Potential Failure Analysis and Maintenance for Computer Lab Hard Disks

Zongqing Lu

2009-01-01

Abstract:This paper discusses the logical and physical damages to computer hard disks.According to the failure analysis,several approaches for solution are proposed.

What problem does this paper attempt to address?

Failure Analysis of Computer Data Storage Disc Drive Systems

K.C. Tripathi

DOI: https://doi.org/10.31399/asm.fach.modes.c9001516

2019-06-01

Abstract:Abstract This paper deals with disk drive failures that occur in the interface area between the head and disk. The failures often lead to the loss of stored data and are characterized by circumferential microscratches that are usually visible to the unaided eye. The recording media in disk drives consists of a metal, glass, ceramic, or plastic substrate coated with a magnetic material. Data errors are classified as ‘soft’ or ‘hard’ depending on their correctability. Examination has shown that hard errors are the result of an abrasive wear process that begins with contact between head and disk asperities. The contact generates debris that, as it accumulates, increases contact pressure between the read-write head and the surface of the disk. Under sufficient pressure, the magnetic coating material begins wearing away, resulting in data loss.
Advancements and Best Practices in Hard Disk Drives (HDDs) Repair: A Comprehensive Review of Diagnostic Techniques, Repair Methods, and Preventive Strategies

Adekuba Usman

DOI: https://doi.org/10.69739/jcsp.v1i1.103

2024-09-19

Abstract:Hard Disk Drives (HDDs) remain a crucial component of data storage systems despite the advent of Solid State Drives (SSDs). However, the mechanical and electronic nature of HDDs makes them susceptible to various failures over time. This article provides a comprehensive examination of HDD repair, focusing on the anatomy and function of these devices, common causes of failure, diagnostic techniques, and effective repair methods. Mechanical failures, such as head crashes and spindle motor issues, electronic failures like PCB damage, and logical failures including file system corruption and bad sectors are explored in detail. Diagnostic approaches utilizing visual inspection, diagnostic software tools, hardware-based diagnostics, and SMART monitoring are discussed to aid in identifying issues accurately. Repair methods covered include data recovery techniques, mechanical repairs, electronic component replacement, and firmware fixes. Real-world case studies illustrate practical challenges and solutions in HDD repair, highlighting the complexities involved. Additionally, best practices and safety measures for handling and maintaining HDDs are provided to prevent future failures and ensure data integrity. This article aims to equip data recovery professionals, IT technicians, and tech enthusiasts with the knowledge and skills necessary to diagnose and repair HDD issues effectively. By integrating technical insights and practical approaches, it serves as a valuable resource for enhancing the reliability and longevity of HDDs in various applications.
Reliability Assessment of Data Storage in Cyber Physical Systems

Heping Jia,Yi Ding,Rui Peng

DOI: https://doi.org/10.1109/icsrs.2018.8688837

2018-01-01

Abstract:With the deep integration of cyber physical systems, data storage systems might suffer from both physical failures and cyber attacks which significantly impact the systems' reliability. The data can be partitioned into several data parts to enhance data security and multiple replications can be created for each data part in the storage systems. In this paper, both the failures of data storage devices and cyber attacks including data theft and data corruption are considered in the proposed system model. Moreover, the multi-valued decision diagram (MDD) technique is developed for considering chronological characteristics of sequential events to achieve the quantitative reliability assessment. A numerical example is provided to validate the proposed model and method.
Significance of Disk Failure Prediction in Datacenters

Jayanta Basak,Randy H. Katz

DOI: https://doi.org/10.48550/arXiv.1707.01952

2017-07-07

Abstract:Modern datacenters assemble a very large number of disk drives under a single roof. Even if economic and technical factors where to make individual drives more reliable (which is not at all clear, given the commoditization of the technology), their sheer numbers combined with their ever increasing utilization in a well-balanced design makes achieving storage reliability a major challenge. In this paper, we assess the challenge of storage system reliability in the modern datacenter, and demonstrate how good disk failure prediction models can significantly improve the reliability of such systems.

