CN109582527A - Storage server and its solid-state disk service life monitoring method - Google Patents
Storage server and its solid-state disk service life monitoring method Download PDFInfo
- Publication number
- CN109582527A CN109582527A CN201710911732.XA CN201710911732A CN109582527A CN 109582527 A CN109582527 A CN 109582527A CN 201710911732 A CN201710911732 A CN 201710911732A CN 109582527 A CN109582527 A CN 109582527A
- Authority
- CN
- China
- Prior art keywords
- hard disk
- solid state
- state hard
- solid
- remaining
- 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.)
- Pending
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F11/00—Error detection; Error correction; Monitoring
- G06F11/30—Monitoring
- G06F11/3003—Monitoring arrangements specially adapted to the computing system or computing system component being monitored
- G06F11/3037—Monitoring arrangements specially adapted to the computing system or computing system component being monitored where the computing system component is a memory, e.g. virtual memory, cache
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F11/00—Error detection; Error correction; Monitoring
- G06F11/30—Monitoring
- G06F11/3003—Monitoring arrangements specially adapted to the computing system or computing system component being monitored
- G06F11/3034—Monitoring arrangements specially adapted to the computing system or computing system component being monitored where the computing system component is a storage system, e.g. DASD based or network based
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F11/00—Error detection; Error correction; Monitoring
- G06F11/30—Monitoring
- G06F11/3055—Monitoring arrangements for monitoring the status of the computing system or of the computing system component, e.g. monitoring if the computing system is on, off, available, not available
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F3/00—Input arrangements for transferring data to be processed into a form capable of being handled by the computer; Output arrangements for transferring data from processing unit to output unit, e.g. interface arrangements
- G06F3/06—Digital input from, or digital output to, record carriers, e.g. RAID, emulated record carriers or networked record carriers
- G06F3/0601—Interfaces specially adapted for storage systems
- G06F3/0628—Interfaces specially adapted for storage systems making use of a particular technique
- G06F3/0653—Monitoring storage devices or systems
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F3/00—Input arrangements for transferring data to be processed into a form capable of being handled by the computer; Output arrangements for transferring data from processing unit to output unit, e.g. interface arrangements
- G06F3/06—Digital input from, or digital output to, record carriers, e.g. RAID, emulated record carriers or networked record carriers
- G06F3/0601—Interfaces specially adapted for storage systems
- G06F3/0668—Interfaces specially adapted for storage systems adopting a particular infrastructure
- G06F3/0671—In-line storage system
- G06F3/0673—Single storage device
- G06F3/0679—Non-volatile semiconductor memory device, e.g. flash memory, one time programmable memory [OTP]
Abstract
A kind of storage server and its solid-state disk service life monitoring method.Solid-state disk service life monitoring method is suitable for inclusion in the storage server of multiple solid state hard disks, comprises the steps of the remaining life reading difference for capturing each solid state hard disk between starting time point and end time point；In starting time point and terminate the data writing for adding up each solid state hard disk between time point；The ratio between difference and data writing, which is read, according to remaining life determines that the solid state hard disk of each solid state hard disk restrains oneself force parameter；And the remaining life reading that each solid state hard disk is put at the end is captured, to restrain oneself the remaining writing that the ratio between force parameter determines that each solid state hard disk puts at the end according to remaining life reading and solid state hard disk.The effect of this disclosure of documents can be by can be used single solid-state disk service life monitoring method, reaching the service life of the solid state hard disk of monitoring variety classes and different size.
Description
Technical field
This disclosure of documents is related to a kind of seagate more particularly to a kind of storage server and its solid-state disk service life monitoring
Method.
Background technique
Solid state hard disk is a kind of Computer Memory Unit using memory body as its memory device, wherein memory body be using
Based on NAND gate flash memory (NAND Flash) technology.NAND gate flash memory utilizes floating gate (Floating Gate)
Electronic data is stored, that is, when electronics is into and out floating gate can be respectively seen as " 0 " and " 1 " two different storage states.
And electronics is stored to the voltage that control gate (Word Line Control Gate) need to be adjusted into floating gate, allow part electronics to measure
Sub- tunneling effect leap floating gate.However, by electronics pour into during floating gate can gradually to around floating gate insulating layer (such as
Compound crystal silicon interlayer dielectric layer (inter-poly dielectric) or tunnel oxide (tunnel oxide)) it causes physically
Loss.After the loss of insulating layer reaches to a certain degree, the memory device will be unable to normal operation.That is, for
For solid state hard disk, with the accumulation of number of operations, it will have more and more memory blocks become can not be erasable state.Cause
This, the service life of solid state hard disk can substantially present with the erasable number of NAND gate flash memory and be positively correlated.
