Skip to main content

Thank you for visiting nature.com. You are using a browser version with limited support for CSS. To obtain the best experience, we recommend you use a more up to date browser (or turn off compatibility mode in Internet Explorer). In the meantime, to ensure continued support, we are displaying the site without styles and JavaScript.

  • Article
  • Published:

Massively parallel probabilistic computing with sparse Ising machines

Nature Electronics volume 5, pages 460–468 (2022)Cite this article

Subjects

Abstract

Solving computationally hard problems using conventional computing architectures is often slow and energetically inefficient. Quantum computing may help with these challenges, but it is still in the early stages of development. A quantum-inspired alternative is to build domain-specific architectures with classical hardware. Here we report a sparse Ising machine that achieves massive parallelism where the flips per second—the key figure of merit—scales linearly with the number of probabilistic bits. Our sparse Ising machine architecture, prototyped on a field-programmable gate array, is up to six orders of magnitude faster than standard Gibbs sampling on a central processing unit, and offers 5–18 times improvements in sampling speed compared with approaches based on tensor processing units and graphics processing units. Our sparse Ising machine can reliably factor semi-primes up to 32 bits and it outperforms competition-winning Boolean satisfiability solvers in approximate optimization. Moreover, our architecture can find the correct ground state, even when inexact sampling is made with faster clocks. Our problem encoding and sparsification techniques could be applied to other classical and quantum Ising machines, and our architecture could potentially be scaled to 1,000,000 or more p-bits using analogue silicon or nanodevice technologies.

This is a preview of subscription content, access via your institution

Access options

Buy this article

  • Purchase on Springer Link
  • Instant access to full article PDF
Buy now

Prices may be subject to local taxes which are calculated during checkout

Fig. 1: Combinatorial optimization with invertible logic.
Fig. 2: Performance scaling between parallel (sIM) and serial (CPU) implementations.
Fig. 3: Exact factorization of a 32-bit number (P = 4,277,546,633) in the sIM.
Fig. 4: Performance of sIM versus competition-winning SAT solvers.
Fig. 5: Overclocked Gibbs sampling with sIM.

Similar content being viewed by others

Data availability

The data that support the plots within this paper and other findings of this study are available from the corresponding authors upon reasonable request.

Code availability

The computer code and problem instances used in this study are available from the corresponding authors upon reasonable request.

References

  1. Metropolis, N., Rosenbluth, A. W., Rosenbluth, M. N., Teller, A. H. & Teller, E. Equation of state calculations by fast computing machines. J. Chem. Phys. 21, 1087–1092 (1953).

    Article  Google Scholar 

  2. Buluc, A. et al. Randomized algorithms for scientific computing (RASC). Preprint at https://arxiv.org/abs/2104.11079 (2021).

  3. Hinton, G. E. A practical guide to training restricted Boltzmann machines. in Neural Networks: Tricks of the Trade 599–619 (Springer, 2012).

  4. Mansinghka, V. K., Jonas, E. M. & Tenenbaum, J. B. Stochastic Digital Circuits for Probabilistic Inference. Report No. MITCSAIL-TR (Massachussets Institute of Technology, 2008).

  5. Bouchard-Côté, A., J Vollmer, S. & Doucet, A. The bouncy particle sampler: a nonreversible rejection-free Markov chain Monte Carlo method. J. Am. Stat. Assoc. 113, 855–867 (2018).

    Article  MathSciNet  Google Scholar 

  6. Kirkpatrick, S., Gelatt, C. D. & Vecchi, M. P. Optimization by simulated annealing. Science 220, 671–680 (1983).

    Article  MathSciNet  Google Scholar 

  7. McMahon, P. L. et al. A fully programmable 100-spin coherent Ising machine with all-to-all connections. Science 354, 614–617 (2016).

    Article  MathSciNet  Google Scholar 

  8. Yamaokam, M. et al. 24.3 20k-spin Ising chip for combinational optimization problem with CMOS annealing. In 2015 IEEE International Solid-State Circuits Conference—(ISSCC) Digest of Technical Papers 1–3 (IEEE, 2015).

  9. Goto, H., Tatsumura, K. & Dixon, A. R. Combinatorial optimization by simulating adiabatic bifurcations in nonlinear Hamiltonian systems. Sci. Adv. 5, eaav2372 (2019).

