A Scalable Black-Box Optimization System for Auto-Tuning VLSI Synthesis Programs

Modern logic and physical synthesis tools provide numerous options and parameters that can drastically impact design quality; however the large number of options leads to a complex design space difficult for human circuit designers to navigate. We tackle this parameter tuning problem with a novel system employing intelligent search strategies and parallel computing, thus automating one of the key design tasks conventionally performed by a human designer. We provide an overview of this system, called SynTunSys, as well as results from employing it during the design of the IBM z13 22nm highperformance server chip, currently in production. During this major processor design, SynTunSys provided significant savings in human design effort and achieved a quality of results beyond what human designers alone could achieve, yielding on average a 36% improvement in total negative slack and a 7% power reduction.

[1]  Luca P. Carloni,et al.  On learning-based methods for design-space exploration with High-Level Synthesis , 2013, 2013 50th ACM/EDAC/IEEE Design Automation Conference (DAC).

[2]  Mark Horowitz,et al.  An integrated framework for joint design space exploration of microarchitecture and circuits , 2010, 2010 Design, Automation & Test in Europe Conference & Exhibition (DATE 2010).

[3]  Matthew M. Ziegler,et al.  POWER8 design methodology innovations for improving productivity and reducing power , 2014, Proceedings of the IEEE 2014 Custom Integrated Circuits Conference.

[4]  Luca P. Carloni,et al.  Scalable Auto-Tuning of Synthesis Parameters for Optimizing High-Performance Processors , 2016, ISLPED.

[5]  Nachiket Kapre,et al.  Driving Timing Convergence of FPGA Designs through Machine Learning and Cloud Computing , 2015, 2015 IEEE 23rd Annual International Symposium on Field-Programmable Custom Computing Machines.

[6]  Vittorio Zaccaria,et al.  A meta-model assisted coprocessor synthesis framework for compiler/architecture parameters customization , 2013, 2013 Design, Automation & Test in Europe Conference & Exhibition (DATE).

[7]  Nando de Freitas,et al.  Bayesian Optimization in High Dimensions via Random Embeddings , 2013, IJCAI.

[8]  Michael F. P. O'Boyle,et al.  Milepost GCC: Machine Learning Enabled Self-tuning Compiler , 2011, International Journal of Parallel Programming.

[9]  James C. Hoe,et al.  Nautilus: Fast automated IP design space search using guided genetic algorithms , 2015, 2015 52nd ACM/EDAC/IEEE Design Automation Conference (DAC).

[10]  Luca Fossati,et al.  Decision-Theoretic Design Space Exploration of Multiprocessor Platforms , 2010, IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems.

[11]  Luca P. Carloni,et al.  A synthesis-parameter tuning system for autonomous design-space exploration , 2016, 2016 Design, Automation & Test in Europe Conference & Exhibition (DATE).

[12]  Nikolaos V. Sahinidis,et al.  Simulation optimization: a review of algorithms and applications , 2014, 4OR.

[13]  Louise Trevillyan,et al.  An integrated environment for technology closure of deep-submicron IC designs , 2004, IEEE Design & Test of Computers.

[14]  Matthew M. Ziegler,et al.  Power reduction by aggressive synthesis design space exploration , 2013, International Symposium on Low Power Electronics and Design (ISLPED).

[15]  Gordon Fraser,et al.  Parameter tuning or default values? An empirical investigation in search-based software engineering , 2013, Empirical Software Engineering.

[16]  Matthew M. Ziegler,et al.  4.1 22nm Next-generation IBM System z microprocessor , 2015, 2015 IEEE International Solid-State Circuits Conference - (ISSCC) Digest of Technical Papers.