However, other than the erasable number of reference material itself, there are also permitted when calculating the service life of a solid state hard disk
Mostly additional parameter (such as processing procedure factor, user's use habit etc.) will be also included in and consider.Various manufacturers can needle at present
The hard disk produced to it imports self-monitoring, analysis and reporting techniques (Self-Monitoring Analysis and
Reporting Technology；S.M.A.R.T.) information.Through S.M.A.R.T. information, user can learn that solid-state is hard
The health status of disk, then by health status go assessment solid state hard disk service life.
Such mode is directed to same label and perhaps can yet be regarded as one when the solid state hard disk of same model assesses remaining life
More objective assessment mode can do the sequence of remaining life for multiple and different hard disks of same label and same model,
So that user replaces the hard disk that will be damaged ahead of time, in order to avoid Missing data.However, if in storage system there are it is multiple not
With the solid state hard disk of label, then the obtained health status of S.M.A.R.T. information of each label can't be one objective and can
The information being compared to each other.As previously mentioned, assessment solid-state disk service life need to refer to many additional factors, different labels
S.M.A.R.T. algorithm used in technology may be different, and referenced user's use condition may be different, therefore even if
By same erasable number, it is also possible to obtain different health status.
Especially limitation user is selected the bullet of solid state hard disk in the storage system of an all solid state hard disk by above situation
Property.Still further, if open user selects different labels, the solid state hard disk of different model forms all solid state storage system
System, then the health status information of the obtained each solid state hard disk of S.M.A.R.T. information of different labels or different model is not
It is a standard that is objective and can mutually comparing, therefore the remaining life for all solid state hard disks that can not be used directly to sort.?
That is when user uses the solid state hard disk of different model with, it may be difficult to learn the remaining life of each solid state hard disk, so
May cause the longer solid state hard disk of remaining life preferentially to be eliminated and changed, increase enterprise safeguard all solid state hard disk storage system at
This.
Summary of the invention
This disclosure of documents is designed to provide single solid-state disk service life monitoring method, reach monitoring variety classes and
The effect of service life of the solid state hard disk of different size.
Therefore, a purpose of this disclosure of documents is to propose a kind of solid-state disk service life monitoring method, is suitable for inclusion in more
The storage server of a solid state hard disk, which is characterized in that monitoring method includes: each solid state hard disk is captured in starting time point and knot
Remaining life between Shu Shidian reads difference；It is write in starting time point and the data for terminating to add up between time point each solid state hard disk
Enter amount；The ratio between difference and data writing, which is read, according to remaining life determines that the solid state hard disk of each solid state hard disk is restrained oneself
Force parameter；And the remaining life reading that each solid state hard disk is put at the end is captured, according to remaining life reading and solid-state
Hard disk restrains oneself the ratio between force parameter and determines the remaining writing that each solid state hard disk is put at the end.
In an embodiment, the base of the remaining life reading of at least two solid state hard disks is different.
In an embodiment, the remaining writing of each solid state hard disk is using data bit as base.
In an embodiment, solid-state disk service life monitoring method also includes: it sorts to the remaining writing of each solid state hard disk,
Generate the remaining life ranking results of solid state hard disk.
In an embodiment, solid state hard disk forms disk array, and solid-state disk service life monitoring method also includes: according to residue
Arranging in lifetime is as a result, have the parity check code of least residue service life person's storage disk array in specified solid state hard disk.
In an embodiment, solid-state disk service life monitoring method also includes: unit of account time average is according to writing；With
And determine that the remaining of each solid state hard disk uses according to the ratio between remaining writing and unit time average data writing
Time.
In an embodiment, solid-state disk service life monitoring method also includes: is arranged using the time the residue of each solid state hard disk
Sequence generates the remaining life ranking results of solid state hard disk.
The another object of this disclosure of documents is to propose a kind of storage server, which is characterized in that storage server includes:
Multiple solid state hard disks, memory body and processing unit.Memory body is configured to store procedure code.Processing unit is electrically coupled to solid-state
Hard disk and memory body, and configure with Self-memory body access program code, to execute solid-state disk service life monitoring method, solid state hard disk
Service life supervision method includes: capturing remaining life of each solid state hard disk between starting time point and end time point and reads difference；
In starting time point and terminate the data writing for adding up each solid state hard disk between time point；According to remaining life read difference and
Ratio between data writing determines that the solid state hard disk of each solid state hard disk restrains oneself force parameter；And it captures each solid state hard disk and is tying
The remaining life of Shu Shidian is read, and is determined respectively with restraining oneself the ratio between force parameter according to remaining life reading and solid state hard disk
The remaining writing that solid state hard disk is put at the end.
In an embodiment, the base of the remaining life reading of at least two solid state hard disks is different.