    Article  Google Scholar 

  10. Wang, T. & Roychowdhury, J. OIM: oscillator-based Ising machines for solving combinatorial optimisation problems. In International Conference on Unconventional Computation and Natural Computation 232–256 (Springer, 2019).

  11. Ahmed, I., Chiu, P.-W. & Kim, C. H. A probabilistic self-annealing compute fabric based on 560 hexagonally coupled ring oscillators for solving combinatorial optimization problems. In 2020 IEEE Symposium on VLSI Circuits 1–2 (IEEE, 2020).

  12. Dutta, S. et al. An Ising Hamiltonian solver based on coupled stochastic phase-transition nano-oscillators. Nat. Electron. 4, 502–512 (2021).

    Article  Google Scholar 

  13. Borders, W. A. et al. Integer factorization using stochastic magnetic tunnel junctions. Nature 573, 390–393 (2019).

  14. Aramon, M. et al. Physics-inspired optimization for quadratic unconstrained problems using a digital annealer. Front. Phys. 7, 48 (2019).

    Article  Google Scholar 

  15. Yamamoto, K. et al. 7.3 STATICA: a 512-spin 0.25M-weight full-digital annealing processor with a near-memory all-spin-updates-at-once architecture for combinatorial optimization with complete spin-spin interactions. In 2020 IEEE International Solid-State Circuits Conference—(ISSCC) 138–140 (IEEE, 2020).

  16. Patel, S., Canoza, P. & Salahuddin, S. Logically synthesized and hardware-accelerated restricted Boltzmann machines for combinatorial optimization and integer factorization. Nat. Electron. 5, 92–101 (2022).

    Article  Google Scholar 

  17. Su, Y., Mu, J., Kim, H. & Kim, B. A 252 spins scalable CMOS Ising chip featuring sparse and reconfigurable spin interconnects for combinatorial optimization problems. In 2021 IEEE Custom Integrated Circuits Conference (CICC) 1–2 (IEEE, 2021).

  18. Smithson, S. et al. Efficient CMOS invertible logic using stochastic computing. IEEE Trans. Circuits Syst. I, Reg. Papers 66, 2263–2274 (2019).

    Article  MathSciNet  Google Scholar 

  19. Cai, F. et al. Power-efficient combinatorial optimization using intrinsic noise in memristor Hopfield neural networks. Nat. Electron. 3, 409–418 (2020).

    Article  Google Scholar 

  20. Tatsumura, K., Yamasaki, M. & Goto, H. Scaling out Ising machines using a multi-chip architecture for simulated bifurcation. Nat. Electron. 4, 208–217 (2021).

    Article  Google Scholar 

  21. Dixit, V., Selvarajan, R., Alam, M. A., Humble, T. S. & Kais, S. Training restricted Boltzmann machines with a D-Wave quantum annealer. Front. Phys. 9, 589626 (2021).

  22. Koller D. & Friedman, N. Probabilistic Graphical Models: Principles and Techniques (MIT Press, 2009).

  23. Finocchio, G. et al. The promise of spintronics for unconventional computing. J. Magn. Magn. Mater. 521, 167506 (2021).

    Article  Google Scholar 

  24. Andriyash, E. et al. Boosting Integer Factoring Performance via Quantum Annealing Offsets. Report No. 14 (D-Wave Technical Report Series, 2016).

  25. Dridi, R. & Alghassi, H. Prime factorization using quantum annealing and computational algebraic geometry. Sci. Rep. 7, 43048 (2017).

    Article  Google Scholar 

  26. Jiang, S., Britt, K. A., McCaskey, A. J., S Humble, T. & Kais, S. Quantum annealing for prime factorization. Sci. Rep. 8, 17667 (2018).

    Article  Google Scholar 

  27. Lucas, A. Ising formulations of many NP problems. Front. Phys. 2, 5 (2014).

  28. Camsari, K. Y. et al. Stochastic p-bits for invertible logic. Phys. Rev. X 7, 031014 (2017).

    Google Scholar 

  29. Onizawa, N. et al. A design framework for invertible logic. In 2019 53rd Asilomar Conference on Signals, Systems, and Computers 312–316 (IEEE, 2019).

