[go: up one dir, main page]

CN113688564A - Method, device, terminal and storage medium for predicting remaining life of SSD (solid State disk) - Google Patents

Method, device, terminal and storage medium for predicting remaining life of SSD (solid State disk) Download PDF

Info

Publication number
CN113688564A
CN113688564A CN202110872186.XA CN202110872186A CN113688564A CN 113688564 A CN113688564 A CN 113688564A CN 202110872186 A CN202110872186 A CN 202110872186A CN 113688564 A CN113688564 A CN 113688564A
Authority
CN
China
Prior art keywords
hard disk
performance state
training
weighted value
count
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Granted
Application number
CN202110872186.XA
Other languages
Chinese (zh)
Other versions
CN113688564B (en
Inventor
孔涛
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Inspur Jinan data Technology Co ltd
Original Assignee
Inspur Jinan data Technology Co ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Inspur Jinan data Technology Co ltd filed Critical Inspur Jinan data Technology Co ltd
Priority to CN202110872186.XA priority Critical patent/CN113688564B/en
Publication of CN113688564A publication Critical patent/CN113688564A/en
Application granted granted Critical
Publication of CN113688564B publication Critical patent/CN113688564B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F30/00—Computer-aided design [CAD]
    • G06F30/20—Design optimisation, verification or simulation
    • G06F30/27—Design optimisation, verification or simulation using machine learning, e.g. artificial intelligence, neural networks, support vector machines [SVM] or training a model
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q10/00—Administration; Management
    • G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F2119/00—Details relating to the type or aim of the analysis or the optimisation
    • G06F2119/04—Ageing analysis or optimisation against ageing

Landscapes

  • Engineering & Computer Science (AREA)
  • Business, Economics & Management (AREA)
  • Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Evolutionary Computation (AREA)
  • Human Resources & Organizations (AREA)
  • Economics (AREA)
  • Strategic Management (AREA)
  • Computer Hardware Design (AREA)
  • Medical Informatics (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Geometry (AREA)
  • Artificial Intelligence (AREA)
  • Software Systems (AREA)
  • Game Theory and Decision Science (AREA)
  • General Engineering & Computer Science (AREA)
  • Development Economics (AREA)
  • Entrepreneurship & Innovation (AREA)
  • Marketing (AREA)
  • Operations Research (AREA)
  • Quality & Reliability (AREA)
  • Tourism & Hospitality (AREA)
  • General Business, Economics & Management (AREA)
  • Debugging And Monitoring (AREA)

Abstract

The invention discloses a method, a device, a terminal and a storage medium for predicting the residual life of an SSD (solid State disk). A plurality of performance state parameters of each training hard disk are obtained based on a hard disk unified standard; establishing a parameter weighted value model for training to obtain the optimal weighted value of each performance state parameter; establishing a function model of performance state parameter weighting and hard disk service life duration based on the optimal weighted value for training to obtain a hard disk service life prediction function model; and predicting the residual life of the hard disk to be predicted based on the hard disk life prediction function model. The performance state parameters for service life prediction are acquired based on the unified hard disk standard, so that the method can be suitable for different types of hard disks produced by different manufacturers, has wide application value, reduces the test cost, provides reliable basis for hard disk maintenance, and effectively reduces the risk of data loss or server downtime caused by hard disk faults.