Distributed, Parallel, and Cluster Computing
What Can We Learn from Four Years of Data Center Hardware Failures?

Guosai Wang,Lifei Zhang,Wei Xu

DOI: https://doi.org/10.1109/DSN.2017.26

2017-01-01

Abstract:Hardware failures have a big impact on the dependability of large-scale data centers. We present studies on over 290,000 hardware failure reports collected over the past four years from dozens of data centers with hundreds of thousands of servers. We examine the dataset statistically to discover failure characteristics along the temporal, spatial, product line and component dimensions. We specifically focus on the correlations among different failures, including batch and repeating failures, as well as the human operatorsu0027 response to the failures. We reconfirm or extend findings from previous studies. We also find many new failure and recovery patterns that are the undesirable by-product of the state-of-the-art data center hardware and software design.
Classification Based Hard Disk Drive Failure Prediction: Methodologies, Performance Evaluation and Comparison.

Ruiyu Xu,Xinming Wang,Jianguo Wu

DOI: https://doi.org/10.1109/case49997.2022.9926720

2022-01-01

Abstract:Considering the reliability of the data storage system, it is essential to accurately and timely predict impending failures of hard disk drives (HDDs) so as to prevent data loss and reduce recovery cost. Over the past decades, taking as input the SMART (Self-Monitoring, Analysis and Reporting Technology) attributes, many supervised machine learning based methods have been proposed for HDD failure prediction. However, these methods are conducted on different datasets or different preprocessing treatments and thus lack comparative analysis. To fill this gap, we provide a systematic study in this paper on three key steps of the failure prediction, i.e., feature selection strategies, data preprocessing treatments and classification models. A feature selection strategy is proposed by testing the significance of difference between healthy and failed samples. Data relabeling, together with some other data preprocessing treatments are applied and proven to be effective in the case study. The performance of seven classification models are compared, among which the Random Forest model achieves the best performance with 53.95% failure detection rate (FDR) and 6.0% false alarm rate (FAR). Moreover, the Gini importance of SMART attributes is calculated, where two attributes, SMART 197 and SMART 187 are found closely related to the HDD failures.
Reliability failure analysis of CNC lathes

Xiao Jun,Hu Wei,Lu Zhizheng,Chen Zhijun,Yang Jianguo

DOI: https://doi.org/10.16731/j.cnki.1671-3133.2006.12.012

2006-01-01

Abstract:According to domestic CNC lathes,the failure of CNC lathes is analyzed from failure part,failure mode and failure reason,The CNC lathes are divided into ten subsystems.The failure probability of every subsystem is counted by data to find out the subsystems that often have many failures.Failure mode and failure reason are analyzed deeply for making reliability improvement design.Key parts and defective elements are found out by criticality analysis to make reliability allotment and improvement design of CNC lathes according to practical condition
Hard Drive Failure Prediction Using Big Data

Wenjun Yang,Dianming Hu,Yuliang Liu,Shuhao Wang,Tianming Jiang

DOI: https://doi.org/10.1109/SRDSW.2015.15

2015-01-01

Abstract:We design a general framework named Hdoctor for hard drive failure prediction. Hdoctor leverages the power of big data to achieve a significant improvement comparing to all previous researches that used sophisticated machine learning algorithms. Hdoctor exhibits a series of engineering innovations: (1) constructing time dependent features to characterize the Self-Monitoring, Analysis and Reporting Technology (SMART) value transitions during disk failures, (2) combining features to enable the model to learn the correlation among different SMART attributes, (3) regarding circumstance data such as cluster workload, temperature, humidity, location as related features. Meanwhile, Hdoctor collects/labels samples and updates model automatically, and works well for all kinds of disk failure prediction in our intelligent data center. In this work, we use Hdoctor to collect 74,477,717 training records from our clusters involving 220,022 disks. By training a simple and scalable model, our system achieves a detection rate of 97.82%, with a false alarm rate (FAR) of 0.3%, which hugely outperforms all previous algorithms. In addition, Hdoctor is an excellent indicator for how to predict different hardware failures efficiently under various circumstances.
ADF2T: an Active Disk Failure Forecasting and Tolerance Software

