A survey of commercial parallel processors

ABSTRACT

This paper compares eight commercial parallel processors along several dimensions. The processors include four shared-bus multiprocessors (the Encore Multimax, the Sequent Balance system, the Alliant FX series, and the ELXSI System 6400) and four network multiprocessors (the BBN Butterfly, the NCUBE, the Intel iPSC/2, and the FPS T Series). The paper contrasts the computers from the standpoint of interconnection structures, memory configurations, and interprocessor communication. Also, the shared-bus multiprocessors are compared in terms of cache-coherence strategies, and the network multiprocessors are compared in terms of node structure. Where possible, price and performance information has been included. The reader is cautioned that this survey is based largely on information submitted by manufacturers; the authors have not performed any independent evaluation.

REFERENCES

• 1: [Al 10/86] "FX/Series Product Summary," Alliant Computer Systems, Acton, MA, October, 1986.
• 2: [Al 85] Almasi, G. S., "Overview of Parallel Processing," Parallel Computing, 1985, pp. 191-203.
• 3: [BB 3/86] BBN Laboratories, "Butterfly Parallel Processor Overview," BBN Report no. 6148, BBN Laboratories, Inc., Version 1, March 6, 1986.
• 4: [BB 87] BBN Advanced Computers, Inc. "The Butterfly GP 1000 Parallel Processor," BBN Advanced Computers, Inc.
• 5: Bob Beck , Bob Kasten , Shreekant Thakkar, VLSI assist for a multiprocessor, Proceedings of the second international conference on Architectual support for programming languages and operating systems, p.10-20, October 1987, Palo Alto, California, United States
• 6: [Bi 7/87] Billig, Richard R. "A fast backplane cluster heralds a 1000-MIPS computer," Electronic Design, July, 1987, pp. 81-66.
• 7: [Br 4/87] Brooks, Eugene D. III, "A Butterfly processor-memory interconnection for a vector processing environment," Parallel Computing, April, 1987, pp. 103-110.
• 8: Jack J. Dongarra, Performance of various computers using standard linear equations software in a FORTRAN environment, ACM SIGARCH Computer Architecture News, v.16 n.1, p.47-69, March, 1988 [doi>10.1145/44571.44576]
• 9: [El 4/86] "Vector processing boosts hypercube's performance," Electronics, April, 1986, pp. 30-31.
• 10: [FP 11/87a] Floating Point Systems, "The FPST series-a parallel vector supercomputer," FPS, Beaverton, OR, Nov., 1987.
• 11: [FP 11/87b] FPS, "Analysis of Fortran and C performance on the FPS T Series parallel supercomputer," FPS, Inc. Beaverton, OR, Nov. 1987.
• 12: Karen A. Frenkel, Evaluating two massively parallel machines, Communications of the ACM, v.29 n.8, p.752-758, Aug. 1986 [doi>10.1145/6424.6427]
• 13: John P Hayes , Trevor Mudge , Quentin F Stout , Stephen Colley , John Palmer, A microprocessor-based hypercube supercomputer, IEEE Micro, v.6 n.5, p.6-17, Oct. 1986 [doi>10.1109/MM.1986.304707]
• 14: [HT 2/87] High Technology Business magazine, "Representative parallel/multiprecessing computers," Feb. 1987.
• 15: [In 9/87] "Direct-Connect^¿ routing solves node communications challenge," iSCurrents, Fall-Winter 1987, Intel Scientific Computers, Beaverton, OR, pp. 5-6.
• 16: [In 10/87] Intel Scientific Computers, "The INTEL iPSC/2 System product-information", Intel Scientific Computers, Beaverton, OR.
• 17: [In 88] Intel Scientific Computers," iPSC (brochure), Intel Scientific Computers, Beaverton, OR.
• 18: [LT 12/87] Lovett, Tom and Thakkar, Shreekant, "The Symmetry multiprocessor system," Sequent Computer Systems, Inc., to be published.
• 19: [Ju 6/86] Jurasek, David, "Microprocessor design in custom VLSI", VLSI Systems Design, June 1986, pp. 26-30.