Description

Method, device, terminal and storage medium for predicting remaining life of SSD (solid State disk)
Technical Field
The invention relates to the field of SSD hard disk life prediction, in particular to a method, a device, a terminal and a storage medium for predicting the residual life of an SSD hard disk.
Background
At present, SSD hard disks are adopted in a server, and in the operation process of the server, the hard disks can cause faults due to various problems such as heat dissipation, vibration and the like, so that the results of hard disk data loss or server downtime and the like are caused, and great influence is brought to users. Therefore, the service life of the hard disk needs to be predicted so as to immediately know the service life condition of the hard disk and immediately replace the hard disk, so as to avoid the consequences of data loss or server downtime and the like caused by hard disk failure. However, the hard disks are of various types, and how to configure a unified prediction method for various types of hard disks is a problem faced by testers.
Disclosure of Invention
In order to solve the above problems, the present invention provides a method, an apparatus, a terminal and a storage medium for predicting the remaining life of an SSD hard disk, which are applicable to different types of hard disks to perform life prediction on different types of hard disks.
In a first aspect, a technical solution of the present invention provides a method for predicting a remaining life of an SSD hard disk, including the steps of:
acquiring a plurality of performance state parameters of each training hard disk based on a hard disk unified standard;
establishing a parameter weighted value model for training to obtain the optimal weighted value of each performance state parameter;
establishing a function model of performance state parameter weighting and hard disk service life duration based on the optimal weighted value for training to obtain a hard disk service life prediction function model;
and predicting the residual life of the hard disk to be predicted based on the hard disk life prediction function model.
Further, establishing a parameter weighted value model for training to obtain the optimal weighted value of each performance state parameter specifically includes:
the performance state parameters of the training hard disk are respectively S1、S2……Sk；
The weighted value of each performance state parameter is A1、A2……Ak(ii) a Wherein
Figure BDA0003189181170000021
And 0 < An< H; h is a positive number;
the performance state parameter is weighted and summed
Figure BDA0003189181170000022
Taking P ten thousand training hard disks, wherein N fault hard disks are included;
every random change of a set of weight values a1、A2……AkCalculating P ten thousand M values once, and taking out the mode M from the P ten thousand M valuesxWherein the mode M is satisfiedxThe number of failed hard disks of (2) is NyThen (M)x，Ny) Can be recorded as a point in the M/N plane coordinate;
weighted value A1、A2……AkVarying groups to obtain points (M)x，Ny) A scatter point distribution map;
the highest point in the scatter plot is denoted as (M)max，Nmax) Then M ismaxThe corresponding weighted value is the optimal weighted value, and is marked as A1i、A2i……Aki。
Further, a function model of performance state parameter weighting and hard disk life duration is established based on the optimal weighting value for training, and a hard disk life prediction function model is obtained, and the method specifically comprises the following steps:
taking Q ten thousand fault hard disks, and obtaining the stored time length of each fault hard disk and recording the time length as Li；
According to the formula
Figure BDA0003189181170000023
Calculating the M value of each fault hard disk and recording the M value as MiObtaining a point (M)i，Li) Fall into the M/L plane coordinates;
according to Q ten thousand points (M)i，Li) Simulating a linear function
Figure BDA0003189181170000024
Wherein X, Y, Z are three constants; the linear function is a hard disk life prediction function model.
Further, the performance state parameters include a remapped sector count, a spindle spin-up retry number, a head calibration retry count, an erase-fail block count, a point-to-point error detection count, an uncorrectable error count, a hardware ECC correction count, a current to-be-mapped sector count, an offline uncorrectable sector count, and an ULTRA DMA parity error rate.
In a second aspect, the present invention provides an apparatus for predicting the remaining life of an SSD hard disk, comprising,
a performance state parameter acquisition module: acquiring a plurality of performance state parameters of each training hard disk based on a hard disk unified standard;
an optimal weighted value training module: establishing a parameter weighted value model for training to obtain the optimal weighted value of each performance state parameter;
the hard disk life prediction function module training module: establishing a function model of performance state parameter weighting and hard disk service life duration based on the optimal weighted value for training to obtain a hard disk service life prediction function model;
hard disk remaining life prediction module: and predicting the residual life of the hard disk to be predicted based on the hard disk life prediction function model.
Further, the optimal weighted value training module establishes a parameter weighted value model for training to obtain the optimal weighted value of each performance state parameter, and specifically includes:
the performance state parameters of the training hard disk are respectively S1、S2……Sk；
The weighted value of each performance state parameter is A1、A2……Ak(ii) a Wherein
Figure BDA0003189181170000031
Mesh 0 < An< H; h is a positive number;
the performance state parameter is weighted and summed
Figure BDA0003189181170000032
Taking P ten thousand training hard disks, wherein N fault hard disks are included;
every random change of a set of weight values a1、A2......AkCalculating P ten thousand M values at a time, and taking out the P ten thousand M valuesMode MxWherein the mode M is satisfiedxThe number of failed hard disks of (2) is NyThen (M)x，Ny) Can be recorded as a point in the M/N plane coordinate;
weighted value A1、A2……AkVarying groups to obtain points (M)x，Ny) A scatter point distribution map;
the highest point in the scatter plot is denoted as (M)max，Nmax) Then M ismaxThe corresponding weighted value is the optimal weighted value, and is marked as A1i、A2i……Aki。
