EDGE COMPUTING

Integrating memristors and CMOS for better AI

Article metrics

By integrating memristor arrays with CMOS circuitry, a computing-in-memory architecture can be created that could provide efficient deep neural network processors.

Access options

Rent or Buy article

Get time limited or full article access on ReadCube.

from$8.99

All prices are NET prices.

Fig. 1: Von Neumann architecture versus nvCIM architecture for AI edge devices.

References

  1. 1.

    LeCun, Y., Bengio, Y. & Hinton, G. Nature 521, 436–444 (2015).

  2. 2.

    Nurvitadhi, E. et al. Int. Symp. Field-Programmable Gate Arrays (FPGA) 5–14 (ACM, 2017).

  3. 3.

    Jouppi, N. P. et al. Int. Symp. Computer Architecture (ISCA) 1–12 (ACM/IEEE, 2017).

  4. 4.

    Xu, X. et al. Nat. Electron. 1, 216–222 (2018).

  5. 5.

    Chi, P. et al. Int. Symp. Computer Architecture (ISCA) 27–39 (ACM/IEEE, 2016).

  6. 6.

    Yang, J. J., Strukov, D. B. & Stewart, D. R. Nat. Nanotechnol. 8, 13–24 (2013).

  7. 7.

    Ambrogio, S. et al. Nature 558, 60–67 (2018).

  8. 8.

    Prezioso, M. et al. Nature 521, 61–64 (2015).

  9. 9.

    Ankit, A. et al. Int. Conf. Architectural Support for Programming Languages and Operating Systems (ASPLOS) 715–731 (ACM, 2019).

  10. 10.

    Chen, W.-H. et al. Nat. Electron. https://doi.org/10.1038/s41928-019-0288-0 (2019).

Download references

Author information

Correspondence to Yiyu Shi.

Rights and permissions

Reprints and Permissions

About this article

Verify currency and authenticity via CrossMark