采用数据降维的固态硬盘故障检测方法

王宇菲; 董小社; 王龙翔; 陈维多; 陈衡

doi:10.7652/xjtuxb202211018

您当前的位置：

首页 >

文章列表页 >

采用数据降维的固态硬盘故障检测方法

—————————————————————— | 更新时间：2025-07-09

- 采用数据降维的固态硬盘故障检测方法
- A Solid-State Drive Fault Detection Method with Dimensionality Reduction
- 西安交通大学学报 2022年56卷第11期页码：176-185
- 作者机构：
  
  西安交通大学电子与信息学部,西安,710049
- 作者简介：
  
  : 2022-05-31。作者简介: 王宇菲(1990—),男,博士生
  王龙翔(通信作者),男,工程师,硕士生导师。基金项目: 国家重点研发计划资助项目(2016YFB1000303)
- 基金信息：
  
  : 2022-05-31。作者简介: 王宇菲(1990—),男,博士生;王龙翔(通信作者),男,工程师,硕士生导师。基金项目: 国家重点研发计划资助项目(2016YFB1000303);国家自然科学基金资助项目(61972311)
- DOI：10.7652/xjtuxb202211018
  中图分类号： TP319
- 网络首发：2022-11-10，
  
  纸质出版：2022
- 稿件说明：
移动端阅览
王宇菲, 董小社, 王龙翔, 等. 采用数据降维的固态硬盘故障检测方法[J]. 西安交通大学学报, 2022,56(11):176-185.

WANG Yufei, DONG Xiaoshe, WANG Longxiang, et al. A Solid-State Drive Fault Detection Method with Dimensionality Reduction[J]. 2022, 56(11): 176-185.
王宇菲, 董小社, 王龙翔, 等. 采用数据降维的固态硬盘故障检测方法[J]. 西安交通大学学报, 2022,56(11):176-185. DOI： 10.7652/xjtuxb202211018.

WANG Yufei, DONG Xiaoshe, WANG Longxiang, et al. A Solid-State Drive Fault Detection Method with Dimensionality Reduction[J]. 2022, 56(11): 176-185. DOI： 10.7652/xjtuxb202211018.

摘要

针对固态硬盘SMART数据包含大量高维数据特征导致故障检测准确率不理想的问题

结合固态硬盘SMART数据的时序特点

基于门控循环单元对传统自动编码器的结构做了改进

提出了一种基于门控循环单元稀疏自动编码器降维的固态硬盘故障检测方法(GAL)。首先利用固态硬盘SMART数据训练GRUAE模型

然后利用GRUAE模型中的编码器作为降维的工具

对固态硬盘的原始高维SMART数据进行降维

减少固态硬盘原始SAMRT数据中噪声特征的影响并突出与数据特点更加相关的特征

以提高故障检测的准确率

最后基于降维过的低维SMART数据利用长短时记忆网络进行故障检测。选取阿里巴巴固态硬盘数据集

对多种人工智能算法在准确率、召回率和F

0.5

方面进行了比较。实验结果表明:相比于没有采用任何降维手段

采用GAL方法可使两种闪存类型的固态硬盘的故障检测准确率、召回率和F

0.5

分别提高4%、5%、4%和4%、8%、5%

分别达到97%、95%、97%和97%、96%、97%; GAL方法的故障检测准确率、召回率和F

0.5

分别超出WEFR对比方法53%、25%、43%。

Abstract

The massive high-dimensional SMART data attributes tend to cause a low precision in the detection of solid-state drive(SSD)faults. For this reason

this paper proposes an SSD fault detection method-GAL based on dimensionality reduction with GRU sparse auto-encoder based on an improved structure of the auto-encoder with gated recurrent unit(GRU)according to temporal characteristics of SSD SMART data. First

the GRUAE model is trained with SSD SMART data. Then the encoder of GRUAE model is used as the dimen

sionality reduction tool to reduce the original high-dimensional SSD SMART data. This is to reduce the influence of noise features in original SSD SAMRT data and highlight the features more relevant to data characteristics to improve the precision of fault detection. Finally

long short-term memory is used for fault detection with low-dimensional SSD SMART data. The precision

