Chunqiao Li
Skip Abstract Section
Abstract
Fast verification for the extremely large-scale power grid is demanding as CMOS technology advances consistently. In this work, we propose ESPSim, an efficient scalable power grid simulator based on a parallel smoothed aggregation-based algebraic multigrid technique. ESPSim has the ability to do fast DC and transient analysis through MPI and adaptive timestep control mechanism. Thanks to the smoother applied on the prolongation operator, ESPSim copes well with the convergence rate on extremely large-scale power grid transient analysis. Extensive experiments are conducted with a variety of serial/parallel solvers. The runtime of ESPSim is linear with case size. With 16 processors, 1,000 timesteps transient analysis of 63.4M nodes can be completed in 22.1 min. Over 22× speedup compared to the well-known direct solver Cholmod is observed.
- [1] Check Memory. 2021. Checking Memory from a Parallel C Program. Retrieved from https://hpcf.umbc.edu/general-productivity/checking-memory-usage/.Google Scholar
- [2] Hypre User Guide. 2021. Hypre Software and Documentation. Retrieved from https://computing.llnl.gov/projects/hypre-scalable-linear-solvers-multigrid-methods.Google Scholar
- [3] xxHash User Guide. 2021. xxHash. Retrieved from https://github.com/Cyan4973/xxHash.Google Scholar
- [4] Parmetis User Guide. 2022. Parmetis Manual Version 4.0. Retrieved from http://glaros.dtc.umn.edu/gkhome/metis/parmetis/overview.Google Scholar
- [5] . 2011. Parallel and scalable transient simulator for power grids via waveform relaxation (PTS-PWR). IEEE Trans. Very Large Scale Integr. Syst. 19, 2 (2011), 319–332. Google ScholarDigital Library
- [6] . 2010. Improvements of a fast parallel poisson solver on irregular domains. In Proceedings of the 10th International Conference on Applied Parallel and Scientific Computing (PARA’10)(
Lecture Notes in Computer Science , Vol. 7133), (Ed.). Springer, 65–74. Google ScholarDigital Library - [7] . 2021. PETSc Web Page. Retrieved from https://petsc.org/.Google Scholar
- [8] . 2004. Adaptive smoothed aggregation (\(\alpha\)SA). SIAM J. Sci. Comput. 25, 6 (2004), 1896–1920. Google ScholarDigital Library
- [9] . 2000. A Multigrid Tutorial, 2nd ed. SIAM.Google ScholarCross Ref
- [10] . 2008. Large scale P/G grid transient simulation using hierarchical relaxed approach. Integration 41, 1 (2008), 153–160. Google ScholarDigital Library
- [11] . 2001. Efficient large-scale power grid analysis based on preconditioned Krylov-subspace iterative methods. In Proceedings of the 38th Design Automation Conference. 559–562. Google ScholarCross Ref
- [12] . 2008. Algorithm 887: CHOLMOD, supernodal sparse cholesky factorization and update/downdate. ACM Trans. Math. Softw. 35, 3 (2008), 22:1–22:14. Google ScholarDigital Library
- [13] . 2019. Machine-learning-driven matrix ordering for power grid analysis. In Proceedings of the Design, Automation Test in Europe Conference Exhibition (DATE’19). 984–987. Google ScholarCross Ref
- [14] . [n.d.]. SuiteSparse. Retrieved from https://people.engr.tamu.edu/davis/suitesparse.html.Google Scholar
- [15] . 2008. Distance-two interpolation for parallel algebraic multigrid. Numer. Linear Algebra Appl. 15, 2–3 (2008), 115–139. Google ScholarCross Ref
- [16] . 2006. Reducing complexity in parallel algebraic multigrid preconditioners. SIAM J. Matrix Anal. Appl. 27, 4 (2006), 1019–1039. Google ScholarDigital Library
- [17] . 2019. AMGCL: An efficient, flexible, and extensible algebraic multigrid implementation. Lobachevskii J. Math. 40, 5 (
May 2019), 535–546. Google ScholarCross Ref - [18] . 1997. Applied Numerical Linear Algebra. SIAM. Google ScholarCross Ref