Further, the hard disk life prediction function module training module establishes a function model of performance state parameter weighting and hard disk life duration based on the optimal weighting value for training to obtain a hard disk life prediction function model, and the method specifically comprises the following steps:
taking Q ten thousand fault hard disks, and obtaining the stored time length of each fault hard disk and recording the time length as Li；
According to the formula
Figure BDA0003189181170000041
Calculating the M value of each fault hard disk and recording the M value as MiObtaining a point (M)i，Li) Fall into the M/L plane coordinates;
according to Q ten thousand points (M)i，Li) Simulating a linear function
Figure BDA0003189181170000042
Wherein X, Y, Z are three constants; the linear function is a hard disk life prediction function model.
Further, the performance state parameters include a remapped sector count, a spindle spin-up retry number, a head calibration retry count, an erase-fail block count, a point-to-point error detection count, an uncorrectable error count, a hardware ECC correction count, a current to-be-mapped sector count, an offline uncorrectable sector count, and an ULTRA DMA parity error rate.
In a third aspect, a technical solution of the present invention provides a terminal, including:
a processor;
a memory for storing instructions for execution by the processor;
wherein the processor is configured to perform any of the methods described above.
In a fourth aspect, the invention provides a computer-readable storage medium storing a computer program which, when executed by a processor, implements the method of any one of the above.
Compared with the prior art, the method, the device, the terminal and the storage medium for predicting the residual life of the SSD have the following beneficial effects: acquiring performance state parameters of the hard disk based on a hard disk unified standard, configuring weighted values for the performance state parameters and training an optimal weighted value, then training a hard disk life prediction function model according to the performance state parameter weighted values and the relation between the performance state parameter weighted values and the hard disk life duration, and predicting the residual life of the hard disk to be predicted based on the hard disk life prediction function model. The performance state parameters for service life prediction are acquired based on the unified hard disk standard, so that the method can be suitable for different types of hard disks produced by different manufacturers, has wide application value, reduces the test cost, provides reliable basis for hard disk maintenance, and effectively reduces the risk of data loss or server downtime caused by hard disk faults.
Drawings
For a clearer explanation of the embodiments or technical solutions of the prior art of the present application, the drawings needed for the description of the embodiments or prior art will be briefly described below, it is obvious that the drawings in the following description are only some embodiments of the present application, and for those skilled in the art, other drawings can be obtained according to these drawings without creative efforts.
Fig. 1 is a schematic flowchart of a method for predicting the remaining life of an SSD hard disk according to an embodiment of the present invention;
FIG. 2 shows a point (M) in M/N plane coordinates according to an embodiment of the present inventionx，Ny) A schematic diagram;
FIG. 3 shows the present inventionEXAMPLE A point (M)x，Ny) A scatter point distribution map;
FIG. 4 is a diagram of a M/L plane coordinate midpoint (M) according to an embodiment of the present inventioni，Li) A schematic diagram;
FIG. 5 shows a point (M) of an embodiment of the present inventioni，Li) A scatter point distribution map;
fig. 6 is a schematic block diagram of a structure of an apparatus for predicting the remaining life of an SSD hard disk according to a second embodiment of the present invention;
fig. 7 is a schematic structural diagram of a terminal according to a third embodiment of the present invention.
Detailed Description
The following explains the english terms related to the present invention.
SSD hard disk: solid State Disk or Solid State Drive, Solid State Disk;
smart technical standard: Self-Monitoring Analysis and Reporting Technology, which is an automatic hard disk state detection and early warning system and specification;
ECC correction: a memory error correction principle;
ULTRA DMA: ultra Direct Memory Access, advanced Direct Memory Access.
In order that those skilled in the art will better understand the disclosure, the following detailed description will be given with reference to the accompanying drawings. It is to be understood that the embodiments described are only a few embodiments of the present application and not all embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present application.
Example one
At present, SSD hard disks are adopted in a server, and in the operation process of the server, the hard disks can cause faults due to various problems such as heat dissipation, vibration and the like, so that the results of hard disk data loss or server downtime and the like are caused, and great influence is brought to users. Therefore, the service life of the hard disk needs to be predicted so as to immediately know the service life condition of the hard disk and immediately replace the hard disk, so as to avoid the consequences of data loss or server downtime and the like caused by hard disk failure. However, the hard disks are of various types, and how to configure a unified prediction method for various types of hard disks is a problem faced by testers.
Therefore, the present embodiment provides a method for predicting the remaining life of an SSD hard disk, which is applicable to different types of hard disks to predict the life of different types of hard disks.
Considering that each type of hard disk has a set of unified standard, such as smart technical standard, and the performance state parameters affecting the hard disk life can be obtained based on the unified standard, the principle of the method of the embodiment is to obtain a plurality of performance state parameters based on the unified standard, and establish a model between the performance state parameters and the hard disk life, so that the hard disk life can be predicted according to the performance state parameters. Certainly, the number of the performance state parameters is multiple, the weight is configured for each performance state parameter, in order to improve the prediction accuracy, the optimal weighted value of each performance state parameter is trained firstly, and then the function model of the performance state parameter weighting and the hard disk service life duration is established based on the optimal weighted value for training to obtain the hard disk service life prediction function model.
