CN103488434A - Method for improving disk array reliability - Google Patents
Method for improving disk array reliability Download PDFInfo
- Publication number
- CN103488434A CN103488434A CN201310434139.2A CN201310434139A CN103488434A CN 103488434 A CN103488434 A CN 103488434A CN 201310434139 A CN201310434139 A CN 201310434139A CN 103488434 A CN103488434 A CN 103488434A
- Authority
- CN
- China
- Prior art keywords
- data
- disk
- dish
- hotspare
- backup
- 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
Images
Landscapes
- Signal Processing For Digital Recording And Reproducing (AREA)
Abstract
The invention discloses a method for improving disk array reliability. The method includes the following steps: providing a data backup hot spare for conducting backup on data of each disk in a data writing process; providing a data transfer hot spare for conducting data transfer on a certain disk when the disk is predicted to fail possibly. The data backup hot spare and the data transfer hot spare are provided. The data backup hot spare is used for conducting the backup on the data of each disk in the data writing process. The data transfer hot spare is used for conducting the data transfer on a certain disk when the disk is predicted to fail possibly. Therefore, data reconstruction time can be reduced, guarantee is provided for smooth completion of reconstruction, and system reliability can be greatly improved.
Description
Technical field
The invention belongs to computer system and computer memory technical field, relate to a kind of method of strengthening the disk array reliability.
Background technology
Field of computer, for the RAID5 system with redundancy check information, can support the inefficacy of member dish, and when a member is coiled inefficacy, whole system is in degrading state, this time can add HotSpare disk to be reconstructed system data.Yet, adding heat dish to be reconstructed system data is a job very consuming time, must travel through all data that the member coils and carry out the data that verification calculates failure disc, and along with the increase of disk size, the time of this process need will be longer, expends time in.
The restructuring procedure of disk array has two kinds of off-line reconstruct and on-line reorganizations, and off-line reconstruct refers to that disk array stops services request in restructuring procedure, after reconstruct completes, reactivates service again; And on-line reorganization refers to that in restructuring procedure, disk array still externally provides service.In actual application environment, the customer requirements storage system can provide continual service, so the study general of reconstructing method is all to consider on-line reorganization.
In restructuring procedure, if occur that again another piece member coils inefficacy, due to the redundancy scheme of RAID5, just system can't have been recovered again, whole disk array will lose efficacy so.And the time of data reconstruction is longer, occur again to have the member to coil the probability of inefficacy just larger, the reliability of system is just lower.
Therefore the defect existed for above-mentioned prior art, be necessary to develop research in fact, so that a kind of scheme to be provided, solve the problem that above-mentioned prior art exists, can reduce the time of data reconstruction, for the completing and provide safeguard smoothly of data reconstruction, and improve the reliability of system.
Summary of the invention
For addressing the above problem, the object of the present invention is to provide a kind of method of strengthening the disk array reliability, to reduce the time of data reconstruction, completing and providing safeguard smoothly, the reliability of raising system for data reconstruction.
For achieving the above object, technical scheme of the present invention is:
A kind of method of strengthening the disk array reliability, comprise the steps:
One blocks of data backup HotSpare disk is provided, for writing the process of data, the data of every dish is backed up;
Provide a blocks of data to shift HotSpare disk, for when certain dish of prediction lost efficacy possibly, the dish that may lose efficacy to this prediction carries out the data transfer;
Wherein, when creating, data backup HotSpare disk and data transfer HotSpare disk are also divided according to the band of RAID5, band corresponding relation and RAID5 are in full accord.
Further, first data block of data backup HotSpare disk backs up the data of first dish, and second data block backs up the second blocks of data.
Further, while carrying out write operation, a data synchronous write to a dish, the data asynchronous write of other dish.
Further, the parameter that increases by two kinds of defect attribute with disk natural quality and disk is analyzed disk, shifts by data the dish that HotSpare disk may lose efficacy to analyses and prediction and carries out the data transfer.
Further, while carrying out data reconstruction, if finally that piece dish of inefficacy is that piece dish predicted and has carried out the migration of partial data, just data are shifted to HotSpare disk and add above, proceed the reconstruct of remaining data; If last that piece dish lost efficacy is that piece dish predicted but does not also carry out the migration of partial data, just the data backup HotSpare disk is added above, proceed the reconstruct of remaining data; If not being the dish predicted or the result that there is no prediction, the last dish lost efficacy occurs, just upper the interpolation of data backup HotSpare disk, proceed the reconstruct of remaining data.
