ; Pouget, K. ; Videau, B. ; Boyer, E. ; Allalen, M. ; Auweter, A. ; Brayford, D. ; Tafani, D. ; Weinberg, V. ; Brömmel, D.FZJ* ; Halver, R.FZJ* ; Meinke, J.FZJ* ; Beivide, R. ; Benito, M. ; Vallejo, E. ; Valero, M. ; Ramirez, A.

2016
IEEE Press Piscataway, NJ, USA

This record in other databases:

Please use a persistent id in citations: doi:10.1109/SC.2016.37

Abstract: High-performance computing (HPC) is recognized as one of the pillars for further progress in science, industry, medicine, and education. Current HPC systems are being developed to overcome emerging architectural challenges in order to reach Exascale level of performance, projected for the year 2020. The much larger embedded and mobile market allows for rapid development of intellectual property (IP) blocks and provides more flexibility in designing an application-specific system-on-chip (SoC), in turn providing the possibility in balancing performance, energy-efficiency, and cost. In the Mont-Blanc project, we advocate for HPC systems being built from such commodity IP blocks, currently used in embedded and mobile SoCs.As a first demonstrator of such an approach, we present the Mont-Blanc prototype; the first HPC system built with commodity SoCs, memories, and network interface cards (NICs) from the embedded and mobile domain, and off-the-shelf HPC networking, storage, cooling, and integration solutions. We present the system's architecture and evaluate both performance and energy efficiency. Further, we compare the system's abilities against a production level supercomputer. At the end, we discuss parallel scalability and estimate the maximum scalability point of this approach across a set of applications.

---

Contributing Institute(s):

1. Jülich Supercomputing Center (JSC)

Research Program(s):

1. 511 - Computational Science and Mathematical Methods (POF3-511) (POF3-511)
2. MONT-BLANC 2 - Mont-Blanc 2, European scalable and power efficient HPC platform based onlow-power embedded technology (610402) (610402)
3. MONT-BLANC - Mont-Blanc, European scalable and power efficient HPC platform based on low-power embedded technology (288777) (288777)

Appears in the scientific report 2016

---