Skip to main content
SpringerLink
Log in

Reference implementation of scalable I/O low-level API on Intel Paragon

  • Regular Papers
  • Published:
Journal of Computer Science and Technology Aims and scope Submit manuscript

Abstract

The Scalable I/O (SIO) Initiative’s Low-Level Application Programming Interface (SIO LLAPI) provides file system implementers with a simple low-Level interface to support high-level parallel I/O interfaces efficiently and effectively. This paper describes a reference implementation and the evaluation of the SIO LLAPI on the Intel Paragon multicomputer. The implementation provides the file system structure and striping algorithm, compatible with the Parallel File System (PFS) of Intel Paragon, and runs either inside the kernel or as a user level library. The scatter-gather addressing read/write, asynchronous I/O, client caching and prefetching mechanism, file access hint mechanism collective I/O and highly efficient file copy have been implemented. The preliminary experience shows that the SIO LLAPI provides opportunities of significant performance improvement and is easy to implement. Some high level file system interfaces and applications, such as PFS, ADIO and Hartree-Fock application, are also implemented on top of SIO. The performance of PFS is at least the same as that of Intel’s native PFS, and in many cases, such as small sequential file access, huge I/O requests and collective I/O, it is stable and much better. The SIO features help to support high level interfaces easily, quickly and more efficiently, and the cache, prefetching, hints are useful to get better performance based on different access models. The scalability and performance of SIO are limited by the network latency, network scalable bandwidth, memory copy bandwidth, memory size and pattern of I/O requests. The tradeoff between generality and efficiency should be considered in implementation.

This is a preview of subscription content, log in via an institution to check access.

Access this article

Log in via an institution

Price excludes VAT (USA)
Tax calculation will be finalised during checkout.

Instant access to the full article PDF.

Institutional subscriptions

Similar content being viewed by others

References

  1. Moyer S A, Sunderam V S. PIOUS: A scalable parallel I/O system for distributed computing environments. InScalable High-performance Computing Conference, May 1994.

  2. Thakur R, Bordawekar R, Choudhary A, Ponnusamy R, Singh T. PASSION runtime library for parallel I/O. InProceedings of Scalable Parallel Libraries, October 1994.

  3. The MPI-IO Committee. MPI-IO: A Parallel File I/O Interface for MPI. April, 1996. Available from http://parallel.nas.nasa.gov/MPI-IO

  4. Feitelson D G, Corbett P F, Prost J-P, Baylor S J. Parallel access to files in the Vesta file system. InProceedings of Supercomputing’93, Portland, Oregon, 1993, pp.472–481.

  5. Intel Supercomputing Systems Division. Paragon System User’s Guide, 1993.

  6. Scalable I/O Low Level I/O Committee. Proposal for a Common Parallel File System Programming Interface. November 1996. Available from http://www.pdl.cs.cmu.edu/index.html

  7. Roman Zajcew, Paul Roy, David Black, Chris Peaket al. An OSF/1 Unix for massively parallel multicomputers. InProceedings of Winter 1993 Usenix Technical Conference, San Diego, January 1993, pp.449–468.

  8. Avadis Tevanian Jr, Ben Smith. MACH: The model for future UNIX.BYTE, November 1989.

  9. Pierce P. The Paragon implementation of the NX message passing interface. InProceedings of Scalable High-performance Computing Conference (SHPCC), 1994.

  10. Rosario J M, Bordawekar R, Choudhary A. Improved parallel I/O via a two-phase run-time access strategy. InIPPS’93 Workshop on Input/Output in Parallel Computer Systems, 1993, pp.56–70.

  11. Thakur R, Gropp W, Lusk E. An abstract device interface for implementing portable parallel I/O interfaces. InProceedings of the 6th Symposiumon the Frontiers of Massively Parallel Computation, October 1996, pp.180–187.

  12. Chen Yuqun, Plank J S, Li Kai. CLIP: A checkpointing library for Intel Paragon. In theProceedings of SuperComputing’97, San Jose, California, November 1997.

  13. Pablo Project: Input/Output Characterization. Available from http://www-pablo.cs.uiuc.edu/Projects/IO/sioDir/apps.html

  14. Salem K, Garcia-Molina H. Disk striping. InThe 2nd IEEE Conference on Data Engineering, 1986.

  15. Patterson D A, Chen P, Gibson G, Katz R. Introduction to Redundant Arrays of Inexpensive Disks (RAID). InCOMPCON’89, Spring, 1989, pp.112–117.

  16. Chen P, Lee E, Gibson G, Katz R, Patterson D. RAID: High-performance, reliable secondary storage.ACM Computing Survey, June 1994, 26(2): 145–188.

    Article  Google Scholar 

  17. Katz R, Chen P, Drapeau Aet al. RAID-II: Design and implementation of a large scale disk array controller. Inthe Proceedings of Symposium on Integrated Systems, 1993.

