SET-CMP: Improving the Lifetime of NVM Cache

  • Bhukya Krishna PriyaEmail author
  • N. Ramasubramanian
Conference paper
Part of the Advances in Intelligent Systems and Computing book series (AISC, volume 1025)


NonVolatile Memory (NVM) technologies have gained more attention and are being actively followed by the industries. Though NVM deliver very low leakage and high-density, they have extremely limited write endurance. The cache replacement policies designed for traditional caches reduces the lifespan of STT-RAM cache, thus causing deeply written cache blocks to fail before the others. The existing cache management policy distributes unequal writes on both the intra and interset on the NVM. And, constant threshold values are used to either flush the data or rewrite data on the cache block to enhance its lifespan. But by using a constant threshold, the location of a cache block can be predicted, as the data will be flushed or shifted after a certain threshold and it may fail against a simple attack like RAA. To solve these problems, a SET-CMP is proposed. SET-CMP uses randomized threshold values, such that the actual cache block location can’t be predicted, it also tolerates partial cell failures by permanently flagging the failed cache lines as INVALID, hence no additional data is written into the failed cache blocks. The SET-CMP reduces the numeral ways in the resultant cache set by one, which in turn increases the lifetime by 15% and the performance gets reduced.


Cache Write endurance Wear leveling Interset write variation Intraset write variation Lifespan improvement 


  1. 1.
    Ahn, J., Choi, K.: LASIC: loop-aware sleepy instruction caches based on STT-RAM technology. IEEE Trans. on VLSI Syst. 22(5), 1197–1201. IEEE (2014)Google Scholar
  2. 2.
    Reza, J.M., Arjomand, M., Azad, H.S.: Sequoia: a high-endurance NVM-based cache architecture. IEEE Trans. on VLSI Syst. 24(3), 954–967. IEEE (2016)Google Scholar
  3. 3.
    Lin, C., Chiou, J.N.: High-endurance hybrid cache design in CMP architecture with cache partitioning and access-aware policies. IEEE Trans. on VLSI Sys. 23(10), 2149–2161. IEEE (2012)Google Scholar
  4. 4.
    Joo, Y., Niu, D., Dong, X., Sun, G., Chang, N., Xie, Y.: Energy- and endurance-aware design of phase change memory caches. In: Proceedings of the Conference on Design, Automation and Test in Europe, pp. 136–141, European Design and Automation Association, Dresden, Germany, IEEE (2010)Google Scholar
  5. 5.
    Xu, J., Feng, D., Tong, W., Liu, J., Zhou, W.: Encoding separately: an energy-efficient write scheme for MLC STT_RAM. In: 2017 IEEE International Conference on Computer Design (ICCD), pp. 581–584, Boston, MA, USA, IEEE (2017)Google Scholar
  6. 6.
    Yin, S., Li, J., Liu, L., Wei, S., Guo, Y.: “g.” In: Proceeding of the 2015 Design, Automation & Test in Europe Conference & Exhibition (DATE), pp. 187–192, EDA Consortium (2015)Google Scholar
  7. 7.
    Raouf, A.H.F., Abouei, J.: Cache replacement scheme based on sliding window and TTL video on demand. In: Proceedings of Iranian Conference on Electrical Engineering (ICEE), pp. 499–504, Mashhad, Iran, IEEE (2018)Google Scholar
  8. 8.
    Duan, G., Wang, S.: Exploiting narrow-width values for improving the non-volatile cache lifetime. In: Proceedings of the 2014 Design, Automation & Test in Europe Conference & Exhibition (DATE), pp. 1–4, Dresden, Germany, IEEE (2014)Google Scholar
  9. 9.
    Cho, S., Lee, H.: Flip-n-write: a simple deterministic technique to improve pram write performance, energy and endurance. In: Proceedings of the IEEE International Symposium on Micro Architecture (MICRO’09), pp. 347–357, New York, USA (2009)Google Scholar
  10. 10.
    Zhou, P., Zhao, B., Yang, J., Zhang, Y.: A durable and energy efficient main memory using phase change memory. In: Proceedings of 36th Annual International Symposium on Computer Architecture ISCA’09, Austin, TX, USA, pp. 14–23, 20–24 June 2009Google Scholar
  11. 11.
    Mittal, S., Vetter, J.S.: ENLIVE: a write-minimization technique for improving the lifetime of non-volatile caches. ORNL/TM-2015/232, ORNL Report, June 2015Google Scholar
  12. 12.
    Wang, J., Dong, X., Xie, Y., Jouppi, N.P.: “i2 WAP: improving non-volatile cache lifetime by reducing inter-and intra-set write variations. In: Proceedings of IEEE International Symposium on High-Performance Computer Architecture (HPCA), Shenzhen, China, pp. 234–245, 23–27 February 2013Google Scholar
  13. 13.
    Mittal, S., Vetter, J.S.: AYUSH: a technique for extending lifetime of SRAM-NVM hybrid caches. IEEE Comput. Archite. Lett. 14(2), 115–118 (2015)CrossRefGoogle Scholar
  14. 14.
    Binert, N., et al.: Gem5: a multiple-ISA full system simulator with detailed memory model. In: Computer Architecture News (2011)Google Scholar
  15. 15.
    Bienia, C., Kumar, S., Singh, J.P., Li, K.: The PARSEC benchmark suite: characterization and architectural implications. In: Proceedings of the 17th International Conference on Parallel Architectures and Compilation Techniques, Toronto, Ontario, Canada, pp. 72–81, 25–29 October 2008Google Scholar

Copyright information

© Springer Nature Singapore Pte Ltd. 2020

Authors and Affiliations

  1. 1.National Institute of Technology, TiruchirappalliTiruchirappalliIndia

Personalised recommendations