Advertisement

The Stollmann Data Flow Machine

  • Michael Jöhnk
  • Ute Schürfeld
Conference paper
Part of the Informatik-Fachberichte book series (INFORMATIK, volume 189)

Abstract

Most of the data flow architectures proposed in the last years aim at number crunching, which naturally seems to imply fine grain paralellism. Just so fine grain seems to lead naturally to (expensive) special hardware. Our aim is to adapt the data flow concept to commercial applications as for example from the database field. These applications more naturally lead to symbolic crunching and coarse respectively variable grain. For this reason and w.r.t. market acceptance we built a machine using off-the-shelf hardware and lifted up the data flow mechanism from hardware to software level. To avoid the classical bottleneck of “firing” we introduced parallel firing.

The dataflow mechanism is implemented on a shared memory architecture. The chosen hardware configuration is based on four processor boards — each equipped with a microprocessor and 4 MByte dual ported RAM — connected to a VMEbus. The host processor runs the UNIX V.3 operating system whereas the other three nodes run SRTX, a realtime operating system kernel. A first prototype will run in October ’88.

Subjects of investigation on the prototype will be among others the correct choice of grain size, the appropriate ratio of firing control units and execution units and questions related to the topic of load balance.

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. [1]
    Adams, D.A.: A Computation Model with Data Flow Sequencing;Technical Report CS 117, School of Humanities and Sciences, Stanford University, Stanford, Calif.; 1968.Google Scholar
  2. [2]
    Boral, H., Dewitt, D.J.: Design Considerations for Data Flow Machines; Proceedings of 1980 International Conference on Management of Data, ACM-SIGMOD; 1980.Google Scholar
  3. [3]
    Boral, H., Dewitt, D.J.: Applying Data Flow Techniques to Data Base Machines; Computer IEEE; 1982.Google Scholar
  4. [4]
    Dennis, J.B.: First Version of Data Flow Procedure Language; in Programming Symp.: Proc.Colloque sur la Programation, (Paris, France, Apr.74), Robinet, B. (Ed.) Lecture notes in computer science, vol. 19, Springer-Verlag, New York; 1974.Google Scholar
  5. [5]
    Dennis, J.B., Misunas, D.P.:A Computer Architecture for Highly Parallel Signal Processing; in Proc. 1974 Nat. Computer Conf. AFIPS Press, Arlington, Va.; 1974.Google Scholar
  6. [6]
    DeWitt, D.J., et al.: GAMMA — A High Performance Data Flow Database Machine; Proceeding of 12th International Conference on Very Large Data Basics; 1986.Google Scholar
  7. [7]
    Eich, M.H., Wells, D.L: Database Concurrently Control Using Data Flow Graphs; ACM Transactions on Database Systems, Vol. 13, No. 2; 1988.Google Scholar
  8. [8]
    Gurd, J.: Data Flow Computers and Languages; ESPRIT Summer School on Future Parallel Computers; 1986.Google Scholar
  9. [9]
    Glück-Hiltrop, E., Jöhnk, M., Kalmer, P., Ramlow, M., Schürfeld, U., Rolf, F.; Expansion Considerations; ESP-415E-STO-032, Deliverable 5 of Stollmann GmbH in ESPRIT Project 415, Stollmann GmbH, Hamburg; 1988.Google Scholar
  10. [10]
    Jipp, T., Friedrich, P., Oldach, H., Fuhlrott, O., Sievers, M.: Definition of Principles and Basic Language; ESP-415E-STO-022, Deliverable 4 of Stollmann GmbH in ESPRIT Project 415, Stollmann GmbH, Hamburg; 1987.Google Scholar
  11. [11]
    Rodriguez, J.E:A Graph Model for Parallel Computation; MIT Technical Report TR-64, Laboratory for Computer Science, MIT, Cambridge, Mass.; 1969.Google Scholar
  12. [12]
    Treleaven, P.C., Brownbridge, D.R., Hopkins, R.P.: Data Driven and Demand Driven Computer Architectures; Computing Surveys, Vol.14, No.1; 1982.Google Scholar
  13. [13]
    von Bültzingloewen, G., Dittrich, K.R., lochpe, C., Liedtke, R.-P., Lockemann, P.C., Schryro, M.: KARDAMOM — A Dataflow Database Machine For Real-Time Application; SIGMOD RECORD, Vol. 17, No. 1; 1988.Google Scholar
  14. [14]
    von Bültzingloewen, G., Liedtke, R.-P., Dittrichs, K.R.: Set-Oriented Memory Management In a Multiprocessor Database Machine; Proceedings of 5th International Workshop on Database Machines; 1987.Google Scholar
  15. [15]
    Zucker, W., Glück-Hiltrop, E., Jipp, T., Rennhak, B., Schäffler, G.: Selection and Determination of the Class of Applications; ESP-415E-STO-014, Deliverable of Stollmann GmbH in ESPRIT Project 415, Stollmann GmbH, Hamburg, 1986.Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 1988

Authors and Affiliations

  • Michael Jöhnk
    • 1
  • Ute Schürfeld
    • 1
  1. 1.Stollmann GmbHHamburg 50Deutschland

Personalised recommendations