Compared to prior art, a kind of method of strengthening the disk array reliability of the present invention provides a blocks of data backup HotSpare disk and a blocks of data to shift HotSpare disk.Wherein, the data backup HotSpare disk is being for writing the process of data, and the data of every dish are backed up; And data shift HotSpare disk for when certain dish of prediction lost efficacy possibly, the dish that may lose efficacy to this prediction carries out the data transfer.Thereby, can reduce time of data reconstruction, be the completing and provide safeguard smoothly of reconstruct, can greatly improve the reliability of system.
The accompanying drawing explanation
Fig. 1 is disc array structure of the present invention and data layout schematic diagram;
Fig. 2 is the machine-processed flow process diagram that the present invention predicts low-quality disk.
Embodiment
In order to make purpose of the present invention, technical scheme and advantage clearer, below in conjunction with drawings and Examples, the present invention is further elaborated.Should be appreciated that specific embodiment described herein, only in order to explain the present invention, is not intended to limit the present invention.
As shown in Figure 1, a kind of method of strengthening the disk array reliability of the present invention provides a blocks of data backup HotSpare disk and a blocks of data to shift HotSpare disk.Wherein, the data backup HotSpare disk is being for writing the process of data, and the data of every dish are backed up; And data shift HotSpare disk for when certain dish of prediction lost efficacy possibly, the dish that may lose efficacy to this prediction carries out the data transfer.Thereby, can reduce time of data reconstruction, be the completing and provide safeguard smoothly of reconstruct, can greatly improve the reliability of system.
Particularly, in the RAID5 system, provide a blocks of data backup HotSpare disk and a blocks of data to shift HotSpare disk, when creating, data backup HotSpare disk and data transfer HotSpare disk are also divided according to the band of RAID5, band corresponding relation and RAID5 are in full accord.Wherein, the data backup HotSpare disk is being for writing the process of data, and a part of data of every dish are backed up.First data block of data backup HotSpare disk backs up the data of first dish, second data block backs up the second blocks of data ... when write operation is arranged, when the data block of the band at place writes, go above also this data block being write to the position of corresponding data backup HotSpare disk.As shown in Figure 1, the 3rd band of take is example, and data C12 is write to the data backup HotSpare disk.
Become performance bottleneck for fear of HotSpare disk, can be only to the data synchronous write of a dish, the data asynchronous write of other dish.For asynchronous write, data are to complete backup to shift when the I/O mouth is relatively idle, and the bitmap record need to be arranged.When carrying out read operation, need not be read the read operation of HotSpare disk data.
Shift HotSpare disk by data and carry out the data transfer to predicting the dish that may lose efficacy, prediction to low-quality disk is judged by the parameter variation of disk, the parameter of disk is divided into measurable parameter and unpredictable parameter, and unpredictable parameter is nonsensical, does not consider.Collect measurable parameter by a thread cycle, these parameters are analyzed, when finding that certain disk parameter is abnormal, think that this dish may lose efficacy.Parameter from two kinds in the embodiment of the present invention is carried out, and as shown in Figure 2, the parameter that increases by two kinds of defect attribute with disk natural quality and disk is analyzed.Wherein, disk natural quality Main Analysis service time and temperature, differentiate algorithm and adopt threshold values to compare; Disk increases defect attribute Main Analysis error count statistics, bad sector quantity, addressing error counting etc., differentiates algorithm and can adopt threshold values comparison, Bayesian probability statistics or rank test statistics scheduling algorithm.After predicting certain dish inefficacy, just the data of this dish are started anew to move to data transfer HotSpare disk and get on, and the position of record migration.The data of the dish that may lose efficacy are shifted out, back up in advance and reconstruct, thereby greatly reduced the probability of loss of data.In the embodiment of the present invention, when the data that this dish has been moved are carried out write operation again, equally also to write data transfer HotSpare disk and get on, to keep consistency.
When RAID5 demotes, occur that single member coils inefficacy, while carrying out data reconstruction, minute several situations are processed: when that piece dish of last inefficacy is that piece dish predicted and has carried out the migration of partial data, just data are shifted to HotSpare disk and add above, proceed the reconstruct of remaining data; Be that piece dish predicted when that piece dish of last inefficacy but also do not carry out the migration of partial data, just the data backup HotSpare disk is added above, proceed the reconstruct of remaining data; When the dish of last inefficacy is not that the dish predicted or the result that there is no prediction occur, just the data backup HotSpare disk is added above, proceed the reconstruct of remaining data.
