It is now recognized that a high level of I/O performance is crucial in making effective use of parallel machines for many scientific application codes. This paper considers the I/O requirements in one particular scientific application area: three-dimensional modeling of continental shelf sea regions. The authors identify some of the scientific aims that drive the model development, and the consequent impact on the I/O needs. As a case study, the authors take a parallel production code running a simulation of the North Sea on a Cray T3D platform and investigate the I/O performance in dealing with the dominant I/O component: dumping of results data to disk. To place the performance issues in a more general framework, the authors construct a simple theoretical model of I/O requirements, and use this to probe the impact of available I/O performance on current and proposed scientific objectives.
Get full access to this article
View all access options for this article.
References
1.
Baretta, J.W., Ebenhoh, W., and Ruardij, P.1995. The European regional seas ecosystem model. A complex marine ecosystem modelNetherlands J. Sea Res.33(3-4): 233-246.
2.
Booth, S.1996. Lattice QCD simulation programs on the Cray T3D. EPCC-TR96-03Edinburgh Parallel Computing Centre, University of Edinburgh .
3.
Booth, S., and Paton, S.1996. I/O on the Cray T3D. EPCC-TR96-05, Edinburgh Parallel Computing Centre, University of Edinburgh.
4.
Bordawekar, R., del Rosario, J.M., and Choudhary, A.1993. Design and evaluation of primitives for parallel I/O. SCCS-564, North-east Parallel Architectures Center, Syracuse University.
5.
Colella, P., and Woodward, P.R.1984. The piecewise parabolic method (PPM) for gas-dynamical simulationsJ. Comput. Phys. 54:174-201.
6.
Corbett, P.F., Prost, J., Demetriou, C., Gibson, G., Riedel, E., Zelenka, J., Chen, Y., Felten, E., Li, K., Hartman, J., Peterson, L., Bershad, B., Wolman, A., and Aydt, R.1996. Proposal for a common parallel file system programming interface, version 1.0. Scalable I/O Initiative.
7.
Coward, A.C.1997. Experiences with a very high resolution global ocean model on a 512 processor CRAY-T3DPhysics and Chemistry of the Earth, in press.
8.
Cray Research Inc.1993. Advanced I/O users guideCray Research Publication SG-3076 8.0.
9.
Gwilliam, C.S., Coward, A.C., de Cuevas, B.A., Webb, D.J., and Rourke, E.1996. The OCCAM global ocean model: Current status and future plans within the UK Oceanographic Community In Proceedings of the First MMARIE Annual Meeting (EC Concerted Action) , Cagliari, January.
10.
Hack, J.J., Rosinski, J.M., Williamson, D.L., Boville, B.A., and Truesdale, J.E.1995. Computational design of the NCAR community climate modelParallel Computing21:1545-1569.
11.
Huthnance, J.M., Allen, J.I., Davies, A.M., Hydes, D.J., James, I.D., Jones, J.E., Millward, G.E., Prandle, D., Proctor, R., Purdie, D.A., Statham, P.J., Tett, P.B., Thomson, S., and Wood, R.G.1994. Towards water quality models In Understanding the North Sea system, edited by H. Charnock, K. R. Dyer, J. M. Huthnance, P. S. Liss, J. H. Simpson, and R B. Tett. London : Chapman and Hall, pp. 191-206.
Jones, P.W., Kerr, C.L., and Hemler, R.S.1995. Practical considerations in development of a parallel SKYHI general circulation modelParallel Computing21:1677-1694.
14.
MPI Forum.1995. MPI: A message passing interface standard, version 1.1University of Tennessee.
15.
MPI Forum.1997. MPI-2: Extensions to the message passing interfaceUniversity of Tennessee .
16.
Proctor, R., and James, I.D.1996. A fine-resolution 3D model of the southern North SeaJ. Marine Sys.8:285-295.
17.
Radach, G., Regener, M., Carlotti, F., Kühn, W., and Moll, A.1994. Modelling water column processes in the North Sea . In Understanding the North Sea system , editor by H. Charnock, K. R. Dyer, J. M. Huthnance, P. S. Liss, J. H. Simpson, and P. B. Tett. London: Chapman and Hall, pp. 131-140.
18.
Thakur, R., Gropp, W., and Lusk, E.1996a. An abstract-device interface for implementing portable parallel-I/O interfaces In Proceedings of the 6th Symposium on the Frontiers of Massively Parallel Computation , pp. 180-187.
19.
Thakur, R., Gropp, W., and Lusk, E.1996b. An experimental evaluation of the parallel I/O systems of the IBM SP and Intel Paragon using a production application In Proceedings of the 3rd International Conference of the Austrian Center for Parallel Computation (ACPC) (with special emphasis on parallel databases and parallel I/O), September.
20.
Thakur, R., Lusk, E., and Gropp, W.1995. I/O characterization of a portable astrophysics application on the IBM SP and Intel Paragon Technical Report MCS-P534-0895, Mathematics and Computer Science Division, Argonne National Laboratory .
21.
Wehner, M.F., Mirin, A.A., Eltgroth, P.G., Dannevik, W.P., Mechoso, C.R., Farrara, J.D., and Spahr, J.A.1995. Performance of a distributed memory finite difference atmospheric general circulation modelParallel Computing21:1655-1675.