Hongzhang Yang,Yahui Yang,Zhengguang Chen,Zongzhao Li,Yaofeng Tu

DOI: https://doi.org/10.1109/issrew51248.2020.00030

2020-01-01

Abstract:The reliability of distributed file system is inevitably affected by hard disk failure. This paper proposes an active disk failure forecasting and tolerance software. Firstly, multiple SMART records in the time window are merged into one sample, and after sliding, tens of times of positive samples are created. Secondly, the features are selected by two-stage sorting method, so that the most conducive features are used in machine learning modeling, and the time for model training can be shortened obviously. Thirdly, through two-stage verification, parameters can be adjusted in time for unreasonable proactive reconstruction strategies. Experiments show that modeling and forecast of ZTE data set and Backblaze data set respectively, the recall rate is 95.66% and 84.28%, and the error rate is 0.23% and 2.45%. The work in this paper has been commercially used for more than one year in ZTE data center. The reliability of distributed file system software is significantly improved.
Reliability Analysis of Distributed Storage Systems Considering Data Loss and Theft

Heping Jia,Rui Peng,Yi Ding,Changzheng Shao

DOI: https://doi.org/10.1177/1748006x19885508

2020-01-01

Abstract:With the advancement of cloud computing and internet of things, data are usually stored on distributed computers and these data may risk being lost or stolen. In this article, we consider a common case where the entirety of the data is partitioned into several parts and each data part can be allocated to one or more computers. In the case where a computer fails, all the data parts on it are lost. Before the failure of any computer, the data parts may also be stolen by hackers. The basic model of computer failure and computer intrusion resulting in the theft of all the data parts on the computer is considered first. Then, the case is extended to a general model where computer failure, as well as data part corruption and theft caused by hacking are embedded. It is essential to study the reliability of distributed storage systems considering both data loss and data theft, which can be a basis for decision making on system structure optimization. In this article, a multi-valued decision diagram–based approach is developed to quantitatively evaluate system reliability for both models considering the time-dependence property of sequential events. The proposed method is applicable to systems where the random time to failure, theft, or corruption follows arbitrary distributions including the commonly used exponential distributions. Illustrative examples are provided to validate the proposed method.
Hard Disk Failure Prediction Based on Lightgbm with CID.

Hengrui Wang,Yahui Yang,Hongzhang Yang

DOI: https://doi.org/10.1109/iscc53001.2021.9631504

2021-01-01

Abstract:In data centers, hard disks are the most prone to failure of IT equipment. Although there is data backup, data reliability still faces challenges due to hard disks failure. In recent years, many hard disk failure prediction approaches based on SMART data have been proposed. In this paper, we proposed a novel disk failure prediction approach based on Lightgbm algorithm with CID (complexity invariant distance). Our failure prediction model has been built and evaluated on SMART data of about 80,000 hard disks from two manufacturers. The experimental result shows that by adding CID features, the TPR is increased from 0.28 to 0.96, and the number of days that the model can predict failures in advance is extended by 1.2 days. Compared with the several existing failure prediction models, our model has better performance on AUC score, f1-score and TPR.
The Life and Death of SSDs and HDDs: Similarities, Differences, and Prediction Models

