ABSTRACT
We report on a compiler for Warp, a high-performance systolic array developed at Carnegie Mellon. This compiler enhances the usefulness of Warp significantly and allows application programmers to code substantial algorithms.
The compiler combines a novel programming model, which is based on a model of skewed computation for the array, with powerful optimization techniques. Programming in W2 (the language accepted by the compiler) is orders of magnitude easier than coding in microcode, the only alternative available previously.
- 1.Aho, A.V. and Ullman J.D. Principles of Compiler DesiSstGoogle Scholar
- 2.Allen, F.E. and Cocke, J. A Catalogue of Optimizing Transformations. In Design and Optimization of Compiler& Rustin, R., Ed., Prentice-Hall, Englewood Cliffs, N.J., 1972, pp, 1-30,Google Scholar
- 3.Annaratone, M., Arnould, E., Gross, T., Kung, H. T., l.am, M. S., Menzilcioglu, O., Sarocky, K., and Webb, J. A. Warp Architecture and Implementation. Proceedings of the 13 Intl. Symposium on Computer Architecture, ACM, June, 1985. Google ScholarDigital Library
- 4.Hoare, C. A. R. "Communicating Sequential Processes". Communications of the ACM 21, 8 (August 1978), 666-677. Google ScholarDigital Library
- 5.Kung, H.T. Systolic Algorithms for the CMU Warp Processor. Proceedings of the Seventh International Conference on Pattern Recognition, International Association for Pattern Recognition, 1984. pp. 570-577.Google Scholar
- 6.Patel, Janak H. and Davidson, Edward S. Improving the Throughput of a Pipeline by Insertion of Delays. Proe. 3rd Annual Symposium on Computer Architecture, Jan., 1976, pp. 159-164. Google ScholarDigital Library
- 7.Rau, B. R. and Glaeser, C. D. Some Scheduling Techniques and an Easily Schcdulable l lorizontaI Architccture for High PerformaxLce Scientific Computing. Proc. 14fin Annum Workshop on Mieroprogramming, October, 1981. Google ScholarDigital Library
- 8.Steenkiste, P. Global dataflow for W2. internal report.Google Scholar
Index Terms
- Compilation for a high-performance systolic array
Recommendations
Compilation for a high-performance systolic array
We report on a compiler for Warp, a high-performance systolic array developed at Carnegie Mellon. This compiler enhances the usefulness of Warp significantly and allows application programmers to code substantial algorithms.
The compiler combines a novel ...
Compilation of Haskell array comprehensions for scientific computing
PLDI '90: Proceedings of the ACM SIGPLAN 1990 conference on Programming language design and implementationMonolithic approaches to functional language arrays, such as Haskell array comprehensions, define elements all at once, at the time the array is created, instead of incrementally. Although monolithic arrays are elegant, a naive implementation can be ...
Efficient compilation of CUDA kernels for high-performance computing on FPGAs
Special issue on application-specific processorsThe rise of multicore architectures across all computing domains has opened the door to heterogeneous multiprocessors, where processors of different compute characteristics can be combined to effectively boost the performance per watt of different ...
Comments