Fig. 1 is a schematic flow chart of a method for predicting the remaining life of an SSD hard disk according to a first embodiment of the present invention, which includes the following steps.
S1, acquiring a plurality of performance state parameters of each training hard disk based on the hard disk unified standard;
s2, establishing a parameter weighted value model for training to obtain the optimal weighted value of each performance state parameter;
s3, establishing a function model of performance state parameter weighting and hard disk life duration based on the optimal weighting value for training, and obtaining a hard disk life prediction function model;
and S4, predicting the residual life of the hard disk to be predicted based on the hard disk life prediction function model.
In this embodiment, a method for predicting the remaining life of an SSD hard disk obtains performance state parameters of the hard disk based on a unified hard disk standard, configures weighted values for the performance state parameters and trains an optimal weighted value, then trains a hard disk life prediction function model according to the performance state parameter weighted values and a relationship with the hard disk life duration, and performs remaining life prediction on the hard disk to be predicted based on the hard disk life prediction function model. The performance state parameters for service life prediction are acquired based on the unified hard disk standard, so that the method can be suitable for different types of hard disks produced by different manufacturers, has wide application value, reduces the test cost, provides reliable basis for hard disk maintenance, and effectively reduces the risk of data loss or server downtime caused by hard disk faults.
In order to improve the prediction accuracy, a plurality of performance state parameters of the hard disk are extracted as a basis to establish the relationship between the performance state parameters and the hard disk before the service life, but the influence degrees of the performance state parameters are different, so the weight is configured for each performance state parameter. Further, the optimal weight should be configured for each performance status parameter, so the weighted value of each performance status parameter needs to be trained first to obtain the optimal weighted value.
In some embodiments, the parameter weighted value model is established for training to obtain the optimal weighted value of each performance status parameter.
(1) The performance state parameters of the training hard disk are respectively S1、S2……Sk。
And k is the total number of the performance state parameters, and the number of the specifically selected performance state parameters is selected according to the requirement.
If a new hard disk, SnIs 0, but as the hard disk is used, some of the parameter values may increase, which may cause the hard disk to malfunction when one or more of the parameter values reach a threshold value. Then the hard disk fails and one or more of the entries has a value greater than 0.
For a hard disk meeting the smart technology standard, the performance state parameters can be obtained from the smart log.
(2) The weighted value of each performance state parameter is A1、A2......Ak(ii) a Wherein
Figure BDA0003189181170000081
And 0 < An< H; h is a positive number.
The specific value of H is set as needed, for example, H may be set to 100.
(3) The performance state parameter is weighted and summed
Figure BDA0003189181170000082
According to the formula
Figure BDA0003189181170000083
Carrying out model training, wherein in the training process, AnIs randomly varied between 0 and 100, but
Figure BDA0003189181170000084
Is constant.
(4) P ten thousand training hard disks are taken, wherein N fault hard disks are included.
(5) Each random variation of a set of weight values A1、A2......AkCalculating P ten thousand M values once, and taking out the mode M from the P ten thousand M valuesxWherein the mode M is satisfiedxThe number of failed hard disks of (2) is NyThen (M)x，Ny) Can be recorded as a point in the M/N plane coordinates.
It will be appreciated that the value of M at each time is according to the formula
Figure BDA0003189181170000085
And (4) calculating.
As shown in fig. 2, is a point (M) in the M/N plane coordinatex，Ny) Schematic representation.
(6) Weighted value A1、A2......AkVarying groups to obtain points (M)x，Ny) A scatter plot of.
(7) The highest point in the scatter plot is denoted as (M)max，Nmax) Then M ismaxThe corresponding weighted value is the optimal weighted value, and is marked as A1i、A2i......Aki。
With a weight value of A1、A2……AkThe continuous random change causes more and more points to fall into the M/N plane coordinate, according to the point (M)x，Ny) The distribution of (A) can be a scatter distribution diagram, as shown in FIG. 3, which is a point (M)x，Ny) A scatter plot of. The scatter distribution is similar to the normal distribution, and the highest point is marked as (M)max，Nmax). At this time NmaxIs the value closest to the number of true hard disk failures N in the sample, then MmaxThe corresponding weight value is regarded as the optimal weight value. And finishing the training of the parameter weighted value model of the hard disk.
After the optimal weighted value is trained for each performance state parameter, a relation model between the performance state parameter and the service life duration of the hard disk can be established, specifically, a function model of the performance state parameter weighted sum and the service life duration of the hard disk is established and trained, and finally, a hard disk service life prediction function model is obtained.
In some specific embodiments, a function model of the performance state parameter weighted sum and the hard disk life duration is established based on the optimal weighted value for training to obtain a hard disk life prediction function model, which specifically includes the following steps.
(1) Taking Q ten thousand fault hard disks, and obtaining the stored time length of each fault hard disk and recording the time length as Li。
It should be noted that the saved active time length can be obtained from smart log of the failed hard disk.
(2) According to the formula
Figure BDA0003189181170000091