In an embodiment, solid-state disk service life monitoring method also includes: unit of account time average is according to writing；With
And determine that the remaining of each solid state hard disk uses according to the ratio between remaining writing and unit time average data writing
Time.
Detailed description of the invention
Fig. 1 is a kind of block diagram of storage server in one embodiment of this disclosure of documents；
Fig. 2 is a kind of flow chart of solid-state disk service life monitoring method in one embodiment of this disclosure of documents；
Fig. 3 is to be shown according to solid state hard disk remaining life in an embodiment of this disclosure of documents with the relationship that data volume is written
It is intended to；And
Fig. 4 is to be shown according to solid state hard disk remaining life in an embodiment of this disclosure of documents with the relationship that data volume is written
It is intended to.
Specific embodiment
It is hereafter to cooperate appended attached drawing to elaborate for embodiment, but provided embodiment is not to limit this case
The range covered, and the non-sequence to limit its execution of the description of structure operation, any structure reconfigured by element,
It is produced that there is equal and other effects device, it is all the range that this case is covered.In addition, " first " used herein, " the
Two " ... etc., the meaning of order or cis-position is not especially censured, it is also non-to limit this case, just for the sake of difference with identical
The element of technical terms description or operation.
Please refer to Fig. 1 and Fig. 2.Fig. 1 is a kind of block diagram of storage server 1 in one embodiment of this disclosure of documents.Fig. 2
For in one embodiment of this disclosure of documents, a kind of flow chart of solid-state disk service life monitoring method 200.
Storage server 1 includes: multiple solid state hard disk 100A, 100B, memory body 102 and processing unit 104.
In an embodiment, solid state hard disk 100A, 100B is respectively such as, but not limited to, flash memory, and can be formed
Full flash memory storage array (All Flash Array；AFA storage device).It is noted that in different embodiments
In, the number of solid state hard disk 100A, 100B also may be greater than two any number, not limited by what the present embodiment was painted.
Memory body 102 is configured to store procedure code 101.In different embodiments, memory body 102 can be remembered for semiconductor-type
Body equipment, random access memory, read-only memory, programmable read-only memory, cache memory, Magnetic memory body equipment with
System, optical memory body equipment and system, fixed memory body or removable memory body.
Processing unit 104 is electrically coupled to solid state hard disk 100A, 100B and memory body 102.In an embodiment, processing
Unit 104 can pass through various suitable buses and solid state hard disk 100A, 100B and 102 electric property coupling of memory body and be led to
News.The configuration of processing unit 104 is with the access of Self-memory body 102 such as, but not limited to, procedure code 101, to carry out calculation process.
Have the advantages that access speed is fast entirely with the storage device that solid state hard disk is formed.Common all solid state storage device is
It is formed by the solid state hard disk of specific model.In such a case, the service life of each solid state hard disk can be directly by corresponding
S.M.A.R.T. information judges.
However, when solid state hard disk 100A, 100B because such as, but not limited to, from different manufacturers or different model
When with different specifications, the parameter of S.M.A.R.T. information will likely because of unit difference, and be difficult to estimate solid state hard disk 100A,
The 100B real service life, can not optimization data storage efficiency.
Therefore, in an embodiment, the configuration of processing unit 104 is solid to execute with 102 access program code 101 of Self-memory body
State hard disk life-span monitoring method 200, effectively monitors solid state hard disk 100A, 100B of different size.
To be arranged in pairs or groups Fig. 2 by Fig. 1 below, and solid-state disk service life monitoring method 200 is described in detail.Solid-state disk service life monitoring
Method 200 can be applied to the storage server 100 that Fig. 1 is painted.Solid-state disk service life monitoring method, which comprises the steps of, (should be appreciated that
It arrives, in the present embodiment mentioned step, in addition to especially chatting its bright sequence person, it is suitable can to adjust its front and back according to actual needs
Sequence, or even can simultaneously or partially be performed simultaneously).
In step 201, processing unit 104 capture each solid state hard disk 100A, 100B starting time point and terminate time point it
Between remaining life read difference DELTA LifeA, Δ LifeB.
In an embodiment, starting time point is T1, and terminating time point is T2.Processing unit 104 can be hard by capturing solid-state
Disk 100A reads LifeAT2 in the remaining life that the remaining life of starting time point reads LifeAT1 and puts at the end, with phase
Subtract to obtain remaining life reading difference DELTA LifeA, and can be expressed from the next:
Δ LifeA=LifeAT2-LifeAT1 (formula 1)
Similarly, processing unit 104 can be read by capturing solid state hard disk 100B in the remaining life of starting time point
LifeBT1 and the remaining life put at the end read LifeBT2, to subtract each other to obtain remaining life reading difference DELTA LifeB,
And it can be expressed from the next:
Δ LifeB=LifeBT2-LifeBT1 (formula 2)
In an embodiment, above-mentioned remaining life reads LifeAT1, LifeAT2, LifeBT1 and LifeBT2, can be by
Processing unit 104 is read such as, but not limited to, the S.M.A.R.T. information (not being painted) of each solid state hard disk 100A, 100B obtains.