recall and F

0.5

score of various algorithms are compared based on the dataset from Alibaba. The experimental results show that GAL improves the fault detection precision

recall and F

0.5

-score of the two types of flash SSDs by 4%

4% and 4%

8% and 5%

respectively

reaching 97%

95%

97% and 97%

96%

97%

respectively

compared with the condition with no dimensionality reduction. In addition

the precision

recall and F

0.5

score of GAL are respectively 53%

25% and 43% higher than those of the method compared

关键词

Keywords

references

高聪明, 石亮, 刘凯, 等. 闪存固态硬盘系统结构与技术 [J]. 计算机研究与发展, 2021, 58(7): 1518-1532.

GAO Congming, SHI Liang, LIU Kai, et al. Architecture and technologies of flash memory based solid state drives [J]. Journal of Computer Research and Development, 2021, 58(7): 1518-1532.

王梓仲, 王海霞, 邵艾然, 等. 一种混合局部恢复码及Hitchhiker码的存储策略 [J]. 计算机学报, 2020, 43(4): 618-630.

WANG Zizhong, WANG Haixia, SHAO Airan, et al. A local-reconstruction-code-and-Hitchhiker-code-mixing storage scheme [J]. Chinese Journal of Computers, 2020, 43(4): 618-630.

NARAYANAN I, WANG Di, JEON M, et al. SSD failures in datacenters: what, when and why? [C]∥Proceedings of the 2016 ACM SIGMETRICS International Conference on Measurement and Modeling of Computer Science. New York, NY, USA: ACM, 2016: 407-408.

XU Yong, SUI Kaixin, YAO R, et al. Improving service availability of cloud systems by predicting disk error [C]∥Proceedings of the 2018 USENIX Conference on Usenix Annual Technical Conference. New York, NY, USA: ACM, 2018: 481-493.

GUNAWI H S, SUMINTO R O, SEARS R, et al. Fail-slow at scale: evidence of hardware performance faults in large production systems [J]. ACM Transactions on Storage, 2018, 14(3): 23.

万洪浩, 邓明翥, 肖侬, 等. 一种面向RAID6固态阵列的数据盘失效快速修复方法 [J]. 计算机研究与发展, 2015, 52(S2): 62-67.

WAN Honghao, DENG Mingzhu, XIAO Nong, et al. A fast recovery method for data disk failure in SSD-based RAID6 [J]. Journal of Computer Research and Development, 2015, 52(S2): 62-67.

HAN Shujie, LEE P P C, SHEN Zhirong, et al. Toward adaptive disk failure prediction via stream mining [C]∥2020 IEEE 40th International Conference on Distributed Computing Systems(ICDCS). Piscataway, NJ, USA: IEEE, 2020: 628-638.

ZHANG Ji, WANG Yuanzhang, WANG Yangtao, et al. Tier-scrubbing: an adaptive and tiered disk scrubbing scheme with improved MTTD and reduced cost [C]∥2020 57th ACM/IEEE Design Automation Conference(DAC). Piscataway, NJ, USA: IEEE, 2020: 1-6.

XIAO Jiang, XIONG Zhuang, WU Song, et al. Disk failure prediction in data centers via online learning [C]∥Proceedings of the 47th International Conference on Parallel Processing. New York, NY, USA: ACM, 2018: 35.

WANG Yufei, DONG Xiaoshe, WANG Longxiang, et al. Optimizing small-sample disk fault detection based on LSTM-GAN model [J]. ACM Transactions on Architecture and Code Optimization, 2022, 19(1): 13.

MAHDISOLTANI F, STEFANOVICI I, SCHROEDER B. Improving storage system reliability with proactive error prediction [C]∥Proceedings of the 2017 USENIX Conference on Usenix Annual Technical Conference. New York, NY, USA: ACM, 2017: 391-402.

ALTER J, XUE Ji, DIMNAKU A, et al. SSD failures in the field: symptoms, causes, and prediction models [C]∥Proceedings of the International Conference for High Performance Computing, Networking, Storage and Analysis. New York, NY, USA: ACM, 2019: 75.