- [19] . 2008. Multigrid on GPU: Tackling power grid analysis on parallel SIMT platforms. In Proceedings of the IEEE/ACM International Conference on Computer-Aided Design. 647–654. Google ScholarCross Ref
- [20] . 2007. Parallel symbolic factorization for sparse LU with static pivoting. SIAM J. Sci. Comput. 29, 3 (2007), 1289–1314. Google ScholarCross Ref
- [21] . 2016. Exponential Time Integration for Transient Analysis of Large-Scale Circuits. Ph.D. Dissertation. Department of Computer Science, University of California, San Diego, CA. Retrieved from https://escholarship.org/uc/item/60d8c2r6.Google Scholar
- [22] . 2014. Parallel power grid analysis using distributed direct linear solver. In Proceedings of the IEEE International Symposium on Electromagnetic Compatibility (EMC’14). 866–871. Google ScholarCross Ref
- [23] . 1952. Methods of conjugate gradients for solving linear systems. J. Res. Natl. Bureau Stand. 49, 6 (1952), 409–436.Google ScholarCross Ref
- [24] . 1998. Inside SPICE. McGraw-Hill.Google ScholarDigital Library
- [25] . 2015. Parallel simulation for VLSI power grid. Ph.D. Dissertation. the Department of Computer Science & Engineering, Texas A&M University.Google Scholar
- [26] . 2002. Power grid transient simulation in linear time based on transmission-line-modeling alternating-direction-implicit method. IEEE Trans. Comput.-Aided Design Integr. Circ. Syst. 21, 11 (2002), 1343–1352. Google ScholarDigital Library
- [27] . 2017. InTraSim: Incremental transient simulation of power grids. IEEE Trans. Comput.-Aided Design Integr. Circ. Syst. 36, 12 (2017), 2052–2065. Google ScholarCross Ref
- [28] . 2021. IBM Power Grid Benchmarks. Retrieved from https://web.ece.ucsb.edu/lip/PGBenchmarks/ibmpgbench.html.Google Scholar
- [29] . 2008. Principles of Parallel Programming. Pearson.Google ScholarDigital Library
- [30] . 2021. feGRASS: Fast and effective graph spectral sparsification for scalable power grid analysis. IEEE Trans. Comput.-Aided Design Integr. Circ. Syst. (2021), 1–1. Google ScholarCross Ref
- [31] . 2000. Fast power grid simulation. In Proceedings of the 37th Design Automation Conference. 156–161. Google ScholarDigital Library
- [32] . 2010. An aggregation-based algebraic multigrid method. Electr. Trans. Numer. Anal. 37 (2010), 123–146.Google Scholar
- [33] . 2021. AGMG Software and Documentation Documentation. Retrieved from http://homepages.ulb.ac.be/ynotay/AGMG.Google Scholar
- [34] . 2015. A massively parallel solver for discrete Poisson-like problems. J. Comput. Phys. 281 (2015), 237–250. Google ScholarDigital Library
- [35] . 2004. Transient analysis of on-chip power distribution networks using equivalent circuit modeling. In Proceedings of the International Symposium on Signals, Circuits and Systems (SCS’03). 63–68. Google ScholarCross Ref
- [36] . 2005. Power grid analysis using random walks. IEEE Trans. Comput.-Aided Design Integr. Circ. Syst. 24, 8 (2005), 1204–1224. Google ScholarDigital Library
- [37] . 2003. Iterative Methods for Sparse Linear Systems. SIAM.Google ScholarCross Ref
- [38] . 2014. Algorithms, Part I, 4th ed. Pearson Education.Google Scholar
- [39] . 2007. Parallel domain decomposition for simulation of large-scale power grids. In Proceedings of the IEEE/ACM International Conference on Computer-Aided Design. 54–59. Google ScholarCross Ref
- [40] . 2001. Convergence of algebraic multigrid based on smoothed aggregation. Numer. Math. 88, 3 (2001), 559–579. Google ScholarCross Ref
- [41] . 1996. Algebraic multigrid by smoothed aggregation for second and fourth order elliptic problems. Computing 56, 3 (
Sep. 1996), 179–196. Google ScholarCross Ref - [42] . 2013. Scalable power grid transient analysis via MOR-assisted time-domain simulations. In Proceedings of the IEEE/ACM International Conference on Computer-Aided Design (ICCAD’13). 548–552. Google ScholarCross Ref