In step 202, processing unit 104 starting time point and terminating add up between time point each solid state hard disk 100A,
Data the writing Δ WDA and Δ WDB of 100B.
In different embodiments, data writing Δ WDA and Δ WDB can be read by processing unit 104 such as, but not limited to,
The S.M.A.R.T. information of each solid state hard disk 100A, 100B obtains, or relevant program record executed by processing unit 104 and
It obtains.
In step 203, processing unit 104 reads difference DELTA LifeA, Δ LifeB and data according to remaining life and is written
The ratio between Δ WDA, Δ WDB is measured, determines that the solid state hard disk of each solid state hard disk 100A, 100B restrains oneself force parameter EDA and EDB.
In more detail, the solid state hard disk of solid state hard disk 100A, which restrains oneself force parameter EDA, to read difference by remaining life
Ratio between Δ LifeA and data writing Δ WDA obtains, and can following formula indicate:
EDA=Δ LifeA/ Δ WDA (formula 3)
On the other hand, the solid state hard disk of solid state hard disk 100B, which restrains oneself force parameter EDB, to read difference DELTA by remaining life
Ratio between LifeB and data writing Δ WDB obtains, and can following formula indicate:
EDB=Δ LifeB/ Δ WDB (formula 4)
In step 204, processing unit 104 captures the remaining life reading that each solid state hard disk 100A, 100B is order at the end
LifeAT2, LifeBT2, with according to remaining life read LifeAT2, LifeBT2 and solid state hard disk restrain oneself force parameter EDA,
Ratio between EDB determines remaining writing RWDA, RWDB that each solid state hard disk 100A, 100B is order at the end.
In more detail, the remaining writing RWDA of solid state hard disk 100A can by remaining life read LifeAT2 and
The ratio that solid state hard disk is restrained oneself between force parameter EDA obtains, and can following formula indicate:
RWDA=LifeAT2/EDA (formula 5)
On the other hand, the remaining writing RWDB of solid state hard disk 100B can be read LifeBT2 and be consolidated by remaining life
The ratio that state hard disk is restrained oneself between force parameter EDB obtains, and can following formula indicate:
RWDB=LifeBT2/EDB (formula 6)
By above-mentioned (formula 3) and (formula 5) it is found that residue writing RWDA can further be expressed from the next:
RWDA=LifeAT2/ (Δ LifeA/ Δ WDA) (formula 7)
Remaining writing RWDB then can further be expressed from the next:
RWDB=LifeBT2/ (Δ LifeB/ Δ WDB) (formula 8)
In an embodiment, the remaining life of solid state hard disk 100A, 100B reading LifeAT1, LifeAT2, LifeBT1 and
The base of LifeBT2 is different, the benchmark list of the remaining life reading difference DELTA LifeA, Δ LifeB that derive accordingly
Position is also different.And on the other hand, data writing Δ WDA, the Δ WDB of solid state hard disk 100A, 100B then can with for example, but
Data bit is not limited to as base.
By (formula 7) and (formula 8) it is found that above-mentioned base relevant to remaining life reading, is read by remaining life
LifeAT2 and remaining life reading difference DELTA LifeA be divided by and remaining life reading LifeBT2 and remaining life read
The counteracting of being divided by of difference DELTA LifeB.Remaining writing RWDA, RWDB being calculated by the above method, will not be by base
The influence of quasi- unit, and equally using data bit as base.
Therefore, even if solid state hard disk 100A, 100B has different specifications and makes the data in its S.M.A.R.T. information
With different bases, above-mentioned method still can effectively be standardized remaining writing RWDA, RWDB, accurately be estimated
The real surplus life-span of solid state hard disk 100A, 100B.
In step 205, in an embodiment, processing unit 104 can further calculate the unit of solid state hard disk 100A, 100B
Time average is according to writing WDPDA, WDPDA.
In different embodiments, unit time average data writing WDPDA, WDPDA can read example by processing unit 104
Such as, it but is not limited to the S.M.A.R.T. information of each solid state hard disk 100A, 100B and obtains, or correlation is executed by processing unit 104
Program record obtain.
Further, in step 206, processing unit 104 can be flat according to remaining writing RWDA, RWDB and unit time
Ratio between equal data writing WDPD determines that the residue of each solid state hard disk 100A, 100B uses time RTA, RTB.