Riccardo Pinciroli,Lishan Yang,Jacob Alter,Evgenia Smirni

DOI: https://doi.org/10.48550/arXiv.2012.12373

IF: 5.414

2020-12-22

Machine Learning

Abstract:Data center downtime typically centers around IT equipment failure. Storage devices are the most frequently failing components in data centers. We present a comparative study of hard disk drives (HDDs) and solid state drives (SSDs) that constitute the typical storage in data centers. Using a six-year field data of 100,000 HDDs of different models from the same manufacturer from the BackBlaze dataset and a six-year field data of 30,000 SSDs of three models from a Google data center, we characterize the workload conditions that lead to failures and illustrate that their root causes differ from common expectation but remain difficult to discern. For the case of HDDs we observe that young and old drives do not present many differences in their failures. Instead, failures may be distinguished by discriminating drives based on the time spent for head positioning. For SSDs, we observe high levels of infant mortality and characterize the differences between infant and non-infant failures. We develop several machine learning failure prediction models that are shown to be surprisingly accurate, achieving high recall and low false positive rates. These models are used beyond simple prediction as they aid us to untangle the complex interaction of workload characteristics that lead to failures and identify failure root causes from monitored symptoms.
A Failure Prediction Approach Based on BiLSTM and Deep Feature Extractor for Hard Disk Drives.

Dongjiang Li,Wenyuan Qiu,Jing Zhang,Kun Xue,Xianbo Zhang,Binbin Sun,Feng Lin,Li Li

DOI: https://doi.org/10.1109/ICISCAE59047.2023.10393131

2023-01-01

Abstract:With the swift implementation of cloud platforms, guaranteeing optimal service dependability is imperative. Disk failure is a common cause of service unreliability, particularly in industrial cloud platforms that house numerous disks. These approaches can forecast disk failures by analyzing disk status data before they occur. Deep neural networks have demonstrated their effectiveness in addressing time series classification problems. Our research paper introduces an innovative and efficient approach to forecasting disk failures. By incorporating Bidirectional Long Short Term Memory (BiLSTM) into deep neural networks, we enhance the ability to extract relevant features. We explore the use of very neural networks with the ResNet structure and Fully Convolutional Network. The attention mechanism is adopted to support BiLSTM in acquiring long-term dependencies within lengthy sequences. Our model’s effectiveness is evaluated on two datasets and compared to advanced approaches. The empirical findings unequivocally establish that our performance surpasses that of alternative techniques for predicting disk failures.
Fuzzy Reliability Analysis of Two-Disk Failure-Tolerant Disk Arrays

MingHua Jiang,Jingli Zhou,Ming Hu,YiXiang Ding

DOI: https://doi.org/10.1109/iccit.2007.219

2007-01-01

Abstract:This paper presents a data placement of a two-disk failure-tolerant disk arrays system, a code-mixing solution RAID5x with XOR-based code and mirror for disk arrays to improve performance of accessing storage system with high reliability. RAID5x is not only just higher in storage efficiency than other code-mixing solutions of the same type, but also it has the higher accessing performance over them and XOR-based code solutions, and then analyzed the reliability of this system by combination of fuzzy logic and Markov modeling. This reliability method is a technique for analyzing fault tolerant designs under considerable uncertainty, such as is seen in compilations of component failure rates, the presented model provides the estimation of the lower and upper boundary of a two-disk failure-tolerant disk array system with a single run of the model.
Research on Safety Resilience Evaluation Model of Data Center Physical Infrastructure: An ANP-Based Approach

Xiaer Xiahou,Jialong Chen,Bangyi Zhao,Zixuan Yan,Peng Cui,Qiming Li,Zhou Yu

DOI: https://doi.org/10.3390/buildings12111911

IF: 3.324

2022-11-08

Buildings

Abstract:With the development of the digital economy, the number and scale of data centers are expanding rapidly. Data centers are playing an increasingly important role in social and economic development. However, a short downtime of a data center may result in huge losses. The safety management of data centers' physical infrastructure is of great significance to address this concern. We applied resilience theory to the safety management of data center physical infrastructures. We analyzed the resilience connotation and evaluated the system resilience using the resilience indexes. The data center infrastructure was regarded as a system of systems. Through theoretical analysis, the resilience framework of data center infrastructures was established, which formed the main dimensions of resilience assessment. The Delphi method determined the resilience indices, and the ANP method was adopted to set up the evaluation model. The results revealed the important indexes affecting data center infrastructure system safety resilience. Based on the findings, this paper argues for improving redundancy and adaptability, paying attention to the resilience management of energy flow and thermal flow, and establishing an automatic systematic data management system. These suggested measures would not only effectively make contributions to the data center infrastructure safety management theory but also provide an important reference for construction industry practices.