Calculating the M value of each fault hard disk and recording the M value as MiObtaining a point (M)i，Li) Fall into the M/L plane coordinates.
FIG. 4 shows the M/L plane coordinate midpoint (M)i，Li) Schematic representation.
(3) According to Q ten thousand points (M)i，Li) Simulating a linear function
Figure BDA0003189181170000101
Wherein X, Y, Z are three constants; the linearityThe function is a hard disk life prediction function model.
Q ten thousand fault hard disks can obtain Q ten thousand scattering points on an M/L plane coordinate, and M can be obtained according to the distribution of the scattering pointsiAnd LiThe two values are inversely related, i.e., the larger the value of M, the smaller the value of L. Shown as a point (M) in FIG. 5i，Li) A scatter plot of.
According to the distribution relation of scattering points, a linear function can be simulated
Figure BDA0003189181170000102
And finishing the training of the hard disk life prediction function model.
According to hard disk life prediction function model
Figure BDA0003189181170000103
The remaining life duration of the hard disk can be calculated as follows: obtaining smart logs of a healthy hard disk, and obtaining the performance state parameters of the health of the hard disk from the smart logs, wherein the performance state parameters are S1、S2......Sk(ii) a The optimal weight value is A1i、A2i......Aki. Then use the formula
Figure BDA0003189181170000104
Calculating the current health coefficient M value of the hard disk and recording the current health coefficient M value as Mj(ii) a Using formulas
Figure BDA0003189181170000105
The service life of the hard disk at this time can be calculated as follows: l isj. The survival time of the hard disk can be obtained from the smart log of the hard disk as follows: l isoThen Ls＝Lj-LoTo LsNamely the remaining life duration of the hard disk. The lifetime may be in hours.
Several performance state parameters of the hard disk have an impact on hard disk life, and in some embodiments, the performance state parameters taken include remapped sector count, spindle spin-up retry number, head calibration retry count, erase-fail block count, point-to-point error detection count, uncorrectable error count, hardware ECC correction count, current to-be-mapped sector count, offline uncorrectable sector count, and ULTRA DMA parity error rate.
Example two
The second embodiment provides a device for predicting the remaining life of an SSD hard disk, which is used to implement the method for predicting the remaining life of an SSD hard disk.
Fig. 6 is a schematic block diagram of a structure of a device for predicting the remaining life of an SSD hard disk according to the second embodiment, which includes the following functional modules.
The performance state parameter acquisition module 101: acquiring a plurality of performance state parameters of each training hard disk based on a hard disk unified standard;
optimal weight training module 102: establishing a parameter weighted value model for training to obtain the optimal weighted value of each performance state parameter;
hard disk life prediction function module training module 103: establishing a function model of performance state parameter weighting and hard disk service life duration based on the optimal weighted value for training to obtain a hard disk service life prediction function model;
hard disk remaining life prediction module 104: and predicting the residual life of the hard disk to be predicted based on the hard disk life prediction function model.
The optimal weighted value training module 102 establishes a parameter weighted value model for training to obtain the optimal weighted value of each performance status parameter, and specifically includes the following processes:
(1) the performance state parameters of the training hard disk are respectively S1、S2……Sk；
(2) The weighted value of each performance state parameter is A1、A2……Ak(ii) a Wherein
Figure BDA0003189181170000111
And 0 < An< H; h is a positive number;
(3) the performance state parameter is weighted and summed
Figure BDA0003189181170000112
(4) Taking P ten thousand training hard disks, wherein N fault hard disks are included;
(5) every random change of a set of weight values a1、A2……AkCalculating P ten thousand M values once, and taking out the mode M from the P ten thousand M valuesxWherein the mode M is satisfiedxThe number of failed hard disks of (2) is NyThen (M)x，Ny) Can be recorded as a point in the M/N plane coordinate;
(6) weighted value A1、A2......AkVarying groups to obtain points (M)x，Ny) A scatter point distribution map;
(7) the highest point in the scatter plot is denoted as (M)max，Nmax) Then M ismaxThe corresponding weighted value is the optimal weighted value, and is marked as A1i、A2i……Aki。
The hard disk life prediction function module training module 103 establishes a function model of performance state parameter weighting and hard disk life duration based on the optimal weighting value to train, and obtains a hard disk life prediction function model, which specifically comprises the following processes:
(1) taking Q ten thousand fault hard disks, and obtaining the stored time length of each fault hard disk and recording the time length as Li；
(2) According to the formula
Figure BDA0003189181170000121
Calculating the M value of each fault hard disk and recording the M value as MiObtaining a point (M)i，Li) Fall into the M/L plane coordinates;
(3) according to Q ten thousand points (M)i，Li) Simulating a linear function
Figure BDA0003189181170000122
Wherein X, Y, Z are three constants; the linear function is a hard disk life prediction function model.
According to hard disk life prediction function model
Figure BDA0003189181170000123
The remaining life duration of the hard disk can be calculated as follows: obtaining smart logs of a healthy hard disk, and obtaining the performance state parameters of the health of the hard disk from the smart logs, wherein the performance state parameters are S1、S2......Sk(ii) a The optimal weight value is A1i、A2i......Aki. Then use the formula
Figure BDA0003189181170000124
Calculating the current health coefficient M value of the hard disk and recording the current health coefficient M value as Mi(ii) a Using formulas
Figure BDA0003189181170000125
The service life of the hard disk at this time can be calculated as follows: l isj. The survival time of the hard disk can be obtained from the smart log of the hard disk as follows: l isoThen Ls＝Lj-LoTo LsNamely the remaining life duration of the hard disk. The lifetime may be in hours.
