Abstract
Growing on-chip wire delays are motivating architectural features that expose on-chip communication to the compiler. EDGE architectures are one example of communication-exposed microarchitectures in which the compiler forms dataflow graphs that specify how the microarchitecture maps instructions onto a distributed execution substrate. This paper describes a compiler scheduling algorithm called spatial path scheduling that factors in previously fixed locations - called anchor points - for each placement. This algorithm extends easily to different spatial topologies. We augment this basic algorithm with three heuristics: (1) local and global ALU and network link contention modeling, (2) global critical path estimates, and (3) dependence chain path reservation. We use simulated annealing to explore possible performance improvements and to motivate the augmented heuristics and their weighting functions. We show that the spatial path scheduling algorithm augmented with these three heuristics achieves a 21% average performance improvement over the best prior algorithm and comes within an average of 5% of the annealed performance for our benchmarks.
- K. Arvind and R.S. Nikhil. Executing a program on the MIT taggedtoken dataflow architecture. IEEE Transactions on Computers, 39(3):300--318, 1990. Google Scholar
Digital Library
- S.J. Beaty and P.H. Sweany. Instruction scheduling using simulated annealing. In International Conference on Massively Parallel Computing Systems, Colorado Springs, CO, Apr. 1998.Google Scholar
- V. Betz and J. Rose. VPR: A new packing, placement and routing tool for FPGA research. In FPL '97: Proceedings of the 7th International Workshop on Field-Programmable Logic and Applications, pages 213--222, London, UK, 1997. Springer-Verlag. Google Scholar
Digital Library
- D. Burger, S.W. Keckler, K.S. McKinley, M. Dahlin, L.K. John, C. Lin, C.R. Moore, J. Burrill, R.G. McDonald, W. Yoder, and others. Scaling to the end of silicon with EDGE architectures. IEEE Computer, pages 44--55, July 2004. Google Scholar
Digital Library
- J.B. Dennis and D.P. Misunas. A preliminary architecture for a basic data-flow processor. In International Symposium on Computer Architecture, pages 126--132, New York, NY, USA, 1975. Google Scholar
Digital Library
- J.R. Ellis. Bulldog: A Compiler for VLIW Architectures. MIT Press, 1986. Google Scholar
Digital Library
- B. Fields, S. Rubin, and R. Bodik. Focusing processor policies via critical-path prediction. In Proceedings of the 28th Annual International Symposium on Computer Architecture, pages 74--85, July 2001. Google Scholar
Digital Library
- J.A. Fisher, J.R. Ellis, J.C. Ruttenberg, and A. Nicolau. Parallel processing: A smart compiler and a dumb machine. In ACM Symposium on Compiler Construction, Montreal, Canada, June 1984. Google Scholar
Digital Library
- E. Gibert, J. Sanchez, and A. Gonzalez. Effective instruction scheduling techniques for an interleaved cache clustered VLIW processor. In Proceedings of the 35th annual ACM/IEEE international symposium on Microarchitecture, pages 123--133, 2002. Google Scholar
Digital Library
- K. Kailas, K. Ebcioglu, and A.K. Agrawala. CARS: A new code generation framework for clustered ILP processors. In International Symposium on High-Performance Computer Architecture, pages 133--143, Jan. 2001. Google Scholar
Digital Library
- C. Kessler and A. Bednarski. Optimal integrated code generation for clustered VLIWarchitectures. In Proceedings of the Joint Conference on Languages, Compilers and Tools for Embedded Systems, pages 102--111, June 2002. Google Scholar
Digital Library
- S. Kirkpatrick, C.D. Gelatt Jr., and M.P. Vecchi. Optimization by simulated annealing. Science, 220(4598):671--680, 1983.Google Scholar
Digital Library
- R.E. Korf. Depth-first iterative-deepening: an optimal admissible tree search. Artif. Intell., 27(1):97--109, 1985. Google Scholar
Digital Library
- W. Lee, D. Puppin, S. Swanson, and S. Amarasinghe. Convergent scheduling. In International Symposium on Microarchitecture, Istanbul, Turkey, Oct. 2002. Google Scholar
Digital Library
