Flexible IO and integration for scientific codes through the adaptable IO system (ADIOS)

ABSTRACT

Scientific codes are all subject to variation in performance depending on the runtime platform and/or configuration, the output writing API employed, and the file system for output. Since changing the IO routines to match the optimal or desired configuration for a given system can be costly in terms of human time and machine resources, the Adaptable IO System provides an API nearly as simple as POSIX IO that also provides developers with the flexibility of selection the optimal IO routines for a given platform, without recompilation. As a side effect, we also gain the ability to transparently integrate more tightly with workflow systems like Kepler and Pegasus and visualization systems like Visit with no runtime impact. We achieve this through our library of highly tuned IO routines and other transport methods selected and configured in an XML file read only at startup. ADIOS-based IO has demonstrated high levels of performance and scalability. For example, we have achieved 20 GB/sec write performance using GTC on the Jaguar Cray XT4 system at Oak Ridge National Labs (about 50\% of peak performance). We can change GTC output among MPI-IO synchronous, MPI-IO collective, POSIX IO, no IO (for baseline testing), asynchronous IO using the Georgia Tech DataTap system, and Visit directly for in situ visualization with no changes to the source code. We designed this initial version of ADIOS based on the data requirements of 7 major scientific codes (GTC, Chimera, GTS, XGC1, XGC0, FLASH, and S3D) and have successfully adapted all of them to use ADIOS for all of their IO needs.

REFERENCES

• 1: Robert Ross , Rajeev Thakur , William Loewe , Robert Latham, Parallel I/O in practice, Proceedings of the 2006 ACM/IEEE conference on Supercomputing, November 11-17, 2006, Tampa, Florida [doi>10.1145/1188455.1188680]
• 2: H. FSIO, "http://institutes.lanl.gov/hecfsio/ docs/hec-fsio-fy07-gaps roadmap.pdf."
• 3: Bertram Ludäscher , Ilkay Altintas , Chad Berkley , Dan Higgins , Efrat Jaeger , Matthew Jones , Edward A. Lee , Jing Tao , Yang Zhao, Scientific workflow management and the Kepler system: Research Articles, Concurrency and Computation: Practice & Experience, v.18 n.10, p.1039-1065, August 2006 [doi>10.1002/cpe.v18:10]
• 4: Ewa Deelman , Gurmeet Singh , Mei-Hui Su , James Blythe , Yolanda Gil , Carl Kesselman , Gaurang Mehta , Karan Vahi , G. Bruce Berriman , John Good , Anastasia Laity , Joseph C. Jacob , Daniel S. Katz, Pegasus: A framework for mapping complex scientific workflows onto distributed systems, Scientific Programming, v.13 n.3, p.219-237, July 2005
• 5: SILO, "https://wci.llnl.gov/codes/visit/3rd party/silo.book.pdf."
• 6: P. J. Braam, "Lustre: a scalable high-performance file system," Nov. 2002. {Online}. Available: http://www.lustre.org/docs/whitepaper.pdf
• 7: R. Ross, R. Latham, N. Miller, and P. Carns, "A next-generation parallel file system for linux clusters," January 2004..
• 8: R. Latham, R. Ross, and R. Thakur, "The impact of file systems on mpi-io scalability," in Proceedings of EuroPVM/MPI 2004, September 2004.
• 9: R. Oldfield, L. Ward, R. Riesen, A. Maccabe, P. Widener, and T. Kordenbrock, "Lightweight i/o for scientific applications," Cluster Computing, 2006 IEEE International Conference on, pp. 1--11, 25-28 Sept. 2006.
• 10: A. A. V. System, "http://www.avs.com."
• 11: J. Ahrens, C. Law, W. Schroeder, K. Martin, and M. Papka, "A parallel approach for efficiently visualizing extremely large," 2000. {Online}. Available: citeseer.ist.psu.edu/ahrens00parallel.html
• 12: K. S. Hasan Abbasi, Matthew Wolf, "Live data workspace: A flexible, dynamic and extensible platform for petascale applications," in Cluster Computing. Austin, TX: IEEE International, September 2007.
• 13: S. O. G. T. Results, "http://users.nccs.gov/ oral/jagregtests/gtc128.html."
• 14: Visit, "http://www.llnl.gov/visit/home.html."

INDEX TERMS

Primary Classification:
  D. Software
  D.4 OPERATING SYSTEMS
      D.4.3 File Systems Management
          Subjects: Access methods

General Terms:
Design, Documentation, Experimentation, Human Factors, Management, Performance, Reliability, Standardization

Keywords:
HDF-5, MPI-IO, modular IO, visualization, workflow