• 20: [Ma 2/87] Martin, W. R. "Monte Carlo Photon transport on shared memory and distributed memory parallel processors", University of Michigan, Feb. 1987.
• 21: [NC 88] NCUBE Corporation, "NCUBE/10, an overview" (brochure), NCUBE, Beaverton, OR.
• 22: [Ol 85] Olson, Robert, "Parallel Processing in a message-based operating system," IEEE Software, 1985, pp. 39-49.
• 23: [Ol 5/86] Olson, Robert, "Real-time response on a message-based multiprocessor," IEEE Software, May 1986, pp. 28-35.
• 24: [Pa 5/86] Palmer, John F., "The VLSI parallel computer," Proe. COMPCON, May 1986, pp. 397-401.
• 25: [Pe 5/86] Perron, Robert, "The architecture of the Alliant FX/8 computer," Proc. COMPCON , May, 1986, pp. 390-396.
• 26: Mark S. Papamarcos , Janak H. Patel, A low-overhead coherence solution for multiprocessors with private cache memories, Proceedings of the 11th annual international symposium on Computer architecture, p.348-354, January 1984
• 27: Randall Rettberg , Robert Thomas, Contention is no obstacle to shared-memory multiprocessing, Communications of the ACM, v.29 n.12, p.1202-1212, Dec. 1986 [doi>10.1145/7902.7905]
• 28: [Re -] Reeves, Anthony P., "Parallel Pascal and the FPS Hypercube supercomputer."
• 29: John Sanguinetti, Performance of a Message-Based Multiprocessor, Computer, v.19 n.9, p.47-55, September 1986 [doi>10.1109/MC.1986.1663330]
• 30: David J. Schanin, The design and development of a very high speed system bus—the encore Mutlimax nanobus, Proceedings of 1986 ACM Fall joint computer conference, p.410-418, November 1986, Dallas, Texas, United States
• 31: [SC 87] Scott, Michael and Cox, Alan, "An empirical study of message-passing overhead," Proceedings of the 7th International Conference on Distributed Computing Systems, Berlin, Sept. 21-25, 1987, pp. 536- 543.
• 32: [Se 11/86] "Balance Technical Summary," Sequent Computer Systems, Inc., Nov. 1986.
• 33: [Sh 2/87] Shar, Leonard E., "Designing a multiple processor environment," Proc. COMPCON, February, 1987, pp. 110-113.
• 34: [St 6/87] Steinberg, Jeffrey A., "A new twist: vectors in parallel," Digital Review, June 29, 1987, pp. 63-66.
• 35: [Te -] Test, Jack A., "Multiprocessor Management in the Concentrix operating system," Alliant Computer Systems Corporation, Acton, MA, pp. 35-43.
• 36: Shreekant Thakkar , Paul Gifford , Gary Fielland, The Balance Multiprocessor System, IEEE Micro, v.8 n.1, p.57-69, January 1988 [doi>10.1109/40.521]
• 37: [TM -] Test, Jack A., Myszewski, M., Swift, Richard C., "The Alliant FX/Series: Automatic parallelism in a multiprocessor mini-supercomputer."
• 38: Lewis W. Tucker , George G. Robertson, Architecture and Applications of the Connection Machine, Computer, v.21 n.8, p.26-38, August 1988 [doi>10.1109/2.74]
• 39: A. W. Wilson, Jr., Hierarchical cache/bus architecture for shared memory multiprocessors, Proceedings of the 14th annual international symposium on Computer architecture, p.244-252, June 02-05, 1987, Pittsburgh, Pennsylvania, United States [doi>10.1145/30350.30378]

INDEX TERMS

Primary Classification:
  C. Computer Systems Organization
  C.1 PROCESSOR ARCHITECTURES
      C.1.2 Multiple Data Stream Architectures (Multiprocessors)
          Subjects: Parallel processors**

Additional Classification:
  C. Computer Systems Organization
  C.1 PROCESSOR ARCHITECTURES
      C.1.2 Multiple Data Stream Architectures (Multiprocessors)
          Subjects: Interconnection architectures (e.g., common bus, multiport memory, crossbar switch)

  K. Computing Milieux
  K.1 THE COMPUTER INDUSTRY
      Subjects: Suppliers; Statistics

General Terms:
Design, Measurement, Performance