Several performance state parameters of the hard disk have an impact on hard disk life, and in some embodiments, the performance state parameters taken include remapped sector count, spindle spin-up retry number, head calibration retry count, erase-fail block count, point-to-point error detection count, uncorrectable error count, hardware ECC correction count, current to-be-mapped sector count, offline uncorrectable sector count, and ULTRA DMA parity error rate.
The device for predicting the remaining life of the SSD hard disk of the present embodiment is used for implementing the foregoing method for predicting the remaining life of the SSD hard disk, and therefore, the specific implementation manner of the device can be seen in the foregoing embodiment section of the method for predicting the remaining life of the SSD hard disk, and therefore, the specific implementation manner thereof can refer to the description of the corresponding respective embodiment sections, and is not further described herein.
In addition, since the device for predicting the remaining life of the SSD hard disk of the present embodiment is used for implementing the method for predicting the remaining life of the SSD hard disk, the function corresponds to the function of the method, and the detailed description is omitted here.
EXAMPLE III
Fig. 7 is a schematic structural diagram of a terminal device 700 according to an embodiment of the present invention, where the terminal device 700 may be used to execute the method for predicting the remaining life of an SSD hard disk according to the embodiment of the present invention.
The terminal device 700 may include: processor 710, memory 720, and communication unit 730. The components communicate via one or more buses, and those skilled in the art will appreciate that the architecture of the servers shown in the figures is not intended to be limiting, and may be a bus architecture, a star architecture, a combination of more or less components than those shown, or a different arrangement of components.
The memory 720 may be used for storing instructions executed by the processor 710, and the memory 720 may be implemented by any type of volatile or non-volatile storage terminal or combination thereof, such as Static Random Access Memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic memory, flash memory, magnetic disk or optical disk. The executable instructions in memory 720, when executed by processor 710, enable terminal 700 to perform some or all of the steps in the method embodiments described below.
The processor 710 is a control center of the storage terminal, connects various parts of the entire electronic terminal using various interfaces and lines, and performs various functions of the electronic terminal and/or processes data by operating or executing software programs and/or modules stored in the memory 720 and calling data stored in the memory. The processor may be composed of an Integrated Circuit (IC), for example, a single packaged IC, or a plurality of packaged ICs connected with the same or different functions. For example, the processor 710 may include only a Central Processing Unit (CPU). In the embodiment of the present invention, the CPU may be a single operation core, or may include multiple operation cores.
A communication unit 730, configured to establish a communication channel so that the storage terminal can communicate with other terminals. And receiving user data sent by other terminals or sending the user data to other terminals.
Example four
The present invention also provides a computer storage medium, wherein the computer storage medium may store a program, and the program may include some or all of the steps in the embodiments provided by the present invention when executed. The storage medium may be a magnetic disk, an optical disk, a read-only memory (ROM) or a Random Access Memory (RAM).
Those skilled in the art will readily appreciate that the techniques of the embodiments of the present invention may be implemented as software plus a required general purpose hardware platform. Based on such understanding, the technical solutions in the embodiments of the present invention may be embodied in the form of a software product, where the computer software product is stored in a storage medium, such as a usb disk, a removable hard disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), a magnetic disk or an optical disk, and the like, and the storage medium can store program codes, and includes instructions for enabling a computer terminal (which may be a personal computer, a server, or a second terminal, a network terminal, and the like) to perform all or part of the steps of the method in the embodiments of the present invention.
The same and similar parts in the various embodiments in this specification may be referred to each other. Especially, for the terminal embodiment, since it is basically similar to the method embodiment, the description is relatively simple, and the relevant points can be referred to the description in the method embodiment.
In the embodiments provided in the present invention, it should be understood that the disclosed system, apparatus and method may be implemented in other ways. For example, the above-described apparatus embodiments are merely illustrative, and for example, the division of the units is only one logical division, and other divisions may be realized in practice, for example, a plurality of units or components may be combined or integrated into another system, or some features may be omitted, or not executed. In addition, the shown or discussed mutual coupling or direct coupling or communication connection may be an indirect coupling or communication connection through some interfaces, devices or units, and may be in an electrical, mechanical or other form.
The units described as separate parts may or may not be physically separate, and parts displayed as units may or may not be physical units, may be located in one place, or may be distributed on a plurality of network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of the embodiment.
In addition, functional units in the embodiments of the present invention may be integrated into one processing unit, or each unit may exist alone physically, or two or more units are integrated into one unit.
The above disclosure is only for the preferred embodiments of the present invention, but the present invention is not limited thereto, and any non-inventive changes that can be made by those skilled in the art and several modifications and amendments made without departing from the principle of the present invention shall fall within the protection scope of the present invention.

