Shingled magnetic recording (SMR) is an extension of Perpendicular magnetic recording (PMR) technology, which can increase areal density by overlapping the neighboring tracks.
However, SMR has some restrictions that may reduce performance during the random writes, and the time-consuming cleaning process for the persistent cache of SMR will cause the response time to suddenly increase.
HSTL (Half-Shingled Translation Layer) is a novel idea of drive-managed SMR, which proposes the half-shingled zone that can not only in-place update the data without data loss but also increase the storage density compared to the traditional hard disk, and the full-shingled zone that retains some advantages of persistent cache to reduce the Read-Merge-Write operations.
The Zone Converter of HSTL can efficiently execute a cleaning process to reclaim more space, and the latency of cleaning can be greatly improved so the system response time can be smooth.
The simulated SMR is implemented to evaluate the experimental results, the results show that the average response time can be reduced by 18.39% on average, and the write amplification can be reduced by 87.99% on average.

Shingled magnetic recording (SMR) is an extension of Perpendicular magnetic recording (PMR) technology, which can increase areal density by overlapping the neighboring tracks.
However, SMR has some restrictions that may reduce performance during the random writes, and the time-consuming cleaning process for the persistent cache of SMR will cause the response time to suddenly increase.
HSTL (Half-Shingled Translation Layer) is a novel idea of drive-managed SMR, which proposes the half-shingled zone that can not only in-place update the data without data loss but also increase the storage density compared to the traditional hard disk, and the full-shingled zone that retains some advantages of persistent cache to reduce the Read-Merge-Write operations.
The Zone Converter of HSTL can efficiently execute a cleaning process to reclaim more space, and the latency of cleaning can be greatly improved so the system response time can be smooth.
The simulated SMR is implemented to evaluate the experimental results, the results show that the average response time can be reduced by 18.39% on average, and the write amplification can be reduced by 87.99% on average.

[1] Charbel Tannous and R. Lawrence Comstock. Magnetic Information- Storage Materials, pages 1–1. Springer International Publishing, 2017. ISBN 978-3-319-48933-9. doi: 10.1007/978-3-319-48933-9 49.
[2] Weiping He and David H.C. Du. Novel address mappings for shingled write disks. In 6th USENIX Workshop on Hot Topics in Storage and File Systems (HotStorage 14), Philadelphia, PA, June 2014. USENIX Association.
[3] Chenlin Ma, Zhaoyan Shen, Yi Wang, and Zili Shao. Alleviating hot data write back effect for shingled magnetic recording storage systems. IEEE Transactions on Computer-Aided Design of Integrated Circuits and Sys- tems, 38(12):2243–2254, 2019. doi: 10.1109/TCAD.2018.2878190.
[4] Weiping He and David H.C. Du. Smart: An approach to shingled magnetic recording translation. In 15th USENIX Conference on File and Storage Technologies (FAST 17), pages 121–134, Santa Clara, CA, February 2017. USENIX Association. ISBN 978-1-931971-36-2.
[5] C. Ma, Z. Shen, J. Wang, Y. Wang, R. Chen, Y. Guan, and Z. Shao. Tiler: An autonomous region-based scheme for smr stor- age. IEEE Transactions on Computers, 70(2):291–304, 2021. doi: 10.1109/TC.2020.2988004.
[6] Yi-Jing Chuang, Shuo-Han Chen, Yuan-Hao Chang, Yu-Pei Liang, Hsin- Wen Wei, and Wei-Kuan Shih. Dstl: A demand-based shingled transla- tion layer for enabling adaptive address mapping on smr drives. ACM Trans. Embed. Comput. Syst., 19(4), July 2020. ISSN 1539-9087. doi: 10.1145/3391892. URL https://doi.org/10.1145/3391892.
[7] Chenlin Ma, Yi Wang, Zhaoyan Shen, and Zili Shao. Kfr: Optimal cache management with k-framed reclamation for drive-managed smr disks. In 2020 57th ACM/IEEE Design Automation Conference (DAC), pages 1–6, 2020. doi: 10.1109/DAC18072.2020.9218636.
[8] Timothy R. Feldman and Garth A. Gibson. Shingled magnetic record- ing: Areal density increase requires new data management. login Usenix Mag., 38, 2013.