In more detail, the residue of solid state hard disk 100A can pass through remaining writing RWDA and unit using time RTA
Time average is obtained according to the ratio between writing WDPD, and can following formula indicate:
RTA=RWDA/WDPD (formula 9)
On the other hand, when the residue of solid state hard disk 100B can be by remaining writing RWDB and unit using time RTB
Between ratio between average data writing WDPD obtain, and can following formula indicate:
RTB=RWDB/WDPD (formula 10)
In an embodiment, when the number of days per unit time of known corresponding unit time average data writing WDPD is distinguished
When for DP, can be calculated according to (formula 9) and (formula 10) remaining using number of days RDA and RDB, and following two formulas indicate respectively:
RDA=(RWDA/WDPD) × DP (formula 11)
RDB=(RWDB/WDPD) × DP (formula 12)
In step 207, processing unit 104 is to remaining writing RWDA, RWDB of each solid state hard disk 100A, 100B or surplus
It is remaining to be ranked up using time RTA, RTB, generate the remaining life ranking results of solid state hard disk 100A, 100B.
In an embodiment, processing unit 104 can generate notice signal and (not draw after generating remaining life ranking results
Show), to be transmitted to such as, but not limited to, user's device of pen electricity.
In another embodiment, processing unit 104 can specify solid state hard disk 100A and consolidate according to remaining life ranking results
The parity check code of remaining life shorter one storage disk array in state hard disk 100B.It whereby, can will be in parity check code collection
It is stored in remaining life shorter one (for example, solid state hard disk 100B).
For example, when the data volume of solid state hard disk 100A write-in reaches the 70% of its capacity, due to parity check code
Be centrally stored in solid state hard disk 100B, so when solid state hard disk 100B write-in data volume be likely to approach the 100% of its capacity,
Therefore storage server 1 can give notice signal automatically, inform that user preferentially replaces solid state hard disk 100B.Accordingly, user
The most short person of remaining life in current storage server 1 can be assessed to instantaneity, and is replaced remaining life in reasonable time
Most short person, to reach the optimal use efficiency of whole disk array.
It will illustrate the solid-state disk service life monitoring method applied to storage server 100 below with actual numerical value example
200。
Assuming that solid state hard disk 100A in January 1 12:00 from its S.M.A.R.T. information obtain starting time point T1 the remaining longevity
Life reading LifeAT1 be 20, and in January 31 the same year 12:00 from its S.M.A.R.T. information obtain terminate time point T2 residue
It is 19 that service life, which reads LifeAT2, then remaining life reading difference DELTA LifeA is 1.In addition, solid state hard disk 100A exists in this example
The data writing Δ WDA in this January 1 to January 31 is 100GB.
On the other hand, solid state hard disk 100B in January 1 12:00 from its S.M.A.R.T. information obtain starting time point T1
Remaining life read LifeBT1 be 30, and in January 31 the same year 12:00 from its S.M.A.R.T. information obtain terminate time point T2
Remaining life reading LifeBT2 be 20, then remaining life reading difference DELTA LifeB be 10.In addition, solid-state is hard in this example
Disk 100B is 50GB in this 1 day 1 month to January 31 data writing Δ WDB.
Please refer to table 1.Table 1 is in one embodiment of this disclosure of documents, above-mentioned solid state hard disk 100A and solid state hard disk 100B's
The numerical value tables of associated lifetime reading and data writing.
Table 1
For solid state hard disk 100A, according to formula (3), processing unit 104 by remaining life read difference DELTA LifeA (such as
1) force parameter EDA=1/ is restrained oneself to calculate the solid state hard disk of solid state hard disk 100A divided by data writing Δ WDA (such as 100GB)
100GB。
In general, remaining life reading difference DELTA LifeA is bigger in the case where data writing Δ WDA is constant, generation
The extent of deterioration of table solid state hard disk 100A caused by this data writing Δ WDA is higher.In other words, solid state hard disk endurance
Parameter EDA represents the degree that the data writing Δ WDA in a period of time causes the service life of solid state hard disk 100A to decline.
Then, according to formula (5), processing unit 104 is again by remaining life reading LifeAT2 (such as 19) divided by solid state hard disk
The solid state hard disk of 100A restrains oneself force parameter EDA (1/100GB), then can calculate the remaining writing RWDA=for terminating time point T2
19/ (1/100GB)=1900GB.
When user be averaged every month write-in unit time average data writing WDPDA be 100GB when, according to formula
(11), remaining writing RWDA (1900GB), unit time average data writing WDPD (a month 100GB) and per unit
Time number of days DP (1 month) is calculated remaining using number of days RDA=(1900GB/100GB) × 1=19 months.
Similarly, for solid state hard disk 100B, processing unit 104 is by the second variation difference (for example, 10) divided by the second number
The durability of a solid state hard disk 100B, for example, 10/50GB are calculated according to writing (for example, 50GB).