SARKAR J, PETERSON C, SANAYEI A. Machine-learned assessment and prediction of robust solid state storage system reliability physics [C]∥2018 IEEE International Reliability Physics Symposium(IRPS). Piscataway, NJ, USA: IEEE, 2018: 3C.6-1-3C.6-8.

PULGAR F J, CHARTE F, RIVERA A J, et al. Choosing the proper autoencoder for feature fusion based on data complexity and classifiers: analysis, tips and guidelines [J]. Information Fusion, 2020, 54: 44-60.

陈小红, 李霞, 王娜. 高维多目标优化中基于稀疏特征选择的目标降维方法 [J]. 电子学报, 2015, 43(7): 1300-1307.

CHEN Xiaohong, LI Xia, WANG Na. Objective reduction with sparse feature selection for many objective optimization problem [J]. Acta Electronica Sinica, 2015, 43(7): 1300-1307.

宋昱, 孙文赟, 陈昌盛. 对数变换主成分分析的图像识别 [J]. 西安交通大学学报, 2021, 55(1): 33-42.

SONG Yu, SUN Wenyun, CHEN Changsheng. Logarithm transformation based principal component analysis for image recognition [J]. Journal of Xi'an Jiaotong University, 2021, 55(1): 33-42.

DUAN Feng, REN Xina, YANG Yikang. A gesture recognition system based on time domain features and linear discriminant analysis [J]. IEEE Transactions on Cognitive and Developmental Systems, 2021, 13(1): 200-208.

POURRAMEZAN R, HASSANI R, KARIMI H, et al. A real-time synchrophasor data compression method using singular value decomposition [J]. IEEE Transactions on Smart Grid, 2022, 13(1): 564-575.

PAUL A, CHAKI N. Dimensionality reduction of hyperspectral image using signal entropy and spatial information in genetic algorithm with discrete wavelet transformation [J]. Evolutionary Intelligence, 2021, 14(4): 1793-1802.

CHEN Jiufang, YUAN Ye, RUAN Tao, et al. Hyper-parameter-evolutionary latent factor analysis for high-dimensional and sparse data from recommender systems [J]. Neurocomputing, 2021, 421: 316-328.

ZHAO Xiaoli, JIA Minping, LIU Zheng. Semisupervised deep sparse auto-encoder with local and nonlocal information for intelligent fault diagnosis of rotating machinery [J]. IEEE Transactions on Instrumentation and Measurement, 2021, 70: 1-13.

GUO Jiao, LI Henghui, NING Jifeng, et al. Feature dimension reduction using stacked sparse auto-encoders for crop classification with multi-temporal, quad-pol SAR data [J]. Remote Sensing, 2020, 12(2): 321.

PULGAR F J, CHARTE F, RIVERA A J, et al. ClEnDAE: a classifier based on ensembles with built-in dimensionality reduction through denoising autoencoders [J]. Information Sciences, 2021, 565: 146-176.

GAO Zhaoqi, LI Chuang, LIU Naihao, et al. Large-dimensional seismic inversion using global optimization with autoencoder-based model dimensionality reduction [J]. IEEE Transactions on Geoscience and Remote Sensing, 2021, 59(2): 1718-1732.

LIOU C Y, CHENG Weichen, LIOU J W, et al. Autoencoder for words [J]. Neurocomputing, 2014, 139: 84-96.

ANSARI M S, BARTO V, LEE B. GRU-based deep learning approach for network intrusion alert prediction [J]. Future Generation Computer Systems, 2022, 128: 235-247.

XU Fan, HAN Shujie, LEE P P C, et al. General feature selection for failure prediction in large-scale SSD deployment [C]∥2021 51st Annual IEEE/IFIP International Conference on Dependable Systems and Networks(DSN). Piscataway, NJ, USA: IEEE, 2021: 263-270.

浏览量

下载量

CSCD

文章被引用时，请邮件提醒。

提交

工具集

关联资源

采用多尺度Mamba架构的装备剩余寿命预测方法

蜉蝣优化双通道网络在齿轮箱故障诊断中的应用

采用门控循环单元与深度进化策略的股票指数量化模型

离子液体改性燃料的电流体动力学效应研究

融合物理信息与深度算子网络的变压器绕组热-流场快速计算方法