DOI QR코드

DOI QR Code

A Design of SSD Dedicated RAID System for Efficient Resource Management

효율적인 자원관리를 위한 SSD 전용 RAID 시스템 설계

  • Hyun-Seob Lee (Division of Computer Engineering, Baekseok University)
  • 이현섭 (백석대학교 컴퓨터공학부)
  • Received : 2024.03.04
  • Accepted : 2024.04.08
  • Published : 2024.04.30

Abstract

Enterprise storage systems that require high data reliability are applying RAID (Redundant Array of Independent Disks) systems to recover from data loss and failure. In particular, RAID 5 ensures space efficiency and reliability by distributing parity across multiple storage devices. However, when storage devices have different capacities, RAID is built based on the smallest capacity storage device, resulting in wasted storage space. Therefore, research is needed to solve this resource management problem. In this paper, we propose a method for RAID grouping of each independent NAND flash memory block in a RAID consisting of SSD (Solid State Disk) with external SSDs as well as internal SSDs. This method is divided into a policy for delivering block information inside SSDs to the RAID system and a policy for RAID grouping of physical addresses delivered from the RAID system. This method allows us to maintain a RAID that does not waste resources when SSDs of different capacities are grouped into RAID5. Finally, we demonstrate the effectiveness of the proposed method through experiments.

높은 데이터 신뢰성을 요구하는 엔터프라이즈 저장시스템은 데이터 손실 및 장애 시 복구를 위해 RAID(Redundant Array of Independent Disks) 시스템을 적용하고 있다. 특히 RAID 5는 패리티를 여러 저장장치에 분산 저장하여 공간 효율성과 안정성을 보장하였다. 그러나 저장장치의 용량이 서로 다를 경우 가장 작은 용량의 저장장치기준으로 RAID가 구축되어 저장공간의 낭비가 발생한다. 따라서 이러한 자원관리 문제를 해결하기 위한 연구가 필요하다. 본 논문에서는 SSD(Solid State Disk)로 구성된 RAID에서 각각의 독립적인 NAND 플래시 메모리 블록을 내부뿐만 아니라 외부 SSD와 RAID 그룹을 묶는 방법을 제안한다. 이 방법은 SSD 내부의 블록 정보를 RAID 시스템에 전달하는 정책과 RAID 시스템으로부터 전달된 물리적인 주소를 RAID 그룹으로 묶는 정책으로 구분된다. 이 방법을 통해 서로 다른 용량의 SSD를 RAID로 묶을 때 자원의 낭비가 발생하지 않는 RAID를 유지할 수 있다. 마지막으로 실험을 통해 제안하는 방법의 효과를 증명한다.

Keywords

Acknowledgement

This paper was supported by 2024 Baekseok University Research Fund

References

  1. T.Ma, Z.Li and N.Liu, "Log-ROC: Log Structured RAID on Open-Channel SSD," IEEE 40th International Conference on Computer Design, pp.332-335, 2022.
  2. Z.Gu, J.Li, Y.Peng, Y.Liu and T.Zhang, "Elastic RAID: Implementing RAID over SSDs with Built-in Transparent Compression," Proceedings of the 16th ACM International Conference on Systems and Storage, pp.83-93, 2023.
  3. G.Si, W.Chen, X.Xu, Y.Sun and R.Gao, "Data Back-Up Algorithm Based on Blockchain with Network Erasure Codes," IEEE 11th Joint International Information Technology and Artificial Intelligence Conference, pp.232-235, 2023.
  4. S.Wang, Q.Cao, Hong.Jiang, Z.Lu, J.Yao, Y.Chen and A.Pan, "Explorations and Exploitation for Parity-based RAIDs with Ultra-fast SSDs," ACM Transactions on Storage, Vol.20, No.6, pp.1-32, 2024. https://doi.org/10.1145/3627992
  5. S.Pashazadeh, L.N.Tazehkand and R.Soltani, "RSS_RAID a Novel Replicated Storage Schema for RAID System," Data Science: From Research to Application, Vol.45, pp.36-43, 2020. https://doi.org/10.1007/978-3-030-37309-2_4
  6. J.H.Choi, J.H.Park and S.J.Lee, "Reassembling Linux-based Hybrid RAID," Journal of Forensic Sciences, pp.966-973, 2019.
  7. R.Pitchumani and Y.S.Kee, "Hybrid Data Reliability for Emerging Key-Value Storage Devices," FAST '20 Open Access Sponsored by NetApp, 2020.
  8. Y.Cao, "Future Development of Redundant Array of Independent Disks," International Conference on Applied Physics and Computing, pp.8-10, 2022.
  9. H.S.Lee, "Performance analysis and prediction through various over-provision on NAND flash memory based storage," Journal of Digital Convergence, Vol.20, No.3, pp.343-348, 2022. https://doi.org/10.14400/JDC.2022.20.3.343
  10. H.S.Lee, "A Memory Mapping Technique to Reduce Data Retrieval Cost in the Storage Consisting of Multi Memories," Journal of Internet of Things and Convergence, Vol.9, No.1, pp.19-24, 2023.
  11. H.S.Lee, "A Study on Characteristics and Techniques that Affect Data Integrity for Digital Forensic on Flash Memory-Based Storage Devices," Journal of Internet of Things and Convergence, Vol.9, No.3, pp.7-12, 2023.
  12. H.S.Lee, "High Efficiency Life Prediction and Exception Processing Method of NAND Flash Memory-based Storage using Gradient Descent Method," Journal of Convergence for Information Technology, Vol.11, No.11, pp.44-50, 2021
  13. H.S.Lee, "A Safety IO Throttling Method Inducting Differential End of Life to Improving the Reliability of Big Data Maintenance in the SSD based RAID," Journal of Digital Convergence, Vol.20, No.5, pp.593-598, 2022. https://doi.org/10.14400/JDC.2022.20.5.593
  14. H.S.Lee, "An Efficient SLC Transition Method for Improving Defect Rate and Longer Lifetime on Flash Memory," Journal of Internet of Things and Convergence, Vol.9, No.3, pp.81-86, 2023.
  15. H.S.Lee, "An Efficient SLC Transition Method for Improving Defect Rate and Longer Lifetime on Flash Memory," Journal of Internet of Things and Convergence, Vol.9, No.3, pp.81-86, 2023.