For solid state hard disk 100B, according to formula (4), processing unit 104 by remaining life read difference DELTA LifeB (such as
10) force parameter EDB=is restrained oneself to calculate the solid state hard disk of solid state hard disk 100B divided by data writing Δ WDB (such as 50GB)
10/50GB。
Then, according to formula (6), processing unit 104 is again by remaining life reading LifeBT2 (such as 20) divided by solid state hard disk
The solid state hard disk of 100B restrains oneself force parameter EDB (10/50GB), then can calculate the remaining writing RWDB=for terminating time point T2
20/ (10/50GB)=100GB.
When user be averaged every month write-in unit time average data writing WDPDA be 100GB when, according to formula
(12), remaining writing RWDA (100GB), unit time average data writing WDPDA (a month 100GB) and per unit
Time number of days DP (1 month) calculates residue using number of days RDA=(100GB/100GB) × 1 and residue and uses number of days RDB=
(100GB/100GB) × 1=1 months.
Please refer to table 2.Table 2 is, according to each numerical value in table 1, to calculate in the above described manner in one embodiment of this disclosure of documents
Obtained by data.
Table 2
By in the example of table 1 and table 2, although it can be seen that the remaining longevity that solid state hard disk 100B is obtained from S.M.A.R.T. information
Life reading LifeBT2 is greater than the remaining life reading LifeAT2 that solid state hard disk 100A is obtained from S.M.A.R.T. information, but passes through
It is less than the remaining writing of solid state hard disk 100A in fact by the remaining writing RWDB that the above method obtains solid state hard disk 100B
RWDA。
Referring to figure 3. and Fig. 4.Fig. 3 and Fig. 4 distinguishes in one embodiment of this disclosure of documents, solid state hard disk remaining life with write
Enter the schematic diagram of the relationship of data volume.
It is noted that storage server 1 is not aware that solid state hard disk 100A and solid-state are hard at the beginning in Fig. 3 and Fig. 4
The complexion of actual life curve LA, LB of disk 100B.However, in order to make it easy to understand, showing solid state hard disk in Fig. 3 and Fig. 4
The complexion of actual life curve LA, LB (it is indicated with solid line) of 100A and solid state hard disk 100B.
In Fig. 3, through above-mentioned solid-state disk service life monitoring method 200, solid state hard disk 100A and solid-state can be calculated
The respective solid state hard disk remaining life of hard disk 100B and write-in data volume relationship point PA, PB, and estimate from it is a certain specific when
The estimation of point (such as starting time point T1) extremely another specific time point (such as terminating time point T2) solid state hard disk 100A and solid state hard disk 100B
Life curve LA ', LB ' (being represented by dotted lines it).
In this example, the data writing of solid state hard disk 100B is slightly below the data writing of solid state hard disk 100A, and pre-
The hard disk remaining life for estimating solid state hard disk 100B is longer.
In an embodiment, after processing unit 104 can set the fixed data volume of every write-in one, that is, re-execute primary solid
State hard disk life-span monitoring method 200.As shown in figure 4, can be calculated solid through above-mentioned solid-state disk service life monitoring method 200
Relationship point PA, PB of state hard disk 100A and the respective solid state hard disk remaining life of solid state hard disk 100B and write-in data volume, and estimate
A specific time point (such as starting time point T1) is calculated to another specific time point (such as terminating time point T2) solid state hard disk 100A and solid state hard disk
Life expectancy curve LA ', the LB ' of 100B.
The data writing of solid state hard disk 100B is still below the data writing of solid state hard disk 100A at this time, and estimates solid-state
The hard disk remaining life of hard disk 100B is longer.And it is more and more with the data volume that person is written, the precision of estimation also can be mentioned gradually
Height, therefore user can be allowed accurately to estimate out whichever solid state hard disk and be only and answer the solid state hard disk changed that preferentially to eliminate (as solid
State hard disk 100A).
Based on the above embodiment, the solid-state disk service life monitoring method 200 that this disclosure of documents is proposed, can make solid-state hard
The remaining life criterion and quantity of disk is allowed whereby in storage system, and the remaining life of the solid state hard disk of different model can also compare each other
Compared with, and then the remaining life for the solid state hard disk that sorts out, the solid state hard disk for overcoming composition storage system need to be the limit of same model
System, and then make system operator more polynary in the selection of solid state hard disk.
In addition, solid-state disk service life monitoring method 200 provided by this disclosure of documents, can be applied to consolidating for different model
State hard disk is formed by array.Also, solid-state disk service life monitoring method 200 can voluntarily adjust the fineness of estimation,
Can dynamically adjust at any time and in the system of sorting out all solid state hard disks current remaining life.
Although this case is disclosed above with embodiment, so it is not limited to this case, and any this field tool is usually known
The knowledgeable, in the spirit and scope for not departing from this case, when can be used for a variety of modifications and variations, therefore the protection scope of this case is when view
Subject to the scope of which is defined in the appended claims.