  30. Brélaz, D. New methods to color the vertices of a graph. Commun. ACM 22, 251–256 (1979).

    Article  MathSciNet  Google Scholar 

  31. De Sa, C., Re, C. & Olukotun, K. Ensuring rapid mixing and low bias for asynchronous Gibbs sampling. In Proc. 33rd International Conference on Machine Learning 1567–1576 (PMLR, 2016).

  32. Ko, G. G., Chai, Y., Rutenbar, R. A., Brooks, D. & Wei, G.-Y. FlexGibbs: reconfigurable parallel Gibbs sampling accelerator for structured graphs. In 2019 IEEE 27th Annual International Symposium on Field-Programmable Custom Computing Machines (FCCM) 334 (IEEE, 2019).

  33. Fang, Y. et al. Parallel tempering simulation of the three-dimensional Edwards–Anderson model with compact asynchronous multispin coding on GPU. Comput. Phys. Commun. 185, 2467–2478 (2014).

    Article  Google Scholar 

  34. Yang, K., Chen, Y.-F., Roumpos, G., Colby, C. & Anderson, J. High performance Monte Carlo simulation of Ising model on TPU clusters. In Proc. International Conference for High Performance Computing, Networking, Storage and Analysis 83 (ACM, 2019).

  35. Yoshimura, C., Hayashi, M., Okuyama, T. & Yamaoka, M. FPGA-based annealing processor for Ising model. In 2016 Fourth International Symposium on Computing and Networking (CANDAR) 436–442 (IEEE, 2016).

  36. Kaminsky, W. M. & Lloyd, S. Scalable architecture for adiabatic quantum computing of NP-hard problems. in Quantum Computing and Quantum Bits in Mesoscopic Systems 229–236 (Springer, 2004).

  37. Block, B., Virnau, P. & Preis, T. Multi-GPU accelerated multi-spin Monte Carlo simulations of the 2D Ising model. Comput. Phys. Commun. 181, 1549–1556 (2010).

    Article  Google Scholar 

  38. Preis, T., Virnau, P., Paul, W. & Schneider, J. J. GPU accelerated Monte Carlo simulation of the 2D and 3D Ising model. J. Comput. Phys. 228, 4468–4477 (2009).

    Article  Google Scholar 

  39. Bhatti, S. et al. Spintronics based random access memory: a review. Mater. Today 20, 530–548 (2017).

    Article  Google Scholar 

  40. Safranski, C. et al. Demonstration of nanosecond operation in stochastic magnetic tunnel junctions. Nano Lett. 21, 2040–2045 (2021).

    Article  Google Scholar 

  41. Hayakawa, K. et al. Nanosecond random telegraph noise in in-plane magnetic tunnel junctions. Phys. Rev. Lett. 126, 117202 (2021).

    Article  Google Scholar 

  42. Sutton, B. et al. Autonomous probabilistic coprocessing with petaflips per second. IEEE Access 8, 157238–157252 (2020).

    Article  Google Scholar 

  43. Mohseni, M. et al. Nonequilibrium Monte Carlo for unfreezing variables in hard combinatorial optimization. Preprint at https://arxiv.org/abs/2111.13628 (2021).

  44. Mosca, M., Marcos Vensi Basso, J. & Verschoor, S. R. On speeding up factoring with quantum SAT solvers. Sci. Rep. 10, 15022 (2020).

    Article  Google Scholar 

  45. Hoos, H. H. & Stützle, T. SATLIB: an online resource for research on SAT. Sat 2000, 283–292 (2000).

    MATH  Google Scholar 

  46. Fleury, A. B. K. F. M. & Heisinger, M. CaDiCaL, KISSAT, PARACOOBA, PLINGELING and TREENGELING entering the SAT competition 2020. SAT COMPETITION 2020, 50 (2020).

  47. Soos, M., Devriendt, J., Gocht, S., Shaw, A. & Meel, K. S. CryptoMiniSat with CCAnr at the SAT competition 2020. SAT COMPETITION 2020, 27 (2020).

  48. Zhang, X., Bashizade, R., Wang, Y., Mukherjee, S. & Lebeck, A. R. Statistical robustness of Markov chain Monte Carlo accelerators. In Proc. 26th ACM International Conference on Architectural Support for Programming Languages and Operating Systems 959–974 (ACM, 2021).