- M. Mercaldi, S. Swanson, A. Peterson, A. Putnam, A. Schwerin, M. Oskin, and S. Eggers. Modeling instruction placement on a spatial architecture. In SPAA '06: Proceedings of the Symposium on Parallel Architectures and Applications, 2006. Google Scholar
Digital Library
- J. Moss, P.E. Utgoff, J. Cavazos, D. Precup, D. Stefanovic, C. Brodley, and D. Scheeff. Learning to schedule straight-line code. In Neural Information Processing Systems - Natural and Synthetic, Denver, CO, Dec. 1997. Google Scholar
Digital Library
- R. Nagarajan, D. Burger, K.S. McKinley, C. Lin, S.W. Keckler, and S.K. Kushwaha. Instruction scheduling for emerging communication-exposed architectures. In The International Conference on Parallel Architectures and Compilation Techniques, pages 74--84, Antibes Juan-les-Pins, France, Oct. 2004. Google Scholar
Digital Library
- R. Nagarajan, X. Chen, R.G. McDonald, D. Burger, and S.W. Keckler. Critical path analysis of the TRIPS architecture. In IEEE International Symposium on Performance Analysis of Systems and Software (ISPASS), March 2006.Google Scholar
Cross Ref
- E. Ozer, S. Banerjia, and T.M. Conte. Unified assign and schedule: A new approach to scheduling for clustered register file microarchitectures. In International Symposium on Microarchitecture, pages 308--315, December 1998. Google Scholar
Digital Library
- P.G. Paulin and J.P. Knight. Force-directed scheduling in automatic data path synthesis. In DAC '87: Proceedings of the 24th ACM/IEEE conference on Design automation, pages 195--202, New York, NY, USA, 1987. ACM Press. Google Scholar
Digital Library
- Y. Qian, S. Carr, and P. Sweany. Optimizing loop performance for clustered VLIW architectures. In The International Conference on Parallel Architectures and Compilation Techniques, pages 271--280, Charlottesville, VA, Sept. 2002. Google Scholar
Digital Library
- A. Smith, J. Burrill, J. Gibson, B. Maher, N. Nethercote, B. Yoder, D. Burger, and K.S. McKinley. Compiling for EDGE architectures. In International Symposium on Code Generation and Optimization, Manhattan, NY, Mar. 2006. Google Scholar
Digital Library
- S. Swanson, K. Michaelson, A. Schwerin, and M. Oskin. WaveScalar. In Proceedings of the 36th Symposium on Microarchitecture, December 2003. Google Scholar
Digital Library
- S. Swanson, K. Michelson, and M. Oskin. Configuration by combustion: Online simulated annealing for dynamic hardware configuration. In ASPLOS X Wild and Crazy Idea Session, 2002.Google Scholar
- E. Waingold, M. Taylor, D. Srikrishna, V. Sarkar, W. Lee, V. Lee, J. Kim, M. Frank, P. Finch, R. Barua, J. Babb, S. Amarasinghe, and A. Agarwal. Baring it all to software: Raw machines. IEEE Computer, pages 86--93, Sept. 1997. Google Scholar
Digital Library
- J. Zalamea, J. Llosa, E. Ayguade, and M. Valero. Software and hardware techniques to optimize register file utilization in VLIW architectures. In Proceedings of the International Workshop on Advanced Compiler Technology for High Performance and Embedded Systems (IWACT), July 2001.Google Scholar
Index Terms
A spatial path scheduling algorithm for EDGE architectures
Recommendations
A spatial path scheduling algorithm for EDGE architectures
ASPLOS XII: Proceedings of the 12th international conference on Architectural support for programming languages and operating systemsGrowing on-chip wire delays are motivating architectural features that expose on-chip communication to the compiler. EDGE architectures are one example of communication-exposed microarchitectures in which the compiler forms dataflow graphs that specify ...
A spatial path scheduling algorithm for EDGE architectures
Proceedings of the 2006 ASPLOS ConferenceGrowing on-chip wire delays are motivating architectural features that expose on-chip communication to the compiler. EDGE architectures are one example of communication-exposed microarchitectures in which the compiler forms dataflow graphs that specify ...
A spatial path scheduling algorithm for EDGE architectures
Proceedings of the 2006 ASPLOS ConferenceGrowing on-chip wire delays are motivating architectural features that expose on-chip communication to the compiler. EDGE architectures are one example of communication-exposed microarchitectures in which the compiler forms dataflow graphs that specify ...






Comments