Abstract
In this paper, we introduce a distributed shared-memory system with a high speed serial communication interface called STAFF-Link. STAFF-Link is adopted in massively parallel computer JUMP-1 as I/O links between processing elements and an I/O subsystem consisting of many I/O units as well as an I/O network among I/O units. In this project, we have designed and manufacutured a STAFF-Link interface and a router board, however the network interface is able to be utilized for a high performance workstation cluster. In the workstation cluster, we have implemented a distributed shared-memory system based on a software controlled cache. We introduce the DSM system of the workstation cluster, and then we insist that high speed serial communication with many links is one of the solutions for future parallel processing environments.
This is a preview of subscription content, log in via an institution to check access.
Preview
Unable to display preview. Download preview PDF.
References
V. S. Sunderam, “PVM: A Framework for Parallel Distributed Computing,” Concurrency: Practice and Experience, Vol. 2, Number 4, pp.315–339 (1990).
Message Passing Interface Forum, “MPI: A Message-Passing Interface Standard”, International Journal of Supercomputer Applications and High Performance Computing, Vol. 8, Number 3/4, pp.159–416 (1994).
Craig C. Douglas, Timothy G. Mattson and Martin H. Schultz, “Parallel Programming Systems for Workstation Clusters,” Technical Report TR-975, Yale University Department of Computer Science Research (1993).
K. Li and P. Hudak, “Memory Coherency in Shared Virtual Memory Systems”, ACM Trans. Comput. Syst., Vol.7, No.4, pp.321–359 (1989).
Pete Keleher, Sandhya Dwarkadas, Alan L.Cox, and Willy Zwaenepoel, “Tread-Marks: Distributed Shared Memory on Standard Workstations and Operating Systems”, Rice COMP TR93-214, (1993).
Mark D. Hill, James R. Larus and David A. Wook, “Tempest: A substrate for Portable Parallel Programs.”, Proc. of COMPCON'95, pp.327–332 (1995).
H. Nakajo, K. Kuramae, Y. Kaneda and S. Maekawa, “The Implementation and Evaluation of Software Distributed Shared-Memory (DSM) for Workstation Clusters”, Trans. IPS Japan, Vol.36, No.7, pp.1719–1728 (1995) (in Japanese).
K. Hiraki et al., Overview of the JUMP-1 an MPP Prototype for General Purpose Parallel Computations, Proc. of ISPAN, pp.427–434 (1994).
H. Nakajo, T. Matsumoto, M. Kohata, H. Matsuda, K. Hiraki and Y. Kaneda, “High Performance I/O System of the Distributed Shared-Memory Massively Parallel Computer JUMP-1”, Proc. of 7th IASTED-ISMM Int. Conf. on Parallel and Distributed Computing and Systems, pp.470–473 (1995).
Hidehiko Tanaka (Chief Editor), “The Massively Parallel Processing System JUMP-1”, Ohmsha & IOS Press (1996).
H. Nakajo, S. Ohtani, Y. Kaneda, “A Simulation-based Evaluation of a Disk I/O Subsystem for a Massively Parallel Computer: JUMP-1”, Proc. of 16th International Conference on Distributed Computing Systems (ICDCS16), pp.562–569 (1996).
H. Nakajo, S. Ohtani, T. Matsumoto, M. Kohata, K. Hiraki and Y. Kaneda, “An I/O Network Architecture of the Distributed Shared-Memory Massively Parallel Computer JUMP-1,” Proc. of 11th Int. Conf. on Supercomputing (ICS97), pp.253–260 (1997).
Advanced Micro Devices, Inc, Am7968 / Am7969-175 TAXI-175 Transmitter / Receiver Data Sheet and Technical Manual (1992).
H. Nakajo, A. Ichikawa and Y. Kaneda, “An Implementation and Evaluation of a Distributed Shared-Memory System on Workstation Clusters using Fast Serial Links”, Proc. of Int. Symp. on High Performance Computing (ICHPC), pp.143–158 (1997).
Daniel E. Lenoski and Wolf-Dietrich Weber, “Scalable Shared-Memory Multiprocessing”, Morgan Kaufmann Publishers (1995).
Hironori Nakajo, Takeshi Yoshinaga, Koichi Wada and Yukio Kaneda, “Ring-Connected Parallel Computer KORP — Coherence Protocol for Distributed Shared-Memory —, Proc. of Int. Conf. on Parallel and Distributed Systems (ICPADS'92), pp.504–511 (1992).
N. J. Boden, D. Cohen, R. E. Felderman, A. E. Kulawik, C. L. Seitz, J. N. Seizovic and W. K. Su, “Myrinet: A Gigabit-per Second Local Area Network.”, IEEE Micro, 15(1), pp.29–36 (1995).
Author information
Authors and Affiliations
Editor information
Rights and permissions
Copyright information
© 1998 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Nakajo, H., Tanaka, H., Nakanishi, Y., Kohata, M., Kaneda, Y. (1998). Distributed shared-memory for a workstation cluster with a high speed serial interface. In: Sloot, P., Bubak, M., Hertzberger, B. (eds) High-Performance Computing and Networking. HPCN-Europe 1998. Lecture Notes in Computer Science, vol 1401. Springer, Berlin, Heidelberg. https://doi.org/10.1007/BFb0037186
Download citation
DOI: https://doi.org/10.1007/BFb0037186
Published:
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-540-64443-9
Online ISBN: 978-3-540-69783-1
eBook Packages: Springer Book Archive