Jianwei Li
Alok Choudhary
William Gropp
Rob Latham
ABSTRACT
Dataset storage, exchange, and access play a critical role in scientific applications. For such purposes netCDF serves as a portable, efficient file format and programming interface, which is popular in numerous scientific application domains. However, the original interface does not provide an efficient mechanism for parallel data storage and access. In this work, we present a new parallel interface for writing and reading netCDF datasets. This interface is derived with minimal changes from the serial netCDF interface but defines semantics for parallel access and is tailored for high performance. The underlying parallel I/O is achieved through MPI-IO, allowing for substantial performance gains through the use of collective I/O optimizations. We compare the implementation strategies and performance with HDF5. Our tests indicate programming convenience and significant I/O performance improvement with this parallel netCDF (PnetCDF) interface.
- {1} B. Fryxell, K. Olson, P. Ricker, F. X. Timmes, M. Zingale, D. Q. Lamb, P. MacNeice, R. Rosner, and H. Tufo. "FLASH: An Adaptive Mesh Hydrodynamics Code For Modelling Astrophysical Thermonuclear Flashes," Astrophysical Journal Suppliment, 2000, pp. 131-273.Google Scholar
- {2} W. Gropp, E. Lusk, N. Doss, and A. Skjellum. "A High-Performance, Portable Implementation of the MPI Message-Passing Interface Standard," Parallel Computing, 22(6):789- 828, 1996. Google ScholarDigital Library
- {3} W. Gropp, E. Lusk, and R. Thakur. Using MPI-2: Advanced Features of the Message Passing Interface, MIT Press, Cambridge, MA, 1999. Google ScholarDigital Library
- {4} HDF4 Home Page. The National Center for Supercomputing Applications. http://hdf.ncsa.uiuc.edu/hdf4.html.Google Scholar
- {5} HDF5 Home Page. The National Center for Supercomputing Applications. http://hdf.ncsa.uiuc.edu/HDF5/.Google Scholar
- {6} J. Li, W. Liao, A. Choudhary, and V. Taylor. "I/O Analysis and Optimization for an AMR Cosmology Application," in Proceedings of IEEE Cluster 2002, Chicago, September 2002. Google ScholarDigital Library
- {7} Message Passing Interface Forum. "MPI-2: Extensions to the Message-Passing Interface", July 1997. http://www.mpi-forum.org/docs/docs.html.Google Scholar
- {8} C. Zender. The NetCDF Operators (NCO). http:// nco.sourceforge.net/.Google Scholar
- {9} R. Rew, G. Davis, S. Emmerson, and H. Davies, "NetCDF User's Guide for C," Unidata Program Center, June 1997. http://www.unidata.ucar.edu/packages/netcdf/guidec/.Google Scholar
- {10} R. Rew and G. Davis, "The Unidata netCDF: Software for Scientific Data Access," Sixth International Conference on Interactive Information and Processing Systems for Meteorology, Oceanography and Hydrology, Anaheim, CA, February 1990.Google Scholar
- {11} R. Ross, D. Nurmi, A. Cheng, and M. Zingale, "A Case Study in Application I/O on Linux Clusters", in Proceedings of SC2001, Denver, November 2001. Google ScholarDigital Library
- {12} J.M. Rosario, R. Bordawekar, and A. Choudhary. "Improved Parallel I/O via a Two-Phase Run-time Access Strategy," IPPS'93 Parallel I/O Workshop, February 9, 1993.Google Scholar
- {13} F. Schmuck and R. Haskin. "GPFS: A Shared-Disk File System for Large Computing Clusters," in Proceedings of FAST'02, January 2002. Google ScholarDigital Library
- {14} R. Thakur, R. Bordawekar, A. Choudhary, R. Ponnusamy, and T. Singh. "PASSION Runtime Library for Parallel I/O", Scalable Parallel Libraries Conference, Oct. 1994.Google Scholar
- {15} R. Thakur and A. Choudhary. "An Extended Two-Phase Method for Accessing Sections of Out-of-Core Arrays," Scientific Programming, 5(4):301-317, Winter 1996. Google ScholarDigital Library
- {16} R. Thakur, W. Gropp, and E. Lusk. "An Abstract-Device interface for Implementing Portable Parallel-I/O Interfaces"(ADIO), in Proceedings of the 6th Symposium on the Frontiers of Massively Parallel Computation, October 1996, pp. 180-187. Google ScholarDigital Library
- {17} R. Thakur, W. Gropp, and E. Lusk. "Data Sieving and Collective I/O in ROMIO," in Proceeding of the 7th Symposium on the Frontiers of Massively Parallel Computation, February 1999, pp. 182-189. Google ScholarDigital Library
- {18} R. Thakur, W. Gropp, and E. Lusk. "On Implementing MPIIO Portably and with High Performance," in Proceedings of the Sixth Workshop on Input/Output in Parallel and Distributed Systems, May 1999, pp. 23-32. Google ScholarDigital Library
- {19} R. Thakur, R. Ross, E. Lusk, and W. Gropp, "Users Guide for ROMIO: A High-Performance, Portable MPI-IO Implementation," Technical Memorandum No. 234, Mathematics and Computer Science Division, Argonne National Laboratory, Revised January 2002.Google Scholar
- {20} M. Zingale. FLASH I/O benchmark. http://flash.uchicago. edu/~zingale/flash_benchmark_io/.Google Scholar
- {21} Where is NetCDF Used? Unidata Program Center. http:// www.unidata.ucar.edu/packages/netcdf/usage.html.Google Scholar
Recommendations
Implementing a Parallel NetCDF Interface for Seamless Remote I/O Using Multi-dimensional Data
High Performance Computing for Computational Science - VECPAR 2008Parallel netCDF supports parallel I/O operations for a view of data as a collection of self-describing, portable, and array-oriented objects that can be accessed through a simple interface. Its parallel I/O operations are realized with the help of an ...
Design and implementation of netCDF markup language (NcML) and its GML-based extension (NcML-GML)
The Network Common Data Form (netCDF) is one of the primary methods of self-documenting data storage and access in the international geosciences research and education community and beyond. NetCDF was designed for use in a networked environment. The ...
Supporting computational data model representation with high-performance I/O in parallel netCDF
HIPC '11: Proceedings of the 2011 18th International Conference on High Performance ComputingParallel computational scientific applications have been described by their computation and communication patterns. From a storage and I/O perspective, these applications can also be grouped into separate data models based on the way data is organized ...
Comments