Claims (10)

1. A method for predicting the residual life of an SSD hard disk is characterized by comprising the following steps:
acquiring a plurality of performance state parameters of each training hard disk based on a hard disk unified standard;
establishing a parameter weighted value model for training to obtain the optimal weighted value of each performance state parameter;
establishing a function model of performance state parameter weighting and hard disk service life duration based on the optimal weighted value for training to obtain a hard disk service life prediction function model;
and predicting the residual life of the hard disk to be predicted based on the hard disk life prediction function model.
2. The method for predicting the remaining life of the SSD hard disk according to claim 1, wherein a parameter weighted value model is established for training to obtain an optimal weighted value of each performance state parameter, which specifically comprises: the performance state parameters of the training hard disk are respectively S1、S2......Sk；
The weighted value of each performance state parameter is A1、A2......Ak(ii) a Wherein
Figure FDA0003189181160000011
And 0 < An< H; h is a positive number;
the performance state parameter is weighted and summed
Figure FDA0003189181160000012
Taking P ten thousand training hard disks, wherein N fault hard disks are included;
every random change of a set of weight values a1、A2......AkCalculating P ten thousand M values once, and taking out the mode M from the P ten thousand M valuesxWherein the mode M is satisfiedxThe number of failed hard disks of (2) is NyThen (M)x，Ny) Can be recorded as a point in the M/N plane coordinate;
weighted value A1、A2......AkVarying groups to obtain points (M)x，Ny) A scatter point distribution map;
the highest point in the scatter plot is denoted as (M)max，Nmax) Then M ismaxThe corresponding weighted value is the optimal weighted value, and is marked as A1i、A2i......Aki。
3. The method for predicting the remaining life of the SSD hard disk according to claim 2, wherein a function model of the performance state parameter weighting and the hard disk life duration is established based on the optimal weighting value for training to obtain a hard disk life prediction function model, specifically comprising:
taking Q ten thousand fault hard disks, and obtaining the stored time length of each fault hard disk and recording the time length as Li；
According to the formula
Figure FDA0003189181160000021
Calculating the M value of each fault hard disk and recording the M value as MiObtaining a point (M)i，Li) Fall into the M/L plane coordinates;
according to Q ten thousand points (M)i，Li) Simulating a linear function
Figure FDA0003189181160000022
Wherein X, Y, Z are three constants; the linear function is a hard disk life prediction function model.
4. The method of predicting SSD hard disk remaining life according to any one of claims 1-3, wherein the performance state parameters comprise a remapped sector count, a spindle spin-up retry count, a head calibration retry count, an erase-fail block count, a point-to-point error detection count, an uncorrectable error count, a hardware ECC correction count, a current to-be-mapped sector count, an offline uncorrectable sector count, and an ULTRADMA parity error rate.
5. An apparatus for predicting the remaining life of an SSD hard disk, comprising,
a performance state parameter acquisition module: acquiring a plurality of performance state parameters of each training hard disk based on a hard disk unified standard;
an optimal weighted value training module: establishing a parameter weighted value model for training to obtain the optimal weighted value of each performance state parameter;
the hard disk life prediction function module training module: establishing a function model of performance state parameter weighting and hard disk service life duration based on the optimal weighted value for training to obtain a hard disk service life prediction function model;
hard disk remaining life prediction module: and predicting the residual life of the hard disk to be predicted based on the hard disk life prediction function model.
6. The device for predicting the remaining life of the SSD hard disk of claim 5, wherein the optimal weighted value training module establishes a parameter weighted value model for training to obtain the optimal weighted value of each performance status parameter, specifically comprising:
the performance state parameters of the training hard disk are respectively S1、S2......Sk；
The weighted value of each performance state parameter is A1、A2......Ak(ii) a Wherein
Figure FDA0003189181160000031
And 0 < An< H; h is a positive number;
the performance state parameter is weighted and summed
Figure FDA0003189181160000032
Taking P ten thousand training hard disks, wherein N fault hard disks are included;
every random change of a set of weight values a1、A2......AkCalculating P ten thousand M values once, and taking out the mode M from the P ten thousand M valuesxWherein the mode M is satisfiedxThe number of failed hard disks of (2) is NyThen (M)x，Ny) Can be recorded as a point in the M/N plane coordinate;
weighted value A1、A2......AkVarying groups to obtain points (M)x，Ny) A scatter point distribution map;
the highest point in the scatter plot is denoted as (M)max，Nmax) Then M ismaxThe corresponding weighted value is the optimal weighted value, and is marked as A1i、A2i......Aki。
7. The device for predicting the remaining life of the SSD hard disk of claim 6, wherein the hard disk life prediction function module training module establishes a function model of the performance state parameter weighting and the hard disk life duration based on the optimal weighting value for training to obtain the hard disk life prediction function model, specifically comprising:
taking Q ten thousand fault hard disks, and obtaining the stored time length of each fault hard disk and recording the time length as Li；
According to the formula
Figure FDA0003189181160000033
Calculating the M value of each fault hard disk and recording the M value as MiObtaining a point (M)i，Li) Fall into the M/L plane coordinates;
according to Q ten thousand points (M)i，Li) Simulating a linear function
Figure FDA0003189181160000034
Wherein X, Y, Z are three constants; the linear function is a hard disk life prediction function model.
8. The apparatus of any of claims 5-7, wherein the performance state parameters include remapped sector count, spindle spin-up retry count, head calibration retry count, erase-fail block count, point-to-point error detection count, uncorrectable error count, hardware ECC correction count, current to-be-mapped sector count, offline uncorrectable sector count, and ULTRADMA parity error rate.
9. A terminal, comprising:
a processor;
a memory for storing instructions for execution by the processor;
wherein the processor is configured to perform the method of any one of claims 1-4.
10. A computer-readable storage medium, in which a computer program is stored which, when being executed by a processor, carries out the method according to any one of claims 1-4.
CN202110872186.XA 2021-07-30 2021-07-30 Method, device, terminal and storage medium for predicting residual life of SSD hard disk Active CN113688564B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202110872186.XA CN113688564B (en) 2021-07-30 2021-07-30 Method, device, terminal and storage medium for predicting residual life of SSD hard disk