  18. Cao P, Lim S, Venkataraman S, Wilkes J. The TickerTAIP parallel RAID architecture. InProceedings of the 20th Symposium on Computer Architecture, May 1993, pp.52–63.

  19. Riedal E, Ingen C V, Gray J. Sequential I/O on Windows NT 4.0 — Achieving Top Performance. Available from http://www.research.microsoft.com/barc, March 1998.

  20. Paul Pierce. A concurrent file system for a highly parallel mass storage subsystem. InProceedings of the Fourth Conference on Hypercubes, Concurrent Computers, and Applications, 1989, pp.155–160.

  21. Huber J, Elford C, Reed D, Chien A, Blumenthal D. PPFS: A high performance portable parallel file system. InProceedings of International Conference on Supercomputing (ICS)’95

  22. Thakur R, Lusk E, Gropp W. User Guide for ROMIO: A high-performance, portable MPI-IO implementation. Technical Memorandum, ANL/MCS-TM-234, Mathematics and Computer Science Division, Argonne National Laboratory, July 1998.

  23. Shillner R A, Felten E W. Simplifying distributed file systems using a shared logical disk. Technical Report TR-524-96, Princeton University, Department of Computer Science, 1996.

  24. Hartman J, Ousterhout J. The Zebra striped network file system.ACM Transactions on Computer Systems, August 1995.

  25. Anderson T, Dahlin M, Neefe Jet al. Serverless network file system. InProceedings of the 15th ACM Symposium on Operating Systems Principles (SOSP), December 1995, pp.109–126.

  26. Chatterjee S, Gilbert J, Schreiber R. Mobile and replicated alignments of arrays in data-parallel programs. InProceedings of Supercomputing’93, Portland, Oregon, pp.420–429.

  27. Bordawekar R, Choudhary A, Kennedy K, Koelbel C, Paleczny M. A model and compilation strategy for out-of-core data parallel programs. InProceedings of the Fifth ACM SIGPLAN Symposium on Principles and Practice of Parallel Programming (PPOPP), pp.1–10.

  28. Thakur R, Choudhary A, Bordawekar R, More S, Kuditipudi S, PASSION: Optimized I/O for parallel applications.IEEE Computer June 1996, 29(6): 70–78.

    Google Scholar 

  29. Thakur R, Bordawekar R, Choudhary A. Compiler and runtime support for out-of-core HPF programs. InProceedings of the 8th ACM International Conference on Supercomputing, July 1994, pp.382–391.

  30. Cao Pei, Felten E W, Karlin A R, Li Kai. Implementation and performance of integrated application-controlled file caching, Prefetching and Disk Scheduling. InACM Transactions on Computer Systems, November 1996, 14(4).

  31. Cao Pei, Felten E W, Karlin A R, Li Kai. A study of integrated prefetching and caching strategies. InProceedings of the 1995 ACM SIGMETRICS, 1995.

  32. Cao Pei, Felten E, Li Kai. Implementation and performance of application-controlled file caching policies. InProceedings of the First USENIX Symposium on Operating System Design and Implementation (OSDI), November 1994, pp.165–178.

  33. Kimbrel Tracy, Karlin A R. Near-optimal parallel prefetching and caching. InProceedings of 36th IEEE Symposium on Foundations of Computer Science (FOCS), 1996.

  34. Patterson R, Gibson G, Ginting E, Stodolsky D, Zelenka J. Informed prefetching and caching. InProceedings of the 15th ACM Symposium on Operating System Principles, Dec. 1995, pp.79–95.

  35. Kimbrel Tracy, Tomkins Andrew, Patterson R Hugoet al. A Trace-Driven Comparison of Algorithms for Parallel Prefetching and Caching. Technical Report 96-09-01, University of Washington, 1996.

Download references

Author information

Authors and Affiliations

  1. National Research Center for Intelligent Computing systems, Institute of Computing Technology, Chinese Academy of Sciences, 100080, Beijing, P.R. China

    Sun Ninghui

Authors
  1. Sun Ninghui
    View author publications

    You can also search for this author in PubMed Google Scholar

Additional information

Sun Ninghui is an Associate Professor of Institute of Computing Technology, Chinese Academy of Sciences. He graduated in 1989 from Beijing University and got his M.S. degree in 1992 from Institute of Computing Technology, Chinese Academy of Sciences. His major research area is parallel processing, architecture and system software of scalable system and scalable I/O.

Rights and permissions

Reprints and permissions

About this article

Cite this article

Sun, N. Reference implementation of scalable I/O low-level API on Intel Paragon. J. of Comput. Sci. & Technol. 14, 206–223 (1999). https://doi.org/10.1007/BF02948509

Download citation

  • Received:

  • Revised:

  • Issue Date:

  • DOI: https://doi.org/10.1007/BF02948509

Keywords

Search

Navigation