An interleaved array-processing architecture

ABSTRACT

This paper describes an array-processing architecture capable of executing high-level vector operations. There are two distinguishing features of this architecture: First, the user can define for later use complex vector operations that involve several arithmetic operations and branching. Once defined, they appear as built-in vector instructions to the user. Second, the algorithms for accessing and aligning vectors are implemented in hardware, eliminating the need for user programs to deal with memory addresses.

REFERENCES

• 1: D. J. Kuck , R. A. Stokes, The Burroughs Scientific Processor (BSP), IEEE Transactions on Computers, v.31 n.5, p.363-376, May 1982 [doi>10.1109/TC.1982.1676014]
• 2: Wulf, W. A., R. Levin, and S. P. Harbison. HYDRA/C.mmp. New York: McGraw-Hill, 1981.
• 3: G. H. Barnes , S. F. Lundstrom, Design and Validation of a Connection Network for Many-Processor Multiprocessor Systems, Computer, v.14 n.12, p.31-41, December 1981 [doi>10.1109/C-M.1981.220293]
• 4: Batcher, K. E. "The Multidimensional Access Memory in STARAN." IEEE Transactions on Computers, C-26 (1977), pp. 174--177.
• 5: J. H. Patel, Performance of Processor-Memory Interconnections for Multiprocessors, IEEE Transactions on Computers, v.30 n.10, p.771-780, October 1981 [doi>10.1109/TC.1981.1675695]
• 6: D. M. Dias , J. R. Jump, Packet Switching Interconnection Networks for Modular Systems, Computer, v.14 n.12, p.43-53, December 1981 [doi>10.1109/C-M.1981.220295]
• 7: E. U. Cohler , J. E. Storer, Functionally Parallel Architecture for Array Processors, Computer, v.14 n.9, p.28-36, September 1981 [doi>10.1109/C-M.1981.220596]
• 8: James E. Smith, Decoupled access/execute computer architectures, Proceedings of the 9th annual symposium on Computer Architecture, p.112-119, April 26-29, 1982, Austin, Texas, United States
• 9: Philip C. Treleaven , David R. Brownbridge , Richard P. Hopkins, Data-Driven and Demand-Driven Computer Architecture, ACM Computing Surveys (CSUR), v.14 n.1, p.93-143, March 1982 [doi>10.1145/356869.356873]
• 10: Sudhir Raman Ahuja, A modular multiprocessor architecture for array processing., Rice University, Houston, TX, 1977
• 11: Ahuja, S. R., and J. R. Jump. "A Modular Vector Processing Unit." In Proceedings of the 1976 International Conference on Parallel Processing. New York: IEEE, August 1976.
• 12: S. R. Ahuja , J. R. Jump, A modular memory scheme for array processing, Proceedings of the 4th annual symposium on Computer architecture, p.90-94, March 23-25, 1977