- [43] . 2001. Performance of algebraic multi-grid solvers based on unsmoothed and smoothed aggregation schemes. Int. J. Numer. Methods Fluids 36, 7 (2001), 743–772.Google ScholarCross Ref
- [44] . 2021. THU Power Grid Benchmarks. Retrieved from http://tiger.cs.tsinghua.edu.cn/PGBench/.Google Scholar
- [45] . 2012. PowerRush: Efficient transient simulation for power grid analysis. In Proceedings of the IEEE/ACM International Conference on Computer-Aided Design (ICCAD’12). 653–659.Google ScholarDigital Library
- [46] . 2014. PowerRush: An efficient simulator for static power grid analysis. IEEE Trans. Very Large Scale Integr. Syst. 22, 10 (2014), 2103–2116. Google ScholarCross Ref
- [47] . 2012. PGT_SOLVER: An efficient solver for power grid transient analysis. In Proceedings of the IEEE/ACM International Conference on Computer-Aided Design (ICCAD’12). 647–652.Google ScholarDigital Library
- [48] . 2012. Efficient parallel power grid analysis via additive Schwarz method. In Proceedings of the International Conference on Computer-Aided Design (ICCAD’12). ACM, New York, NY, 399–406. Google ScholarDigital Library
- [49] . 2016. Parallel power grid analysis based on enlarged partitions. ACM Trans. Design Autom. Electr. Syst. 21, 2 (2016), 26:1–26:21. Google ScholarDigital Library
- [50] . 2021. pGRASS-Solver: A parallel iterative solver for scalable power grid analysis based on graph spectral sparsification. Retrieved from https://numbda.cs.tsinghua.edu.cn/papers/iccad21_2.pdf.Google Scholar
- [51] . 2003. Power network analysis using an adaptive algebraic multigrid approach. In Proceedings of the Design Automation Conference (DAC’03). 105–108. Google ScholarCross Ref
- [52] . 2014. MATEX: A distributed framework for transient simulation of power distribution networks. In Proceedings of the 51st Annual Design Automation Conference (DAC’14). ACM, New York, NY, 1–6. Google ScholarDigital Library
- [53] . 2015. An algorithmic framework for efficient large-scale circuit simulation using exponential integrators. In Proceedings of the 52nd ACM/EDAC/IEEE Design Automation Conference (DAC’15). 1–6. Google ScholarDigital Library
- [54] . 2008. Power grid analysis and optimization using algebraic multigrid. IEEE Trans. Comput.-Aided Design Integr. Circ. Syst. 27, 4 (2008), 738–751. Google ScholarDigital Library
Index Terms
- ESPSim: An Efficient Scalable Power Grid Simulator Based on Parallel Algebraic Multigrid
Recommendations
Parallel and scalable transient simulator for power grids via waveform relaxation (PTS-PWR)
This paper presents a fast algorithm for transient simulation of power grids in very large scale integration systems using waveform relaxation (WR) techniques. Novel partitioning methods and convergence accelerators are developed for fast convergence of ...
Parallel forward and back substitution for efficient power grid simulation
ICCAD '12: Proceedings of the International Conference on Computer-Aided DesignParallelizing power grid simulation with factorization-based direct or preconditioned iterative methods is a challenging task due to the data dependency among forward and back substitution. In this paper, we propose two approaches to parallelize forward ...
Power grid reduction based on algebraic multigrid principles
DAC '03: Proceedings of the 40th annual Design Automation ConferenceWith the scaling of technology, power grid noise is becoming increasingly significant for circuit performance. A typical power grid circuit contains millions of linear elements, making noise analysis and verification challenging in terms of both run ...
Comments