Due to the position of having recorded migration, can start order downwards according to this position to the standby operation of coiling of data heat of transfer and carry out data reconstruction.For the data backup HotSpare disk, as shown in Figure 1, if the 3rd dish lost efficacy, after changing HotSpare disk, due to the write operation before, data block has been write to HotSpare disk, originally the data block on the 3rd dish can have a part, as shown in Figure 1 the C12 data block on HotSpare disk, and and the relation of band be corresponding, therefore, for this data block, do not need reconstruct directly to skip just passable, need not read in, even if also need not carry out verification, so just greatly reduced the time of recovering.And the C0 of HotSpare disk, other data blocks such as C1, because the data with failure disc are not corresponding, directly covered just passable by verification result of calculation.When other dish lost efficacy, situation is also identical.
The foregoing is only preferred embodiment of the present invention, not in order to limit the present invention, all any modifications of doing within the spirit and principles in the present invention, be equal to and replace and improvement etc., within all should being included in protection scope of the present invention.
Claims (5)
1. a method of strengthening the disk array reliability, is characterized in that, comprises the steps:
One blocks of data backup HotSpare disk is provided, for writing the process of data, the data of every dish is backed up;
Provide a blocks of data to shift HotSpare disk, for when certain dish of prediction lost efficacy possibly, the dish that may lose efficacy to this prediction carries out the data transfer;
Wherein, when creating, data backup HotSpare disk and data transfer HotSpare disk are also divided according to the band of RAID5, band corresponding relation and RAID5 are in full accord.
2. strengthen as claimed in claim 1 the method for disk array reliability, it is characterized in that: first data block of data backup HotSpare disk backs up the data of first dish, and second data block backs up the second blocks of data.
3. strengthen as claimed in claim 2 the method for disk array reliability, it is characterized in that: while carrying out write operation, a data synchronous write to a dish, the data asynchronous write of other dish.
4. strengthen as claimed in claim 3 the method for disk array reliability, it is characterized in that: the parameter that increases by two kinds of defect attribute with disk natural quality and disk is analyzed disk, shifts by data the dish that HotSpare disk may lose efficacy to analyses and prediction and carries out the data transfer.
5. strengthen as claimed in claim 4 the method for disk array reliability, it is characterized in that: while carrying out data reconstruction, if last that piece dish lost efficacy is that piece dish predicted and has carried out the migration of partial data, just data are shifted to HotSpare disk and add above, proceed the reconstruct of remaining data; If last that piece dish lost efficacy is that piece dish predicted but does not also carry out the migration of partial data, just the data backup HotSpare disk is added above, proceed the reconstruct of remaining data; If not being the dish predicted or the result that there is no prediction, the last dish lost efficacy occurs, just upper the interpolation of data backup HotSpare disk, proceed the reconstruct of remaining data.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201310434139.2A CN103488434A (en) | 2013-09-23 | 2013-09-23 | Method for improving disk array reliability |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201310434139.2A CN103488434A (en) | 2013-09-23 | 2013-09-23 | Method for improving disk array reliability |
Publications (1)
Publication Number | Publication Date |
---|---|
CN103488434A true CN103488434A (en) | 2014-01-01 |
Family
ID=49828703
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201310434139.2A Pending CN103488434A (en) | 2013-09-23 | 2013-09-23 | Method for improving disk array reliability |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN103488434A (en) |
Cited By (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN104461771A (en) * | 2014-11-03 | 2015-03-25 | 北京百度网讯科技有限公司 | Data backup processing method and device |
CN109962857A (en) * | 2017-12-26 | 2019-07-02 | 中国电信股份有限公司 | Flow control methods, device and computer readable storage medium |
CN110058808A (en) * | 2018-01-17 | 2019-07-26 | 富士通株式会社 | Reduce the device of the Data Migration time of data rearrangement between storage hierarchical layer |
Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN1692335A (en) * | 2002-11-01 | 2005-11-02 | 国际商业机器公司 | Method and means for tolerating multiple dependent or arbitrary double disk failures in a disk array |
