1 / 36

Tri-Level-Cell Phase Change Memory (PCM): Toward an Efficient and Reliable Memory System

Tri-Level-Cell Phase Change Memory (PCM): Toward an Efficient and Reliable Memory System. Nak Hee Seong Sungkap Yeo Hsien-Hsin S. Lee. School of Electrical and Computer Engineering Georgia Institute of Technology Atlanta, GA 30332 nakhee.seong@gmail.com {sungkap, leehs}@gatech.edu.

miach
Download Presentation

Tri-Level-Cell Phase Change Memory (PCM): Toward an Efficient and Reliable Memory System

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript


  1. Tri-Level-Cell Phase Change Memory (PCM): Toward an Efficient and Reliable Memory System Nak Hee Seong Sungkap Yeo Hsien-Hsin S. Lee School of Electrical and Computer Engineering Georgia Institute of Technology Atlanta, GA 30332 nakhee.seong@gmail.com {sungkap, leehs}@gatech.edu Presented By: Anand Dhole Shalini Satre

  2. Contents • PCM • Background and Motivation • Tri-Level-Cell (3LC) PCM • 3LC PCM in Practice • Evaluation • Conclusion

  3. PCM • Background and Motivation • Tri-Level-Cell (3LC) PCM • 3LC PCM in Practice • Evaluation • Conclusion

  4. Phase Change Memory (PCM) • Promising alternative memory technology • Two states • Crystalline (SET) • Amorphous (RESET) • Multi-level-cell PCM • Intermediate states • Store more data per cell

  5. Single Level Cell (SLC) [1] Set Reset High resistivity Low resistivity

  6. SLC vs MLC Four Storage Levels Two Storage Levels 0 1 002 012 102 112 2LC or SLC = one bit per cell 4LC = two bits per cell

  7. SLC PCM SET RESET i i t t # of Cells  103 106 103 Difference

  8. MLC PCM i SET RESET i i t t t # of Cells Storage Level 0 Storage Level 1 Storage Level 2 Storage Level 3  1k  1M

  9. Error Model • Critical problems • Resistance Drift • Resistance of PCM cell increases over time • Soft errors • Not permanent failure • Have solutions to resolve • Soft error caused by resistance drift • Error rate is proportional to initial resistance value • Error rate is negligible in SLC PCM • In MLC PCM, resistance drift at intermediate levels • Iterative-writing mechanism • Degrades write latency • For 4LC, 4x~8x slower than that of SLC [1]

  10. Resistance Drift [1] T = 1 # of Cells SET RESET Storage Level 0 Storage Level 1 Storage Level 2 Storage Level 3  Programmed Boundaries Decision Boundaries

  11. Resistance Drift T = 2 # of Cells SET RESET Storage Level 0 Storage Level 1 Storage Level 2 Storage Level 3 

  12. Resistance Drift T = 4 # of Cells SET RESET Storage Level 0 Storage Level 1 Storage Level 2 Storage Level 3 

  13. Resistance Drift T = 8 # of Cells SET RESET Storage Level 0 Storage Level 1 Storage Level 2 Storage Level 3  Drift-induced Soft Errors!!!

  14. Drifted Resistance • Power Law Equation

  15. Proposed Solution • Proposed tri-level-cell PCM • Soft error rate matches that of DRAM • Gain performance of SLC PCM

  16. PCM • Background and Motivation • Tri-Level-Cell (3LC) PCM • 3LC PCM in Practice • Evaluation • Conclusion

  17. Background and Motivation • Flash Memory w.r.t. PCM • Switching mem. ele. requires more voltage & time. • Degrades more rapidly • More susceptible to radiation • PCM w.r.t NAND • Better read/write latency. • Consumes significantly less read/write energy. • PCM Advantages • Higher information density. • Cheaper when in mass production.

