Abstract
Petascale supercomputers rely on highly efficient Petascale I/O subsystems. This work describes the tuning and scaling behavior of the GPFS parallel file system on JUGENE, the largest IBM Blue Gene/P installation worldwide and the first PetaFlop/s HPC resource within the European PRACE Research Infrastructure.
Article PDF
Similar content being viewed by others
References
IBM Blue Gene team (2008) Overview of the IBM Blue Gene/P project. IBM J Res Dev 52(1/2):199–220. Online at http://dx.doi.org/10.1147/rd.521.0199
Gara A, et al. (2005) Overview of the Blue Gene/L system architecture. IBM J Res Dev 49(2/3):213–248. Online at http://dx.doi.org/10.1147/rd.492.0195
Coteus P, et al. (2005) Packaging the Blue Gene/L supercomputer. IBM J Res Dev 49(2/3):213–248. Online at http://dx.doi.org/10.1147/rd.492.0195
Moreira J, et al. (2005) Blue Gene/L programming and operating environment. IBM J Res Dev 49(2/3):367–376
Mohr B, Frings W (2010) Jülich Blue Gene/P extreme scaling workshop 2009. Technical report FZJ-JSC-IB-2010-02. Online at http://www.fz-juelich.de/jsc/docs/printable/ib/ib-10/ib-2010-02.pdf
Mohr B, Frings W (2010) Jülich Blue Gene/P extreme scaling workshop 2010. Technical report FZJ-JSC-IB-2010-03. Online at http://www.fz-juelich.de/jsc/docs/printable/ib/ib-10/ib-2010-03.pdf
Schmuck F, Haskin R (2002) GPFS: A shared-disk file system for large computing clusters. In: Proceedings of the first USENIX conference on file and storage technologies. Monterey, CA, January 28–30, 2002, pp 231–244, 2002. Online at http://www.usenix.org/publications/library/proceedings/fast02/
Mextorf O, Schmidt U, Wollschläger L, Hennecke M, Kutzer K (2010) Storage and network design for the JUGENE Petaflop system. inSiDE 8(1):62–66 Online at http://inside.hlrs.de/htm/editions.htm
Vishwanath V, et al. (2010) Accelerating I/O Forwarding in IBM Blue Gene/P Systems. Online at http://www.mcs.anl.gov/uploads/cels/papers/P1745.pdf
Latham R (2008) Parallel I/O for high performance and scalability. In: 14th annual meeting of ScicomP, NY, 19–23 May 2008. Online at http://www.spscicomp.org/ScicomP14/talks/Latham.pdf
Parallel-NetCDF: a high performance API for NetCDF file access. Online at http://www.mcs.anl.gov/parallel-netcdf/
The HDF Group Parallel HDF5. Online at http://www.hdfgroup.org/HDF5/PHDF5/
Unidata. Parallel I/O with netCDF-4. Online at http://www.unidata.ucar.edu/software/netcdf/netcdf-4/
Gropp W, Huss-Lederman S, Lumsdaine A, Lusk E, Netzberg B, Saphir W, Snir M (1998) MPI: the complete reference. The MPI-2 extensions, vol 2. MIT-Press, Cambridge, p 10
SIONlib: scalable massively parallel I/O to task-local files. Online at http://www.fz-juelich.de/jsc/sionlib/
Frings W, Wolf F, Petkov V (2009) Scalable massively parallel I/O to task-local files. In: Proceedings of SC09, Portland, OR, USA, November 14–20. Online at http://dx.doi.org/10.1145/1654059.1654077
GPFS Version 3.3 (2009) Concepts, planning, and installation guide. IBM publication GA76-0413-03, September 2009
GPFS Version 3.3 (2009) Administration and programming reference. IBM publication SC23-2221-03, September 2009
GPFS Version 3.3 Advanced administration guide. IBM publication SC23-5182-03, September 2009
IBM developerWorks “HPC Central” Wiki. Online at http://www.ibm.com/developerworks/wikis/display/hpccentral/
Lakner G (2010) IBM system Blue Gene solution: Blue Gene/P system administration. Configuring I/O nodes. Chap 12. IBM redbook SG24-7417-03. Online at http://www.redbooks.ibm.com/abstracts/sg247417.html
Hennecke M (2006) GPFS multicluster with the IBM System Blue. Gene solution and eHPS clusters. IBM redpaper REDP4168. Online at http://www.redbooks.ibm.com/abstracts/redp4168.html
IOR HPC benchmark. Online at http://sourceforge.net/projects/ior-sio/
Shan H, Shalf J (2007) Using IOR to analyze the I/O performance for HPC Platforms. LBNL technical report. Online at http://escholarship.org/uc/item/9111c60j
Author information
Authors and Affiliations
Corresponding author
Additional information
This work was supported by the FZJ/IBM Exascale Innovation Center, EIC cooperation agreement T/Z1213.02.09.
∗IBM, Blue Gene and GPFS are trademarks of IBM in USA and/or other countries. Linux is a registered trademark of Linus Torvalds in the United States, other countries, or both.
Rights and permissions
Open Access This is an open access article distributed under the terms of the Creative Commons Attribution Noncommercial License (https://creativecommons.org/licenses/by-nc/2.0), which permits any noncommercial use, distribution, and reproduction in any medium, provided the original author(s) and source are credited.
About this article
Cite this article
Frings, W., Hennecke, M. A system level view of Petascale I/O on IBM Blue Gene/P. Comput Sci Res Dev 26, 275–283 (2011). https://doi.org/10.1007/s00450-011-0154-4
Published:
Issue Date:
DOI: https://doi.org/10.1007/s00450-011-0154-4