Low-Level Interfaces for High-Level Parallel I/O

As the I/O needs of parallel scientific applications increase, file systems for multiprocessors are being designed to provide applications with parallel access to multiple disks. Many parallel file systems present applications with a conventional Unix-like interface that allows the application to access multiple disks transparently. By tracing all the activity of a parallel file system in a production, scientific computing environment, we show that many applications exhibit highly regular, but non-consecutive I/O access patterns. Since the conventional interface does not provide an efficient method of describing these patterns, we present three extensions to the interface that support strided, nested-strided, and nested-batched I/O requests. We show how these extensions can be used to express common access patterns.

[1]  E. DeBenedictis,et al.  nCUBE parallel I/O software , 1992, Eleventh Annual International Phoenix Conference on Computers and Communication [1992 Conference Proceedings].

[2]  Mary Baker,et al.  Measurements of a distributed file system , 1991, SOSP '91.

[3]  David J. DeWitt,et al.  SPIFFI-A Scalable Parallel File System for the Intel Paragon , 1996, IEEE Trans. Parallel Distributed Syst..

[4]  Samuel J. Leffler,et al.  The design and implementation of the 4.3 BSD Unix operating system , 1991, Addison-Wesley series in computer science.

[5]  David Kotz,et al.  Dynamic file-access characteristics of a production parallel scientific workload , 1994, Proceedings of Supercomputing '94.

[6]  Carla Schlatter Ellis,et al.  Characterizing parallel file-access patterns on a large-scale multiprocessor , 1995, IPPS.

[7]  Dror G. Feitelson,et al.  Mpi-io: a parallel file i/o interface for mpi , 1995 .

[8]  Dror G. Feitelson,et al.  Design and implementation of the Vesta parallel file system , 1994, Proceedings of IEEE Scalable High Performance Computing Conference.

[9]  John A. Kunze,et al.  A trace-driven analysis of the UNIX 4.2 BSD file system , 1985, SOSP '85.

[10]  John Kunze,et al.  A trace-driven analysis of the unix 4 , 1985, SOSP 1985.

[11]  Armando P. Stettner The design and implementation of the 4.3BSD UNIX operating system , 1988 .

[12]  Dror G. Feitelson,et al.  Overview of the Vesta parallel file system , 1993, CARN.

[13]  Carla Schlatter Ellis,et al.  File-Access Characteristics of Parallel Scientific Workloads , 1996, IEEE Trans. Parallel Distributed Syst..

[14]  Randy H. Katz,et al.  Input/output behavior of supercomputing applications , 1991, Proceedings of the 1991 ACM/IEEE Conference on Supercomputing (Supercomputing '91).

[15]  Richard Wheeler,et al.  it/sfs: A Parallel File System for the CM-5 , 1993, USENIX Summer.

[16]  David Kotz,et al.  Disk-directed I/O for MIMD multiprocessors , 1994, OSDI '94.

[17]  Michael L. Best,et al.  CMMD I/O: a parallel Unix I/O , 1993, [1993] Proceedings Seventh International Parallel Processing Symposium.

[18]  Dror G. Feitelson,et al.  Parallel I/O Systems and Interfaces for Parallel Computers , 1995 .

[19]  David Kotz,et al.  Performance of the Galley Parallel File System , 1996 .