Claims (10)
1. a kind of solid-state disk service life monitoring method, is suitable for inclusion in the storage server of multiple solid state hard disks, which is characterized in that
The monitoring method includes:
It captures remaining life of each solid state hard disk between starting time point and end time point and reads difference；
In the starting time point and the data writing for terminating to add up each solid state hard disk between time point；
The ratio between difference and the data writing, which is read, according to the remaining life determines each solid state hard disk
Solid state hard disk restrains oneself force parameter；And
Capture each solid state hard disk it is described terminate time point remaining life reading, with according to the remaining life reading and
The solid state hard disk restrains oneself the ratio between force parameter and determines each solid state hard disk in the remaining writing for terminating time point.
2. solid-state disk service life monitoring method according to claim 1, which is characterized in that at least two solid state hard disks
The base of the remaining life reading is different.
3. solid-state disk service life monitoring method according to claim 1, which is characterized in that each solid state hard disk it is described
Remaining writing is using data bit as base.
4. solid-state disk service life monitoring method according to claim 1, which is characterized in that the solid-state disk service life monitoring
Method also includes:
The remaining writing sequence to each solid state hard disk, generates the remaining life sequence knot of the multiple solid state hard disk
Fruit.
5. solid-state disk service life monitoring method according to claim 4, which is characterized in that the multiple solid state hard disk is formed
Disk array, the solid-state disk service life monitoring method also includes:
According to the remaining life ranking results, specifying, there is least residue service life person to store institute in the multiple solid state hard disk
State the parity check code of disk array.
6. solid-state disk service life monitoring method according to claim 1, which is characterized in that the solid-state disk service life monitoring
Method also includes:
Unit of account time average is according to writing；And
Each solid-state is determined according to the ratio between the remaining writing and the unit time average data writing
The residue of hard disk uses the time.
7. solid-state disk service life monitoring method according to claim 6, which is characterized in that the solid-state disk service life monitoring
Method also includes:
To the described remaining using time-sequencing of each solid state hard disk, the remaining life sequence of the multiple solid state hard disk is generated
As a result.
8. a kind of storage server, which is characterized in that the storage server includes:
Multiple solid state hard disks；
It configures to store the memory body of procedure code；
It is electrically coupled to the processing unit of the multiple solid state hard disk and the memory body, and is configured to be deposited from the memory body
Described program code is taken, to execute solid-state disk service life monitoring method, the solid-state disk service life monitoring method includes:
It captures remaining life of each solid state hard disk between starting time point and end time point and reads difference；
In the starting time point and the data writing for terminating to add up each solid state hard disk between time point；
The ratio between difference and the data writing, which is read, according to the remaining life determines each solid state hard disk
Solid state hard disk restrains oneself force parameter；And
Capture each solid state hard disk it is described terminate time point remaining life reading, with according to the remaining life reading and
Solid state hard disk restrains oneself the ratio between force parameter and determines each solid state hard disk in the remaining writing for terminating time point.
9. storage server according to claim 8, which is characterized in that the remaining longevity of at least two solid state hard disks
The base for ordering reading is different.
10. storage server according to claim 9, which is characterized in that the solid-state disk service life monitoring method is also wrapped
Contain:
Unit of account time average is according to writing；And
Each solid-state is determined according to the ratio between the remaining writing and the unit time average data writing
The residue of hard disk uses the time.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710911732.XA CN109582527A (en) | 2017-09-29 | 2017-09-29 | Storage server and its solid-state disk service life monitoring method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710911732.XA CN109582527A (en) | 2017-09-29 | 2017-09-29 | Storage server and its solid-state disk service life monitoring method |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN109582527A true CN109582527A (en) | 2019-04-05 |
Family
ID=65919133
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710911732.XA Pending CN109582527A (en) | 2017-09-29 | 2017-09-29 | Storage server and its solid-state disk service life monitoring method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109582527A (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110544503A (en) * | 2019-08-28 | 2019-12-06 | 深圳忆联信息系统有限公司 | test method for evaluating total write-in quantity of SSD and computer equipment |
