Paca: a Distributed File System Cache for Parallel Machines. Performance under Unix-like Workload

Parallel computers have a great computing power while I/O performance is far behind and it is, very often, the machine bottleneck. In this paper we tackle this problem presenting PACA, a cache directed le system aimed at parallel machines which run both Unix-like and parallel applications. PACA is a spe-ciic cooperative caching mechanism that de-nes a parallel global cache from all nodes local caches. Simplicity, scalability and performance are the main objectives in the design. We also present performance gures under Unix-like workload and compare them with other parallel/distributed caches already implemented .

[1]  Anoop Gupta,et al.  The directory-based cache coherence protocol for the DASH multiprocessor , 1990, ISCA '90.

[2]  David Kotz,et al.  Prefetching and caching techniques in file systems for MIMD multiprocessors , 1991 .

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

[4]  M. Franklin,et al.  Global Memory Management in Client-Server DBMS Architectures , 1992 .

[5]  John H. Hartman,et al.  Zebra: A Striped Network File System , 1992 .

[6]  A. Choudhary,et al.  Design and Evaluation of primitives for Parallel I/O , 1993, Supercomputing '93.

[7]  Philip S. Yu,et al.  Replication Algorithms in a Remote Caching Architecture , 1993, IEEE Trans. Parallel Distributed Syst..

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

[9]  Jesús Labarta,et al.  The Paros Operating System Microkernel , 1994 .

[10]  Steven A. Moyer,et al.  PIOUS: a scalable parallel I/O system for distributed computing environments , 1994, Proceedings of IEEE Scalable High Performance Computing Conference.

[11]  Michael Dahlin,et al.  Cooperative caching: using remote client memory to improve file system performance , 1994, OSDI '94.

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

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

[14]  Michael Gerndt,et al.  Intel Paragon XP/S - Architecture, Software Environment, and Performance , 1994 .

[15]  David Kotz.  Expanding the potential for disk-directed I/O , 1995, Proceedings.Seventh IEEE Symposium on Parallel and Distributed Processing.

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

[17]  Markus Ast,et al.  A general approach for an automatic parallelization applied to the finite element code PERMAS , 1995, HPCN Europe.

[18]  Evangelos P. Markatos,et al.  Implementation and Evaluation of a Remote Memory Pager , 1995 .

[19]  Toni Cortes,et al.  A Parallel Program Development Environment a Parallel Program Development Environment , 1995 .