臺灣博碩士論文加值系統

自我修復NAND型快閃記憶體是近年被提出的新硬體技術，相對於傳統NAND型快閃記憶體經過多次讀寫而造成的寫穿狀態，自我修復NAND型快閃記憶體可以透過高溫使電子離開隧道氧化物讓電晶體可以重新儲存電子解除寫穿狀態；雖然高溫能夠恢復電晶體的狀態，但在現今材料科學的製程中，電晶體僅能承受約攝氏250度，加上自我修復的操作需要一定的時間而可能阻擋快閃記憶體內部其他操作造成效能降低，因此我們提出了一種基於自我修復的NAND型快閃記憶體之資料恢復快閃記憶體轉換層，並以此轉換層來解決因為恢復操作而無法進行存取特定資料的問題。

Self-Healing NAND flash memory is an interesting hardware technology in recently years. It has ability to recover the transistor which was worn-out due to repeatedly writes and erases. It needs to use high-temperature to heat the transistor and let the electron left the tunnel oxide by getting energy. Unfortunately, the more quickly you want to recover the transistor, the higher temperature you have to heat the transistor which can only endure about 250℃. Furthermore, healing operations in flash memory may block the normal requests and can degrade the performance. Therefore, we propose a data restoration flash translation layer for self-healing NAND flash memory. The objective of the thesis is to avoid the problem that data in NAND flash memory cannot be accessed due to the healing operations.

1. Introduction 4
2. Related Works 7
2.1. Self-healing NAND Flash Memory 7
2.2. RAID 8
2.3. RAID-based Flash Translation Layer 10
2.4. Partial Parity Cache 11
3. System Architecture 13
3.1. System Overview 13
3.2. Glossary and Background Knowledge 14
3.3. Address Translation 16
3.4. Data Recovery 19
3.5. Parity Management 22
3.6. Garbage Collection 25
4. Experiment 29
4.1. Experiment Setup 29
4.2. Performance and Overhead 30
4.3. policy of garbage collection 32
5. Conclusion 34
6. Reference 35

[1] G. D. a. T. Z. Qi Wu, A First Study on Self-Healing Solid-State Drives, IEEE International on Memory Workshop (IMW), 2011.
[2] Y. W. D. L. Z. S. a. S. J. Renhai Chen, Heating Dispersal for Self-Healing NAND Flash Memory, IEEE Transactions on Computers, 2016.
[3] Y.-H. C. J.-J. C. T.-W. K. H.-P. L. H.-T. L. Yu-Ming Chang, On Trading Wear-leveling with Heal-leveling, Proceedings of the 51st Annual Design Automation Conference, 2014.
[4] Y.-H. C. a. T.-W. K. Ming-Chang Yang, Virtual Flash Chips: Rethinking the Layer Design of Flash Devices to Improve Data Recoverability, Proceedings of the 52nd Annual Design Automation Conference, 2015.
[5] S. L. B. L. a. K. K. Hyokyung Bahn, A Lifespan-aware Reliability Scheme for RAID-based Flash Storage, Proceedings of the 2011 ACM Symposium on Applied Computing, 2011.
[6] Y. C. a. L. Chang, A Self-Balancing Striping Scheme for NAND-Flash Storage Systems, Proceedings of the 2008 ACM symposium on Applied computing, 2008.
[7] S. J. a. Y. H. S. Yangsup Lee, FRA: A Flash-aware Redundancy Array of Flash Storage Devices, Proceedings of the 7th IEEE/ACM international conference on Hardware/software codesign and system synthesis, 2009.
[8] W. W. Y. Z. L. S. L. H. L. a. Y. C. Jie Guo, DA-RAID-5: A Disturb Aware Data Protection Technique for NAND Flash Storage Systems, Proceedings of the Conference on Design, Automation and Test in Europe, 2013.
[9] S. I. a. D. Shin, Delayed Partial Parity Scheme for Reliable and High-Performance Flash Memory SSD, IEEE 26th Symposium on Mass Storage Systems and Technologies (MSST), 2010.
[10] J.-W. H. a. C.-J. Su, Parity Management Scheme for a Hybrid-Storage RAID, Proceedings of the 31st Annual ACM Symposium on Applied Computing, 2016}.
[11] A. a. Y. a. B.Urgaonkar, A Flash Translation Layer Employing Demand-based Selective Caching of Page-level Address Mappings, Proceedings of the 14th international conference on Architectural support for programming languages and operating systems(ASPLOS), 2009.
[12] M. a. W.Zwaenepoel, eNVy: a non-volatile, main memory storage system, Proceeding ASPLOS VI Proceedings of the sixth international conference on Architectural support for programming languages and operating systems, 1994.
[13] J.-W. H. a. H.-Y. L. a. D.-L. Yang, Multi-Channel Architecture-Based FTL for Reliable and High-Performance SSD, IEEE Transactions on Computers, 2013.
[14] D. J. a. J.-U. K. a. H. J. a. J.-S. K. a. J. Lee, Superblock FTL: A superblock-based flash translation layer with a hybrid address translation scheme, ACM Transactions on Embedded Computing Systems (TECS), 2010.
[15] S. L. a. D. S. a. Y.-J. K. a. J. Kim, LAST: locality-aware sector translation for NAND flash memory-based storage systems, ACM SIGOPS Operating Systems Review, 2008.
[16] Transcend, “Comparison of SLC,MLC,and TLC : http://tw.transcend-info.com/Embedded/Essay-7,” [線上]. Available: http://tw.transcend-info.com/Embedded/Essay-7.
[17] J.-W. a. L. M.-X. Hsieh, Configurable Reliability Framework for SSD-RAID, Non-Volatile Memory Systems and Applications Symposium (NVMSA), 2014.