  18. Background and Motivation cont… • MLC PCM • Many intermediate states between SET and RESET • E.g. 8LC PCM stores three bits per cell • Soft error rate(SER) is higher than that of DRAM • SER increases over time along with resistance • Error correction Methods • Time-aware error correction scheme • Scrub mechanism

  19. Background and Motivation cont… • Time-aware error correction scheme [3] • Uses extra cells for storing predefined reference resistance values • While reading, reference values are used to compensate the resistance drift in corresponding cell. • Reduced SER from 10-3 ~ 10-1 to 10-4 ~ 10-2

  20. Background and Motivation cont… • Scrub Mechanism [2] • Reduced 99.6% of uncorrectable errors • Memory controller spend more time in scrubbing • DRAM-style self refresh [3] • Cells with correct information also gets refreshed • Higher chip-level power • Frequent write decreases lifespan • Slower responsiveness

  21. PCM • Background and Motivation • Tri-Level-Cell (3LC) PCM • 3LC PCM in Practice • Evaluation • Conclusion

  22. 3LC PCM • Each cell has three storage levels • Removed most error-prone state from 4LC PCM i.e. Third storage level • Drift is proportional to resistance • Removes errors generated by third as well as most of the errors generated by second storage level

  23. 3LC PCM Two Storage Levels 0 1 Three Storage Levels 03 13 23 2LC or SLC = one bit per cell Four Storage Levels 3LC 002 012 102 112 ~ 1.5 bits per cell ≠three bits per cell 4LC = two bits per cell Binary System TernarySystem

  24. PCM • Background and Motivation • Tri-Level-Cell (3LC) PCM • 3LC PCM in Practice • Evaluation • Conclusion

  25. 3LC PCM • 4-level cell PCM • unreliable • Tri-level cell PCM • Removing the most error-prone state i i i t t t L2 L0 L1

  26. Bandwidth Expanded 3LC PCM i Relaxing programming range Reducing programming latency Increasing write bandwidth or t SET RESET i i i t t t # of Cells L2 L0 L1 L1

  27. Configuration variable of 4LC PCM Configurationvariable of 3LC PCM

  28. Efficient Conversion Method [1] • In theory 11 bits of binary = 2048 states 7 ternary cells = 2187 states ~94% utilization • Proposed approach 3 bits of binary = 8 states 2 ternary cells = 9 states ~89% utilization Notation: <3,2> conversion

  29. Number Mapping Method 00 000 01 10 001 010 100 02 11 20 011 101 110 12 21 111 22 Binary Ternary

  30. ECC for Tri-Level-Cell PCM Single Bit Error Single Bit Error Ternary Binary • Legacy ECC for binary can be used • Simple (72, 64) Hamming Code • Memory controller requires minimal change

  31. PCM • Background and Motivation • Tri-Level-Cell (3LC) PCM • 3LC PCM in Practice • Evaluation • Conclusion

  32. Drift Induced Error Rate

  33. Information Density Data block size- 256 bits Bits Per Cell Number of Correctable Bits

  34. PCM • Background and Motivation • Tri-Level-Cell (3LC) PCM • 3LC PCM in Practice • Evaluation • Conclusion

  35. Conclusion [1] • Results (over 4LC PCM) • 105 lower soft error rates • 36.4% performance improvement • Results (over SLC PCM) • 1.33x higher information density

  36. References • Nak Hee Seong, Sungkap Yeo, Hsien-Hsin S. Lee, "Tri-Level-Cell Phase Change Memory: Toward an Efficient and Reliable Memory System",ISCA'13 • M. Awasthi, M. Shevgoor, K. Sudan, B. Rajendran, R. Balasubramonian, and V. Srinivasan, “Efficient Scrub Mechanisms for Error-Prone Emerging Memories,” in Proceedings of the International Symposium on High Performance Computer Architecture, 2012.vol. 19, no. 8, pp. 1357–1367, 2011 • W. Xu and T. Zhang, “A time-aware fault tolerance scheme to improve reliability of multilevel phase-change memory in the presence of significant resistance drift,” IEEE Transactions on Very Large Scale Integration (VLSI) Systems, vol. 19, no. 8, pp. 1357–1367, 2011. • T. Nirschl, J. Phipp, T. Happ, G. Burr, B. Rajendran, M. Lee, A. Schrott, M. Yang, M. Breitwisch, C. Chen et al., “Write strategies for 2 and 4-bit multi-level phase-change memory,” in IEEE International Electron Devices Meeting (IEDM), 2007, pp. 461–464. • N. Papandreou, H. Pozidis, T. Mittelholzer, G. Close, M. Breitwisch, C. Lam, and E. Eleftheriou, “Drift-tolerant multilevel phase-change memory,” in 2011 3rd IEEE International Memory Workshop (IMW). IEEE, pp. 1–4. • R. Hamming, “Error detecting and error correcting codes,” Bell System Technical Journal, vol. 29, no. 2, pp. 147–160, 1950.

More Related