相比單階層非揮發性記憶體，因為多階層非揮發性記憶體密度的優勢，使得它更適合作為儲存裝置使用。但是多階層非揮發性記憶體有更高的寫入延遲與更快磨損的問題。先前已有研究利用多階層非揮發性記憶體的雙留存特性改善寫入延遲與延長裝置壽命。我們發現檔案系統因為使用寫時複製保護檔案資料，有許多的檔案資料塊被頻繁丟棄。因此在這篇研究中，我們提出了一個稱為DB-Dual的設計。該設計可以辨識資料庫的檔案訪問模式，預測很快就會被丟棄的檔案資料塊，並對這些資料使用較短留存時間的寫入方式。根據實驗的結果，相比於未使用DB-Dual的原始檔案系統，資料庫的速度平均提升了31.0%，多階層非揮發性記憶體的壽命平均提升了61.9%。

Due to the high capacity of MLC (multi-level cell) NVM, it is more suitable as a storage system device than SLC NVM. However, MLC NVM has problems related to higher write latency and faster device wear. Previous studies have used the dual retention of MLC NVM to improve write latency and increase device lifetime. We found that the file system uses a copy-on-write consistency mechanism for file data, which causes many file data blocks to be discarded on a frequent basis. Therefore, in this study, we propose a design called DB-Dual, which recognizes the file access patterns in the database to predict the file data blocks that will be discarded and uses short retention for these data. According to the experimental results, compared with the baseline, the performance of the database was improved by 31.0%, on average, and the lifetime of MLC NVM was improved by 61.9%, on average.

[1] R.-S. Liu, D.-Y. Shen, C.-L. Yang, S.-C. Yu, and C.-Y. M. Wang, ‘‘NVM duet: Unified working memory and persistent store architecture,’’ ACM SIGARCH Comput. Archit. News, vol. 42, no. 1, pp. 455–470, 2014.
[2] M. Zhang, L. Zhang, L. Jiang, Z. Liu, and F. T. Chong, “Balancing performance and lifetime of MLC PCM by using a region retention monitor,” in Symp. on High-Performance Computer Architecture (HPCA), 2017, pp. 385–396.
[3] M. Zhang, L. Zhang, L. Jiang, F. T. Chong and Z. Liu, "Quick-and-Dirty: Improving Performance of MLC PCM by Using Temporary Short Writes," 2017 IEEE International Conference on Computer Design (ICCD), Boston, MA, 2017, pp. 585-588, doi: 10.1109/ICCD.2017.101.
[4] Y. Luo, Y. Cai, S. Ghose, J. Choi and O. Mutlu, "WARM: Improving NAND flash memory lifetime with write-hotness aware retention management," 2015 31st Symposium on Mass Storage Systems and Technologies (MSST), Santa Clara, CA, 2015, pp. 1-14, doi: 10.1109/MSST.2015.7208284.
[5] L. Sun, Y. Lu, and J. Shu, “DP2: Reducing Transaction Overhead with Differential and Dual Persistency in Persistent Memory,” in Proceedings of the 12th ACM International Conference on Computing Frontiers, Italy, 2015, article 24.
[6] S.-H. Chen et al., “Enhancing the energy efficiency of journaling file system via exploiting multi-write modes on MLC NVRAM,” in Proc. Int. Symp. Low Power Electron. Design (ISLPED), 2018, pp. 23-1–23-6.
[7] S. Chen, Y. Chang, Y. Chang and W. Shih, "mwJFS: A Multiwrite-Mode Journaling File System for MLC NVRAM Storages," in IEEE Transactions on Very Large Scale Integration (VLSI) Systems, vol. 27, no. 9, pp. 2060-2073, Sept. 2019, doi: 10.1109/TVLSI.2019.2919907.
[8] J. Condit et al., ‘‘Better I/O through byte-addressable, persistent memory,’’ in Proc. 22nd ACM Symp. Oper. Syst. Princ., 2009, pp. 133–146.
[9] S. R. Dulloor et al., ‘‘System software for persistent memory,’’ in Proc. 9th Eur. Conf. Comput. Syst., 2014, pp. 1–15.
[10] J. Xu and S. Swanson, ‘‘NOVA: A log-structured file system for hybrid volatile/non-volatile main memories,’’ in Proc. 14th USENIX Conf. File Storage Technol., 2016, pp. 323–338.
[11] E. H. -. Sha, X. Chen, Q. Zhuge, L. Shi and W. Jiang, "Designing an efficient persistent in-memory file system," 2015 IEEE Non-Volatile Memory System and Applications Symposium (NVMSA), Hong Kong, 2015, pp. 1-6, doi: 10.1109/NVMSA.2015.7304365.
[12] E. Sha, X. Chen, Q. Zhuge, L. Shi, and W. Jiang, ‘‘A new design of in-memory file system based on file virtual address framework,’’ IEEE Trans. Comput., vol. 65, no. 10, pp. 2959–2972, Oct. 2016.
[13] S.-H. Chen, T.-Y. Chen, Y.-H. Chang, H.-W. Wei, and W.-K. Shih, “UnistorFS: A union storage file system design for resource sharing between memory and storage on persistent RAM-based systems,” ACM Trans. Storage, vol. 14, nos. 1–3, pp. 1–3, Feb. 2018.
[14] (2020, Jul.). SQLite. [Online]. Available: https://www.sqlite.org/
[15] (2020, Jul.). MySQL. [Online]. Available: http:// http://www.mysql.com/
[16] (2020, Jul.). tpcc-mysql. [Online]. Available: https://github.com/Percona-Lab/tpcc-mysql
[17] “DDR4 SDRAM NVRDIMM MTA36ASS4G72XF1Z,” Micron Technology Inc., 2020. Micron Inc.
[18] (2020, Jul.). Kyoto Cabinet. [Online]. Available: https://dbmx.net//kyotocabinet/
[19] (2020, Jul.). PostgreSQL. [Online]. Available: http://www. postgresql.org/
[20] (2020, Jul.). linkbench. [Online]. Available: https://github.com/mdcallag/linkbench/
[21] (2020, Jul.). linkbench. [Online]. Available: https://github.com/facebookarchive/linkbench/
[22] (2020, Jul.). Mobibench. [Online]. Available: https://github.com/ESOS-Lab/Mobibench/
[23] L. Tang, Q. Huang, W. Lloyd, S. Kumar, and K. Li, “RIPQ: Advanced photo caching on flash for Facebook,” in Proc. 13th USENIX Conf. File Storage Technol., 2015, pp. 373–386.
[24] R. Karedla, J. S. Love and B. G. Wherry, "Caching strategies to improve disk system performance," in Computer, vol. 27, no. 3, pp. 38-46, March 1994, doi: 10.1109/2.268884.
[25] L. Cherkasova and G. Ciardo, "Role of Aging, Frequency, and Size in Web Cache Replacement Policies", HPCN Europe 2001, LNCS 2110, pp.114-123, 2001
[26] A. Eisenman, A. Cidon, and et al., “Flashield: a hybrid key-value cache that controls flash write amplification,” in the 16th NSD, 2019.
[27] J. Chen, L. Chen, S. Wang, G. Zhu, Y. Sun, H. Liu, and F. Li, “HotRing: A Hotspot-Aware In-Memory Key-Value Store,” in Proc. 18th USENIX Conference on File and Storage Technologies, 2020, pp. 239-252