  49. Biere, A. CaDiCaL, LINGELING, PLINGELING, TREENGELING and YALSAT entering the SAT competition 2017. In Proc. SAT Competition 13–14 (2017).

Download references

Acknowledgements

We are grateful to B. M. Sutton, D. Eppens, A. Ho and M. Mohseni for useful discussions. N.A.A. is grateful to G. Eschemann for useful discussions on airhdl. We acknowledge Xilinx for the hardware support. K.Y.C. and L.T. acknowledge support from the Institute of Energy Efficiency, University of California, Santa Barbara. K.Y.C. and N.A.A. acknowledge support from the National Science Foundation through CCF 2106260. The research of A.G., M.C. and G.F. has been supported by project no. PRIN 2020LWPKH7 funded by the Italian Ministry of University and Research and by Petaspin association (https://www.petaspin.com/). Computational facilities were purchased with funds from the National Science Foundation (CNS-1725797) and administered by the Center for Scientific Computing (CSC). The CSC is supported by the California NanoSystems Institute and the Materials Research Science and Engineering Center (MRSEC; NSF DMR 1720256) at University of California, Santa Barbara.

Author information

Authors and Affiliations

  1. Department of Electrical and Computer Engineering, University of California, Santa Barbara, CA, USA

    Navid Anjum Aadit, Luke Theogarajan & Kerem Y. Camsari

  2. Department of Mathematical and Computer Sciences, Physical Sciences and Earth Sciences, University of Messina, Messina, Italy

    Andrea Grimaldi & Giovanni Finocchio

  3. Department of Electrical and Information Engineering, Politecnico di Bari, Bari, Italy

    Mario Carpentieri

  4. Department of Physics, University of California, Santa Barbara, CA, USA

    John M. Martinis

  5. Quantala, Santa Barbara, CA, USA

    John M. Martinis

  6. Zyphra Technologies, San Francisco, CA, USA

    John M. Martinis

Authors
  1. Navid Anjum Aadit
    View author publications

    You can also search for this author in PubMed Google Scholar

  2. Andrea Grimaldi
    View author publications

    You can also search for this author in PubMed Google Scholar

  3. Mario Carpentieri
    View author publications

    You can also search for this author in PubMed Google Scholar

  4. Luke Theogarajan
    View author publications

    You can also search for this author in PubMed Google Scholar

  5. John M. Martinis
    View author publications

    You can also search for this author in PubMed Google Scholar

  6. Giovanni Finocchio
    View author publications

    You can also search for this author in PubMed Google Scholar

  7. Kerem Y. Camsari
    View author publications

    You can also search for this author in PubMed Google Scholar

Contributions

N.A.A. and K.Y.C. conceived the idea and planned the study. N.A.A. set up the FPGA and performed the experimental measurements. A.G., N.A.A., G.F. and K.Y.C. performed the data analysis. All the authors have participated in useful discussions and in writing the manuscript.

Corresponding authors

Correspondence to Navid Anjum Aadit, Giovanni Finocchio or Kerem Y. Camsari.

Ethics declarations

Competing interests

The authors declare no competing interests.

Peer review

Peer review information

Nature Electronics thanks Kosuke Tatsumura, Masanao Yamaoka and the other, anonymous, reviewer(s) for their contribution to the peer review of this work.

Additional information

Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Supplementary information

Supplementary Information

Supplementary Discussion, Figs. 1–9 and Tables 1 and 2.

Rights and permissions

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Aadit, N.A., Grimaldi, A., Carpentieri, M. et al. Massively parallel probabilistic computing with sparse Ising machines. Nat Electron 5, 460–468 (2022). https://doi.org/10.1038/s41928-022-00774-2

Download citation

  • Received:

  • Accepted:

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1038/s41928-022-00774-2

This article is cited by

Search

Advanced search

Quick links

Nature Briefing AI and Robotics

Sign up for the Nature Briefing: AI and Robotics newsletter — what matters in AI and robotics research, free to your inbox weekly.

Get the most important science stories of the day, free in your inbox. Sign up for Nature Briefing: AI and Robotics