US20120151253A1 (en) * | 2010-12-14 | 2012-06-14 | Western Digital Technologies, Inc. | System and method for maintaining a data redundancy scheme in a solid state memory in the event of a power loss |
CN102508620A (en) * | 2011-12-01 | 2012-06-20 | 浪潮电子信息产业股份有限公司 | Method for processing RAID5 (Redundant Array of Independent Disks) bad sector |
US8209587B1 (en) * | 2007-04-12 | 2012-06-26 | Netapp, Inc. | System and method for eliminating zeroing of disk drives in RAID arrays |
CN103064753A (en) * | 2012-10-25 | 2013-04-24 | 浪潮电子信息产业股份有限公司 | Disk array three-stage recovery system designing method |
CN103207761A (en) * | 2013-04-17 | 2013-07-17 | 浪潮(北京)电子信息产业有限公司 | Data backup method and data reconfiguration method for RAID (redundant arrays of independent disks) 5 system hot backup disks |
-
2013
- 2013-09-23 CN CN201310434139.2A patent/CN103488434A/en active Pending
Patent Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN1692335A (en) * | 2002-11-01 | 2005-11-02 | 国际商业机器公司 | Method and means for tolerating multiple dependent or arbitrary double disk failures in a disk array |
US8209587B1 (en) * | 2007-04-12 | 2012-06-26 | Netapp, Inc. | System and method for eliminating zeroing of disk drives in RAID arrays |
US20120151253A1 (en) * | 2010-12-14 | 2012-06-14 | Western Digital Technologies, Inc. | System and method for maintaining a data redundancy scheme in a solid state memory in the event of a power loss |
CN102508620A (en) * | 2011-12-01 | 2012-06-20 | 浪潮电子信息产业股份有限公司 | Method for processing RAID5 (Redundant Array of Independent Disks) bad sector |
CN103064753A (en) * | 2012-10-25 | 2013-04-24 | 浪潮电子信息产业股份有限公司 | Disk array three-stage recovery system designing method |
CN103207761A (en) * | 2013-04-17 | 2013-07-17 | 浪潮(北京)电子信息产业有限公司 | Data backup method and data reconfiguration method for RAID (redundant arrays of independent disks) 5 system hot backup disks |
Cited By (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN104461771A (en) * | 2014-11-03 | 2015-03-25 | 北京百度网讯科技有限公司 | Data backup processing method and device |
CN109962857A (en) * | 2017-12-26 | 2019-07-02 | 中国电信股份有限公司 | Flow control methods, device and computer readable storage medium |
CN109962857B (en) * | 2017-12-26 | 2022-08-30 | 中国电信股份有限公司 | Flow control method, flow control device and computer readable storage medium |
CN110058808A (en) * | 2018-01-17 | 2019-07-26 | 富士通株式会社 | Reduce the device of the Data Migration time of data rearrangement between storage hierarchical layer |
CN110058808B (en) * | 2018-01-17 | 2022-08-02 | 富士通株式会社 | Apparatus for reducing data migration time for rearranging data between storage hierarchy levels |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
US9798620B2 (en) | Systems and methods for non-blocking solid-state memory | |
CN102184129B (en) | Fault tolerance method and device for disk arrays | |
US9128846B2 (en) | Disk array device, control device and data write method | |
US8417989B2 (en) | Method and system for extra redundancy in a raid system | |
US20150371684A1 (en) | Ultra high capacity ssd | |
CN101916173B (en) | RAID (Redundant Array of Independent Disks) based data reading and writing method and system thereof | |
CN104484251B (en) | A kind of processing method and processing device of hard disk failure | |
CN103246478B (en) | A kind of based on the disc array system of software PLC support without packet type overall situation HotSpare disk | |
CN103513942B (en) | The reconstructing method of raid-array and device | |
JP2006259894A (en) | Storage control device and method | |
CN104035830A (en) | Method and device for recovering data | |
CN101567211A (en) | Method for improving usability of disk and disk array controller | |
US20140245062A1 (en) | Preventing unrecoverable errors during a disk regeneration in a disk array | |
US9785438B1 (en) | Media cache cleaning based on workload | |
CN102968361A (en) | RAID (Redundant Array of Independent Disk) data self-repairing method | |
CN102591746B (en) | Data reconstruction method and storage equipment | |
CN103019894B (en) | Reconstruction method for redundant array of independent disks | |
CN103870352A (en) | Method and system for data storage and reconstruction | |
US9378092B2 (en) | Storage control apparatus and storage control method | |
CN103019623A (en) | Memory disc processing method and device | |
CN102023902A (en) | Disc array reconstruction method | |
US20140281316A1 (en) | Data management device and method for copying data | |
CN103488434A (en) | Method for improving disk array reliability | |
CN102789370B (en) | A kind of RAID array synchronous method and device | |
CN107728943B (en) | Method for delaying generation of check optical disc and corresponding data recovery method |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
C06 | Publication | ||
PB01 | Publication | ||
C10 | Entry into substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20140101 |
|
WD01 | Invention patent application deemed withdrawn after publication |