[9] Saurabh Kadekodi, Swapnil Pimpale, and Garth A. Gibson. Caveat- scriptor: Write anywhere shingled disks. In 7th USENIX Workshop on Hot Topics in Storage and File Systems (HotStorage 15), Santa Clara, CA, July 2015. USENIX Association.
[10] Chao Jin, Wei-Ya Xi, Zhi-Yong Ching, Feng Huo, and Chun-Teck Lim. Hismrfs: A high performance file system for shingled storage array. In 2014 30th Symposium on Mass Storage Systems and Technologies (MSST), pages 1–6, 2014. doi: 10.1109/MSST.2014.6855539.
[11] Ting Yao, Zhihu Tan, Jiguang Wan, Ping Huang, Yiwen Zhang, Chang- sheng Xie, and Xubin He. Sealdb: An efficient lsm-tree based kv store on smr drives with sets and dynamic bands. IEEE Transac- tions on Parallel and Distributed Systems, 30(11):2595–2607, 2019. doi: 10.1109/TPDS.2019.2918219.
[12] Ping-Xiang Chen, Shuo-Han Chen, Yuan-Hao Chang, Yu-Pei Liang, and Wei-Kuan Shih. Facilitating the efficiency of secure file data and meta- data deletion on smr-based ext4 file system. In Proceedings of the 26th Asia and South Pacific Design Automation Conference, ASPDAC ’21, page 728–733, New York, NY, USA, 2021. Association for Computing Machinery. ISBN 9781450379991. doi: 10.1145/3394885.3431517. URL https://doi.org/10.1145/3394885.3431517.
[13] Wenguo Liu, Hui Li, and Lingfang Zeng. Star: A zone translation scheme to improve the performance of host-aware smr. In 2021 IEEE 6th International Conference on Big Data Analytics (ICBDA), pages 267–271, 2021. doi: 10.1109/ICBDA51983.2021.9402952.
[14] Lingfang Zeng, Zehao Zhang, Yang Wang, Dan Feng, and Kenneth B. Kent. Cosafs: A cooperative shingle-aware file system. ACM Trans. Storage, 13(4), November 2017. ISSN 1553-3077. doi: 10.1145/3149482. URL https://doi.org/10.1145/3149482.
[15] Anand Suresh, Garth A. Gibson, and G. Ganger. Shingled magnetic recording for big data applications, 2012.
[16] Lawrence Ying and Ted Ts’o. Dynamic hybrid-smr: an ocp proposal to improve data center disk drives, 2017.
[17] Abutalib Aghayev and Peter Desnoyers. Skylight—a window on shingled disk operation. In 13th USENIX Conference on File and Storage Tech- nologies (FAST 15), pages 135–149, Santa Clara, CA, February 2015. USENIX Association. ISBN 978-1-931971-201.
[18] Ming-Chang Yang, Yuan-Hao Chang, Fenggang Wu, Tei-Wei Kuo, and David H.C. Du. Virtual persistent cache: Remedy the long latency behavior of host-aware shingled magnetic recording drives. In 2017 IEEE/ACM International Conference on Computer-Aided Design (IC- CAD), pages 17–24, 2017. doi: 10.1109/ICCAD.2017.8203755.
[19] Tianming Yang, Haitao Wu, Ping Huang, and Fei Zhang. A shingle- aware persistent cache management scheme for dm-smr disks. In 2017 IEEE International Conference on Computer Design (ICCD), pages 81– 88, 2017. doi: 10.1109/ICCD.2017.21.
[20] Joseph Born Kadane. Kadane’s algorithm.
[21] Fenggang Wu, Bingzhe Li, zhichao Cao, Baoquan Zhang, Ming-Hong Yang, Hao Wen, and David H.C. Du. Zonealloy: Elastic data and space management for hybrid SMR drives. In 11th USENIX Workshop on Hot Topics in Storage and File Systems (HotStorage 19), Renton, WA, July 2019. USENIX Association.
[22] Timothy Feldman. Flex dynamic recording. login Usenix Mag., 43, 2018.
[23] Msr cambridge block i/o traces. URL http://iotta.snia.org/traces/388.
[24] Mansour Shafaei and Peter Desnoyers. Virtual guard: A track-based translation layer for shingled disks. In 9th USENIX Workshop on Hot Topics in Storage and File Systems (HotStorage 17), Santa Clara, CA, July 2017. USENIX Association.