Abstract
Many high-level synthesis tools offer degrees of freedom in mapping high-level specifications to Register-Transfer Level descriptions. These choices do not affect the functional behavior but span a design space of different cost-performance tradeoffs. In this paper we present a novel machine learning-based approach that efficiently determines the Pareto-optimal designs while only sampling and synthesizing a fraction of the design space. The approach combines three key components: (1) A regression model based on Gaussian processes to predict area and throughput based on synthesis training data. (2) A "smart" sampling strategy, GP-PUCB, to iteratively refine the model by carefully selecting the next design to synthesize to maximize progress. (3) A stopping criterion based on assessing the accuracy of the model without access to complete synthesis data. We demonstrate the effectiveness of our approach using IP generators for discrete Fourier transforms and sorting networks. However, our algorithm is not specific to this application and can be applied to a wide range of Pareto front prediction problems.
- C. Coello, G. B. Lamont, and D. Veldhuizen. Evolutionary Algorithms for Solving Multi-Objective Problems (Genetic and Evolutionary Computation). Springer-Verlag New York, Inc., Secaucus, NJ, USA, 2006. Google Scholar
Digital Library
- L. Deng, K. Sobti, and C. Chakrabarti. Accurate Models for Estimating Area and Power of FPGA Implementations. In Proc. of International Conference on Acoustics, Speech and Signal Processing (ICASSP), pages 1417--1420, 2008.Google Scholar
Cross Ref
- M. Ehrgott and X. Gandibleux. A Survey and Annotated Bibliography of Multiobjective Combinatorial Optimization. OR Spektrum, (22):425--460, 2000.Google Scholar
- M. Emmerich, K. Giannakoglou, and B. Naujoks. Single- and Multiobjective Evolutionary Optimization Assisted by Gaussian Random Field Metamodels. IEEE Transactions on Evolutionary Computation, 10(4):421--439, 2006. Google Scholar
Digital Library
- J. Knowles. ParEGO: a Hybrid Algorithm with On-line Landscape Approximation for Expensive Multiobjective Optimization Problems. IEEE Transactions on Evolutionary Computation, 10(1):50--66, 2006. Google Scholar
Digital Library
- S. Künzli, L. Thiele, and E. Zitzler. Modular Design Space Exploration Framework for Embedded Systems. IEE Proceedings Computers & Digital Techniques, 152(2):183--192, 2005.Google Scholar
Cross Ref
- P. A. Milder, M. Ahmad, J. C. Hoe, and M. Püschel. Fast and Accurate Resource Estimation of Automatically Generated Custom DFT IP Cores. In Proc. of International Symposium on Field-Programmable Gate Arrays (FPGA), pages 211--220, 2006. Google Scholar
Digital Library
- P. A. Milder, F. Franchetti, J. C. Hoe, and M. Püschel. Formal Datapath Representation and Manipulation for Implementing DSP Transforms. In Proc. of Design Automation Conference (DAC), pages 385--390, 2008. Google Scholar
Digital Library
- G. Palermo, C. Silvano, and V. Zaccaria. ReSPIR: A Response Surface-Based Pareto Iterative Refinement for Application-Specific Design Space Exploration. Computer-Aided Design of Integrated Circuits and Systems, IEEE Transactions on, 28(12):1816--1829, dec. 2009. Google Scholar
Digital Library
- C. Rasmussen and H. Nickisch. Gaussian Process Regression and Classification Toolbox Version 3.1 for Matlab 7.x, 2010.Google Scholar
- C. Rasmussen and C. K. I. Williams. Gaussian Processes for Machine Learning. MIT Press, 2006. Google Scholar
Digital Library
- B. So, M. W. Hall, and P. C. Diniz. A Compiler Approach to Fast Hardware Design Space Exploration in FPGA-Based Systems. In Proc. of Programming Language Design and Implementation (PLDI), pages 165--176, 2002. Google Scholar
Digital Library
- N. Srinivas, A. Krause, S. Kakade, and M. Seeger. Gaussian Process Optimization in the Bandit Setting: No Regret and Experimental Design. In Proc. of International Conference on Machine Learning (ICML), 2010.Google Scholar
- L. Yan, T. Srikanthan, and N. Gang. Area and Delay Estimation for FPGA Implementation of Coarse-Grained Reconfigurable Architectures. In Proc. of Languages, Compilers, and Tools for Embedded Systems, pages 182--188, 2006. Google Scholar
Digital Library
- Q. Zhang, W. Liu, E. Tsang, and B. Virginas. Expensive Multiobjective Optimization by MOEA/D with Gaussian Process Model. IEEE Transactions on Evolutionary Computation, 14(3):456--474, 2010. Google Scholar
Digital Library
- E. Zitzler, D. Brockhoff, and L. Thiele. The Hypervolume Indicator Revisited: on the Design of Pareto-compliant Indicators via Weighted Integration. In Proc. of the 4th International Conference on Evolutionary Multi-criterion Optimization (EMO), pages 862--876, 2007. Google Scholar
Digital Library
- E. Zitzler and S. Künzli. Indicator-based Selection in Multiobjective Search. In Proc. of the 8th International Conference on Parallel Problem Solving from Nature, pages 832--842, 2004.Google Scholar
Cross Ref
- E. Zitzler, M. Laumanns, and L. Thiele. SPEA2: Improving the Strength Pareto Evolutionary Algorithm for Multiobjective Optimization. In Evolutionary Methods for Design, Optimisation, and Control, pages 95--100, 2002.Google Scholar
- M. Zuluaga, P. Milder, and M. Püschel. Computer Generation of Streaming Sorating Nnetworks. In Proceedings of the 45th Annual ACM/IEEE Conference on Design Automation (DAC), 2012. Google Scholar
Digital Library
Index Terms
"Smart" design space sampling to predict Pareto-optimal solutions
Recommendations
"Smart" design space sampling to predict Pareto-optimal solutions
LCTES '12: Proceedings of the 13th ACM SIGPLAN/SIGBED International Conference on Languages, Compilers, Tools and Theory for Embedded SystemsMany high-level synthesis tools offer degrees of freedom in mapping high-level specifications to Register-Transfer Level descriptions. These choices do not affect the functional behavior but span a design space of different cost-performance tradeoffs. ...
Accelerating FPGA Prototyping through Predictive Model-Based HLS Design Space Exploration
DAC '19: Proceedings of the 56th Annual Design Automation Conference 2019One of the advantages of High-Level Synthesis (HLS), also called C-based VLSI-design, over traditional RT-level VLSI design flows, is that multiple micro-architectures of unique area vs. performance can be automatically generated by setting different ...
Efficient System-Level Design Space Exploration for High-Level Synthesis Using Pareto-Optimal Subspace Pruning
ASPDAC '23: Proceedings of the 28th Asia and South Pacific Design Automation ConferenceHigh-level synthesis (HLS) is a rapidly evolving and popular approach to designing, synthesizing, and optimizing embedded systems. Many HLS methodologies utilize design space exploration (DSE) at the post-synthesis stage to find Pareto-optimal hardware ...






Comments