| WO2021143133A1 (en) * | 2020-01-19 | 2021-07-22 | 苏州浪潮智能科技有限公司 | Residual life prediction method, apparatus and device for nonvolatile memory device, and medium |
| CN116774948A (en) * | 2023-08-18 | 2023-09-19 | 合肥联宝信息技术有限公司 | Data processing method, device, electronic device and storage medium |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101079324A (en) * | 2007-06-15 | 2007-11-28 | 华为技术有限公司 | Storage device, its life monitoring device and monitoring method |
| US20080140918A1 (en) * | 2006-12-11 | 2008-06-12 | Pantas Sutardja | Hybrid non-volatile solid state memory system |
| JP2013097575A (en) * | 2011-10-31 | 2013-05-20 | Oki Electric Ind Co Ltd | Semiconductor disk lifetime monitoring device |
| CN106297898A (en) * | 2015-06-03 | 2017-01-04 | 杭州海康威视数字技术股份有限公司 | The life-span method for early warning of a kind of NAND flash storage and device |
| CN106558346A (en) * | 2016-11-16 | 2017-04-05 | 杭州华澜微电子股份有限公司 | The method and device that a kind of solid-state disk service life based on RAIM frameworks is calculated |
-
2017
- 2017-09-29 CN CN201710911732.XA patent/CN109582527A/en active Pending
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20080140918A1 (en) * | 2006-12-11 | 2008-06-12 | Pantas Sutardja | Hybrid non-volatile solid state memory system |
| CN101079324A (en) * | 2007-06-15 | 2007-11-28 | 华为技术有限公司 | Storage device, its life monitoring device and monitoring method |
| JP2013097575A (en) * | 2011-10-31 | 2013-05-20 | Oki Electric Ind Co Ltd | Semiconductor disk lifetime monitoring device |
| CN106297898A (en) * | 2015-06-03 | 2017-01-04 | 杭州海康威视数字技术股份有限公司 | The life-span method for early warning of a kind of NAND flash storage and device |
| CN106558346A (en) * | 2016-11-16 | 2017-04-05 | 杭州华澜微电子股份有限公司 | The method and device that a kind of solid-state disk service life based on RAIM frameworks is calculated |
Non-Patent Citations (1)
| Title |
|---|
| 方才华: "《全程优化的固态硬盘垃圾回收方法》", 《计算机应用》 * |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110544503A (en) * | 2019-08-28 | 2019-12-06 | 深圳忆联信息系统有限公司 | test method for evaluating total write-in quantity of SSD and computer equipment |
| WO2021143133A1 (en) * | 2020-01-19 | 2021-07-22 | 苏州浪潮智能科技有限公司 | Residual life prediction method, apparatus and device for nonvolatile memory device, and medium |
| CN116774948A (en) * | 2023-08-18 | 2023-09-19 | 合肥联宝信息技术有限公司 | Data processing method, device, electronic device and storage medium |
| CN116774948B (en) * | 2023-08-18 | 2023-11-28 | 合肥联宝信息技术有限公司 | Data processing method, device, electronic device and storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US8472274B2 (en) | Using temperature sensors with a memory device | |
| US9799083B2 (en) | Correlating consumption and voltage data to identify line loss in an electric grid | |
| CN106462500A (en) | Apparatuses and methods for performing wear leveling operations | |
| US7992061B2 (en) | Method for testing reliability of solid-state storage medium | |
| CN105579970B (en) | The method of electronic system and its operation with storage disc life estimation mechanism | |
| TW200839773A (en) | Wear-leveling management and file distribution management of hybrid density memory | |
| US10126970B2 (en) | Paired metablocks in non-volatile storage device | |
| CN109582527A (en) | Storage server and its solid-state disk service life monitoring method | |
| US9229640B2 (en) | Inexpensive solid-state storage by throttling write speed in accordance with empirically derived write policy table | |
| CN104346286B (en) | Data storage device and flash memory control method | |
| CN105469834A (en) | Testing method for embedded flash memory | |
| CN106708912B (en) | Junk file identification and management method, identification device, management device and terminal | |
| US11199968B2 (en) | Using recurring write quotas to optimize utilization of solid state storage in a hybrid storage array | |
| CN107636623A (en) | The period is scrubbed in inspection based on power state | |
| CN103914397B (en) | Flash memory device and its management method | |
| WO2019076613A1 (en) | Device and method for controlling the data refresh cycles in reprogrammable non-volatile memories | |
| CN105825884B (en) | Memory operating method and relevant memory device | |
| EP3507809B1 (en) | Device and method for controlling refresh cycles of non-volatile memories | |
| CN105302679A (en) | Detection method and system for intelligent terminal storage stability | |
| CN110431534A (en) | Method and system for user experience event handling and analysis | |
| CN108052441A (en) | A kind of test method, system, device and the storage medium of hard disk performance level | |
| CN102237117A (en) | Method and device for calculating service life of mobile storage device | |
| JP2011150774A (en) | Stored multi-bit data characterized by multiple-dimensional memory states | |
| CN110444244A (en) | Store the test device of equipment read error error correcting capability | |
| CN116661683B (en) | Wear balance management method, system, equipment and medium for flash memory |
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 | ||
| AD01 | Patent right deemed abandoned |
Effective date of abandoning: 20220802 |
|
| AD01 | Patent right deemed abandoned |