construction & building technology,engineering, civil
An In-Depth Study Of Correlated Failures In Production Ssd-Based Data Centers

Shujie Han,Patrick P. C. Lee,Fan Xu,Yi Liu,Cheng He,Jiongzhou Liu

2021-01-01

Abstract:Flash-based solid-state drives (SSDs) are increasingly adopted as the mainstream storage media in modern data centers. However, little is known about how SSD failures in the field are correlated, both spatially and temporally. We argue that characterizing correlated failures of SSDs is critical, especially for guiding the design of redundancy protection for high storage reliability. We present an in-depth data-driven analysis on the correlated failures in the SSD-based data centers at Alibaba. We study nearly one million SSDs of 11 drive models based on a dataset of SMART logs, trouble tickets, physical locations, and applications. We show that correlated failures in the same node or rack are common, and study the possible impacting factors on those correlated failures. We also evaluate via trace-driven simulation how various redundancy schemes affect the storage reliability under correlated failures. To this end, we report 15 findings. Our dataset and source code are now released for public use.
Using machine learning to forecast hard drive failures

Nikita Kodatsky,Elena Revyakina,Andrei Gazizov,Vitaliy Gunko

DOI: https://doi.org/10.1051/e3sconf/202454908024

2024-07-16

E3S Web of Conferences

Abstract:The aim of this research is to develop an effective tool for assessing the condition of computer hardware based on an intelligent method. The outcome of this work not only facilitates the implementation of a data analysis tool into the realm of computer equipment servicing but also offers a practical example of its use for enhancing the reliability and efficiency of information systems. The findings of this research can be valuable for IT specialists responsible for technical support as well as for organizations seeking to optimize equipment servicing processes and improve their competitiveness in the market.
Investigating Power Outage Effects on Reliability of Solid-State Drives

Saba Ahmadian,Farhad Taheri,Mehrshad Lotfi,Maryam Karimi,Hossein Asad

DOI: https://doi.org/10.23919/DATE.2018.8342004

2018-04-29

Abstract:Solid-State Drives (SSDs) are recently employed in enterprise servers and high-end storage systems in order to enhance performance of storage subsystem. Although employing high speed SSDs in the storage subsystems can significantly improve system performance, it comes with significant reliability threat for write operations upon power failures. In this paper, we present a comprehensive analysis investigating the impact of workload dependent parameters on the reliability of SSDs under power failure for variety of SSDs (from top manufacturers). To this end, we first develop a platform to perform two important features required for study: a) a realistic fault injection into the SSD in the computing systems and b) data loss detection mechanism on the SSD upon power failure. In the proposed physical fault injection platform, SSDs experience a real discharge phase of Power Supply Unit (PSU) that occurs during power failure in data centers which was neglected in previous studies. The impact of workload dependent parameters such as workload Working Set Size (WSS), request size, request type, access pattern, and sequence of accesses on the failure of SSDs is carefully studied in the presence of realistic power failures. Experimental results over thousands number of fault injections show that data loss occurs even after completion of the request (up to 700ms) where the failure rate is influenced by the type, size, access pattern, and sequence of IO accesses while other parameters such as workload WSS has no impact on the failure of SSDs.