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202110872186.XA CN113688564B (en) 2021-07-30 2021-07-30 Method, device, terminal and storage medium for predicting residual life of SSD hard disk

Publications (2)

Publication Number Publication Date
CN113688564A true CN113688564A (en) 2021-11-23
CN113688564B CN113688564B (en) 2024-02-27

Family

ID=78578434

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202110872186.XA Active CN113688564B (en) 2021-07-30 2021-07-30 Method, device, terminal and storage medium for predicting residual life of SSD hard disk

Country Status (1)

Country Link
CN (1) CN113688564B (en)

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN114882937A (en) * 2022-04-30 2022-08-09 苏州浪潮智能科技有限公司 Solid state disk durability testing method, sample amount calculating method and device
CN116503164A (en) * 2023-06-25 2023-07-28 杭银消费金融股份有限公司 Business processing method and equipment based on financial terminal
CN118093324A (en) * 2024-04-28 2024-05-28 深圳市领德创科技有限公司 Solid state disk life prediction method and system

Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20110238907A1 (en) * 2007-10-10 2011-09-29 Juniper Networks, Inc. Wake-ahead based on patterns
US20180080853A1 (en) * 2016-09-16 2018-03-22 Siemens Aktiengesellschaft Additive life consumption model for predicting remaining time-to-failure of machines
CN110413430A (en) * 2019-07-19 2019-11-05 苏州浪潮智能科技有限公司 A kind of life-span prediction method of solid state hard disk, device and equipment
WO2020114313A1 (en) * 2018-12-05 2020-06-11 中兴通讯股份有限公司 Method and apparatus for predicting hard disk fault occurrence time, and storage medium
CN111736768A (en) * 2020-05-29 2020-10-02 苏州浪潮智能科技有限公司 Prediction method, system and storage medium for remaining usage time of server hard disk
US20210225405A1 (en) * 2018-01-31 2021-07-22 Hewlett-Packard Development Company, L.P. Hard disk drive lifetime forecasting

Patent Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20110238907A1 (en) * 2007-10-10 2011-09-29 Juniper Networks, Inc. Wake-ahead based on patterns
US20180080853A1 (en) * 2016-09-16 2018-03-22 Siemens Aktiengesellschaft Additive life consumption model for predicting remaining time-to-failure of machines
US20210225405A1 (en) * 2018-01-31 2021-07-22 Hewlett-Packard Development Company, L.P. Hard disk drive lifetime forecasting
WO2020114313A1 (en) * 2018-12-05 2020-06-11 中兴通讯股份有限公司 Method and apparatus for predicting hard disk fault occurrence time, and storage medium
CN110413430A (en) * 2019-07-19 2019-11-05 苏州浪潮智能科技有限公司 A kind of life-span prediction method of solid state hard disk, device and equipment
CN111736768A (en) * 2020-05-29 2020-10-02 苏州浪潮智能科技有限公司 Prediction method, system and storage medium for remaining usage time of server hard disk

Non-Patent Citations (2)

* Cited by examiner, † Cited by third party
Title
梁泽明: "多参数相似性信息融合的剩余寿命预测", 计算机集成制造系统, vol. 24, no. 4, pages 813 - 819 *
潘新祥: "船舶动力设备最佳维修周期的确定", 大连海事大学学报, vol. 23, no. 3, pages 63 - 66 *

Cited By (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN114882937A (en) * 2022-04-30 2022-08-09 苏州浪潮智能科技有限公司 Solid state disk durability testing method, sample amount calculating method and device
CN114882937B (en) * 2022-04-30 2024-10-15 苏州浪潮智能科技有限公司 A solid state hard disk durability testing method, sample size calculation method and device
CN116503164A (en) * 2023-06-25 2023-07-28 杭银消费金融股份有限公司 Business processing method and equipment based on financial terminal
CN116503164B (en) * 2023-06-25 2023-09-19 杭银消费金融股份有限公司 Business processing method and equipment based on financial terminal
CN118093324A (en) * 2024-04-28 2024-05-28 深圳市领德创科技有限公司 Solid state disk life prediction method and system

Also Published As

Publication number Publication date
CN113688564B (en) 2024-02-27

Similar Documents

Publication Publication Date Title
CN113688564A (en) Method, device, terminal and storage medium for predicting remaining life of SSD (solid State disk)
CN109522175B (en) A method and device for memory evaluation
CN107025153B (en) Disk failure prediction method and device
US10223224B1 (en) Method and system for automatic disk failure isolation, diagnosis, and remediation
CN112771504A (en) Multi-factor cloud service storage device error prediction
US11165668B2 (en) Quality assessment and decision recommendation for continuous deployment of cloud infrastructure components
CN115348159A (en) Microservice Fault Location Method and Device Based on Autoencoder and Service Dependency Graph
CN111858108B (en) Hard disk fault prediction method and device, electronic equipment and storage medium
CN111104293A (en) Method, apparatus and computer program product for supporting disk failure prediction
WO2022001125A1 (en) Method, system and device for predicting storage failure in storage system
CN112882796A (en) Abnormal root cause analysis method and apparatus, and storage medium
CN108173672B (en) Method and apparatus for detecting faults
CN114327241A (en) Method, electronic device and computer program product for managing disk
CN108897665A (en) Blog management method, device, computer equipment and storage medium
US11436069B2 (en) Method and apparatus for predicting hard drive failure
US8799608B1 (en) Techniques involving flaky path detection
CN113946983A (en) Method and device for evaluating weak links of product reliability and computer equipment
CN108763027A (en) A kind of hard disk monitoring method, device and server
US20170132056A1 (en) Durability and availability evaluation for distributed storage systems
CN118330439B (en) Semiconductor chip test circuit and test method
CN111367732B (en) Memory application grade prediction method, system, terminal and storage medium
CN116057902B (en) Health index of service
CN115454704A (en) A server fault diagnosis test method, device, terminal and storage medium
CN115114070A (en) Fault diagnosis method, device, equipment and medium
CN109739210B (en) Method and device for assessing health status of equipment components

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant