Abstract
In this paper, a synthesis method (CAPSYS) of dedicated architectures is introduced. Its aim is to produce optimized systems derived from the algorithmic expression of a numerical application. The approach addresses the design of dedicated systems for applications that require high numerical computations. An efficient utilization of hardware resources is achieved through the use of vector processing with an SIMD implementation. The synthesis algorithm realizes simultaneously the design of SIMD structures and the generation of the microcode needed for implementing a software pipelining of operations of the source program. CAPSYS considers a generic model composed of both mechanisms required to manage the flow of controls in a SIMD machine and the description of a parallel data memory. All the synthesized architectures derive from this generic model. Capabilities of CAPSYS are illustrated through the design of an image convolution processor and a two-dimensional median filtering processor. This last example shows also an interesting feature of CAPSYS which permits to instantiate dedicated hardware components in the program of the target application: a hardware realization of conditional schemes in loops allows to get an efficient vectorization of the algorithm and an efficient dedicated architecture.
Similar content being viewed by others
References
B.W. Wah et al., “High Performance Computing and Communication for Grand Challenge Applications: Computer vision, Speech and Natural Language Processing, and Artificial Intelligence,”IEEE Trans. on Knowledge and Data Engineering, Vol. 5, 1993.
P.G. Paulin, J.P. Knight, and E.F. Girczyc, “HAL: A Multi-Paradigm Approach to Automatic Data-Path Synthesis,”Proceedings 23rd Design Automation Conf., New York, pp. 263–270, 1986.
A.C. Parker, J. Pizarro, and M. Mlinar, “MAHA: A program for Datapath Synthesis,”Proceedings 23rd Design Automation Conf., New York, pp. 461–466, 1986.
T.A. Ly, W.L. Elwood, and E.F. Girczic, “A generalized interconnect model for data path synthesis,”Proceedings EDAC, pp. 168–173, 1990.
B.S. Haroun and M.I. Elmasry, ”Automatic synthesis of multibus architectures for DSP,”Proceedings ISCAS, pp. 44–47, 1988.
B.M. Pangrle, “SLICER: a state synthesis for intelligent silicon compiler,”Proceedings Int. Conf. on Computer Design, pp. 436–541, 1988.
M. Breternitz and J.P. Shen, “Architecture synthesis of high performance application specific processors,”Proceedings 27th Design Automation Conf., Orlando, pp. 542–548, 1990.
M.C. McFarland, A.C. Parker, and R. Camposano, “The high level synthesis of digital systems,”Proceedings of IEEE., Vol. 78, pp. 301–318, 1990.
G. Menez, M. Auguin, F. Boeri and C. Carriere, “A partitioning algorithm for system level synthesis,”Proceedings ICCAD92, pp. 482–487, 1992.
J.A. Fisher and J.J. O'Donnel, “VLIW machines: Multiprocessor we can actually program,”Proceedings IEEE Compcon, pp. 299–305, 1984.
A.V. Aho, R. Sethi, and J.D. Ullman,Compilers, principles, techniques and tools, Addison-Wesley, 1986.
D. Landskov, S. Davidson, and B. Shriver, “Local microcode compaction techniques,”ACM Computing Surveys, Vol. 12, 1980.
C. Eisenbeis, “Optimization of horizontal microcode generation for loop structures,”Proceedings International Conf. on Supercomputing, Saint Malo, 1988.
C. Carriere,Definition et optimisation de l'interconnexion dans un processeur specialise a commande synchrone, Ph.D. thesis, Univ. Nice Sophia/Antipolis, September, 1992.
F. Boeri and M. Auguin, “OPSILA: A Vector and Parallel Processor,”IEEE Transactions on Computers, Vol. 42, pp. 76–82, 1993.
M. Auguin and F. Boeri, “Parallel memory management in a SIMD computer,”Proceedings IFIP 10.3 Working Conference on Highly Parallel Computers, Sophia-Antipolis, France, 1986.
J. Lenfant, “Parallel permutation of data: A Benes network control algorithm for frequently used permutations,”IEEE Transactions on Computers, Vol. C-27, pp. 637–647, 1978.
D.H. Lawrie, “Access and alignment of data in an array processor,”IEEE Transactions on Computers, Vol. C-24, pp. 1145–1155, 1975.
P.M. Kogge,The architecture of pipelined computers, McGraw-Hill, 1981.
P. Duclos,Etude du parallelisme en traitement des images, realisation sur une architecture mixte SIMD/SPMD, Ph.D. thesis, Univ. Nice Sophia/Antipolis, October 1988.
Author information
Authors and Affiliations
Rights and permissions
About this article
Cite this article
Auguin, M., Boéri, F., Carrière, C. et al. Synthesis of dedicated SIMD processors. Journal of VLSI Signal Processing 9, 167–179 (1995). https://doi.org/10.1007/BF02407083
Received:
Revised:
Published:
Issue Date:
DOI: https://doi.org/10.1007/BF02407083