Other Computer Science
Failure Analysis and Quantification for Contemporary and Future Supercomputers

Li Tan,Nathan DeBardeleben

DOI: https://doi.org/10.48550/arXiv.1911.02118

2019-11-06

Abstract:Large-scale computing systems today are assembled by numerous computing units for massive computational capability needed to solve problems at scale, which enables failures common events in supercomputing scenarios. Considering the demanding resilience requirements of supercomputers today, we present a quantitative study on fine-grained failure modeling for contemporary and future large-scale computing systems. We integrate various types of failures from different system hierarchical levels and system components, and summarize the overall system failure rates formally. Given that nowadays system-wise failure rate needs to be capped under a threshold value for reliability and cost-efficiency purposes, we quantitatively discuss different scenarios of system resilience, and analyze the impacts of resilience to different error types on the variation of system failure rates, and the correlation of hierarchical failure rates. Moreover, we formalize and showcase the resilience efficiency of failure-bounded supercomputers today.

Distributed, Parallel, and Cluster Computing
Reliability analysis of the main drive system of a CNC machine tool including early failures

He Li,Zhi-Ming Deng,Noorbakhsh Amiri Golilarz,C. Guedes Soares

DOI: https://doi.org/10.1016/j.ress.2021.107846

2021-11-01

Abstract:<p>Early failures occur in the initial operation period of complicated systems such as Main Drive Systems (MDSs) of Computerized Numerical Control machine tools (CNC machine tools). In this paper, a Bayesian network model is developed to conduct a comprehensive reliability analysis of the MDS of a heavy boring and milling CNC machine tool, in which early failures of the MDS is considered. The primary contributions of this study over the existing analyses are: (i) the early failure of the MDS is investigated. (ii) the reliability analysis is conducted under the complicated system assumption, accordingly, multiple working states of the MDS and its subsystems are considered that are working, have a soft failure or a hard failure. (iii) reliability of the MDS in different working stages and for various manufacturing tasks are analysed. With the Bayesian network model, reliability and mean time to failure of the MDS and its subsystems are predicted. Meanwhile, this study identified risky failure items that potentially give rise to malfunctions of the MDS. The error of the predicted results is 8% at the early-wear stage and 10.5% at the stable-working stage when comparison with collected field data. Recommendations on improvements of maintenance and inspection activities are suggested, which may play a role in overall cost saving and guarantee the reliability of the MDS of the heavy boring and milling CNC machine tool.</p>

engineering, industrial,operations research & management science

Potential Failure Analysis and Maintenance for Computer Lab Hard Disks

Failure Analysis of Computer Data Storage Disc Drive Systems

Advancements and Best Practices in Hard Disk Drives (HDDs) Repair: A Comprehensive Review of Diagnostic Techniques, Repair Methods, and Preventive Strategies

Reliability Assessment of Data Storage in Cyber Physical Systems

Significance of Disk Failure Prediction in Datacenters

What Can We Learn from Four Years of Data Center Hardware Failures?

Classification Based Hard Disk Drive Failure Prediction: Methodologies, Performance Evaluation and Comparison.

Reliability failure analysis of CNC lathes

Hard Drive Failure Prediction Using Big Data

ADF2T: an Active Disk Failure Forecasting and Tolerance Software

Reliability Analysis of Distributed Storage Systems Considering Data Loss and Theft

Hard Disk Failure Prediction Based on Lightgbm with CID.

The Life and Death of SSDs and HDDs: Similarities, Differences, and Prediction Models

A Failure Prediction Approach Based on BiLSTM and Deep Feature Extractor for Hard Disk Drives.

Fuzzy Reliability Analysis of Two-Disk Failure-Tolerant Disk Arrays

Research on Safety Resilience Evaluation Model of Data Center Physical Infrastructure: An ANP-Based Approach

An In-Depth Study Of Correlated Failures In Production Ssd-Based Data Centers

Using machine learning to forecast hard drive failures

Investigating Power Outage Effects on Reliability of Solid-State Drives

Failure Analysis and Quantification for Contemporary and Future Supercomputers

Reliability analysis of the main drive system of a CNC machine tool including early failures