CN100367181C

CN100367181C - Hard disk re-recognition system and method in redundancy array of inexpensive disks

Info

Publication number: CN100367181C
Application number: CNB2005100046576A
Authority: CN
Inventors: 宋建福; 刘文涵; 陈蕴弘
Original assignee: Inventec Corp
Current assignee: Inventec Corp
Priority date: 2005-01-21
Filing date: 2005-01-21
Publication date: 2008-02-06
Anticipated expiration: 2025-01-21
Also published as: CN1808366A

Abstract

The present invention relates to a reidentifying system for hard disks in a redundant array of an inexpensive disk (RAID), which can be executed by a computer, and a method of the system. The system comprises a comparison temporary storage table, a hard disk identifying module, an RAID establishing module and an RAID repairing module. When an RAID causes errors on computer access because of the movement of hard disks, the increase of hard disks, sequence regulation, etc., the system can be utilized to reidentify hard disks and establish the comparison temporary storage table to modify a hard disk information table, and then, the system modifies a sub area comparison table in an RAID controlling module to ensure that the arrangement corresponding relationship of the hard disks and the RAID can be established again to maintain the normal running of the RAID.

Description

Hard disk recognition system and method thereof again in the Redundant Array of Inexpensive Disc

Technical field

The present invention relates to the executable hard disk of a kind of computing machine recognition system and method thereof again, be particularly related to and be used for Redundant Array of Inexpensive Disc, when the configuration change of hard disk, produce a contrast scratch list, Redundant Array of Inexpensive Disc can be matched with hard disk again, and the system that can resume operation and method thereof.

Background technology

Be in the present information age, all files all store with the electronic record form, have replaced to put down in writing with the form of paper originally, have not only saved paper, have saved the space of depositing archives more greatly.The data in whole library, an a plurality of rigid disk of need can all store after electronization, but also so has more highlighted the importance of the normal operation of keeping disk.

Keep the security of storage and the integrality of data, avoid causing incalculable damage because of the fault of disk, in 1987, by Patterson, Gibson and these three people of Katz in UC Berkley, delivered the paper of " A Case of Redundant Array of Inexpensive Disks (Redundant Array of Inexpensive Disc scheme) " by name, its basic idea is exactly that less, the relatively inexpensive disk of a plurality of capacity is made up, make it in data security, and the big capacity hard disk that surpasses a costliness on the access usefulness.This designed concept is accepted soon, and since then, the technology of Redundant Array of Inexpensive Disc has obtained widespread use, and data storing has entered quicker, safer, the more cheap New Times.

Originally in design, Redundant Array of Inexpensive Disc is divided into grade 0 to 5, by each disk sector being divided into block subregion of the same size, and get a subregion from different disks respectively and come out, be reassembled as a volume (volume), again all clove hitch are combined into a volume group (VG, Volume Group), its benefit is the file that the capacity of the single disk of script can't store, do not need the user to cut apart voluntarily, can deposit in by the Redundant Array of Inexpensive Disc technology, and the volume group of forming by a plurality of disk.

In addition, one file is when access, originally can only in a disk, carry out, yet volume group (VG at Redundant Array of Inexpensive Disc, Volume Group) in, because a file all is to disperse simultaneously to deposit in each disk partition of a volume, the mode of parallel processing has increased the speed that stores greatly, and is as the same when reading.

Originally when making up Redundant Array of Inexpensive Disc, be to take to control by software, yet this kind mode quite expends the resource of CPU (central processing unit) (CPU), because when computing machine in when operation, the action of I/O (I/O) continues to carry out, if each data the time all will be monitored in read-write, then CPU (central processing unit) (CpU) will not have unnecessary resource and handle other scheduling, cause the computing machine overall efficiency not increase counter subtracting.Therefore, way afterwards all is to take the mode of hardware controls, make up whole Redundant Array of Inexpensive Disc by a raid controller, after this, computing machine only needs data are given raid controller, how to store distribution as for these data and then is responsible for by raid controller.

When setting up Redundant Array of Inexpensive Disc, the corresponding relation of respectively rolling up in the subregion of each hard disk and the Redundant Array of Inexpensive Disc is fixed, in case hard disk occur in sequence change, after perhaps increasing newly, removing hard disk, will make Redundant Array of Inexpensive Disc write fashionable making a mistake reading, originally in the data of this subregion, because of the change that occurs in sequence of hard disk, and seek less than, cause system to move at last.

Summary of the invention

In view of above problem, fundamental purpose of the present invention is to provide a kind of Redundant Array of Inexpensive Disc hard disk recognition system and method thereof again, when Redundant Array of Inexpensive Disc writes data reading, when finding that the hard disk order and the hard disk order of the subregion table of comparisons are wrong, promptly must discern hard disk again, Redundant Array of Inexpensive Disc can be moved once more.

Therefore, for reaching above-mentioned purpose, the present invention discloses a kind of Redundant Array of Inexpensive Disc hard disk recognition system again, comprises at least:

One Redundant Array of Inexpensive Disc is set up module,, and occurs in sequence when changing record and revise the Redundant Array of Inexpensive Disc partition information of this hard disk in the hard disk information table in order to a plurality of hard disks are built into a Redundant Array of Inexpensive Disc at hard disk;

One hard disk identification module in order to reading the Redundant Array of Inexpensive Disc partition information that is write down in each hard disk information table, and when the hard disk order changes, begins to set up the contrast scratch list;

One contrast scratch list, stored the hard disk order before change with change after the contrast that changes of subregion device number; And

One Redundant Array of Inexpensive Disc is repaired module, in order to when the hard disk identification module finds that hard disk occurs in sequence change, according to subregion device number in the amended hard disk information table construction subregion table of comparisons again.

According to above-mentioned conception, wherein this contrast scratch list is just set up in an internal memory when each described hard disk occurs in sequence change.

According to above-mentioned conception, wherein the field of this contrast scratch list comprises a hard disk code name, a Redundant Array of Inexpensive Disc name, a former subregion device number, a new subregion device number, and the order in the Redundant Array of Inexpensive Disc.

According to above-mentioned conception, wherein should the contrast scratch list revise the respectively subregion device number of this hard disk information table in order to contrast.

In addition, the present invention also discloses hard disk recognition methods again in a kind of Redundant Array of Inexpensive Disc, and this method comprises the following step: confirm in a plurality of hard disk information tables have more than one hard disk code name that change is arranged; Stop the operation of this Redundant Array of Inexpensive Disc; Read respectively this subregion table of comparisons; Ask for a plurality of new subregion device numbers to an operating system; Set up a contrast scratch list; Revise respectively this hard disk information table according to this contrast scratch list; And rebuild a subregion table of comparisons according to this hard disk information table respectively.

According to above-mentioned conception, wherein this affirmation respectively this hard disk information table be to be responsible for inspection by a hard disk identification module.

According to above-mentioned conception, wherein this contrast scratch list is just to set up in an internal memory when each described hard disk occurs in sequence change.

Disclosed method has comprised following steps:

At first, can the hard disk identification module continue to check that the hard disk code name has change? to avoid the situation of data read write error; If hard disk changes because of increasing newly, remove, change to make to put in order, and reading of causing that computing machine can't be correct writes data, then stop the operation of Redundant Array of Inexpensive Disc, read the subregion table of comparisons of Redundant Array of Inexpensive Disc, then ask for subregion device number new on each hard disk, set up the contrast scratch list of new and old subregion device number to operating system.

Set up module by Redundant Array of Inexpensive Disc and revise the subregion device number that the hard disk information table of each hard disk is write down according to contrasting scratch list, Redundant Array of Inexpensive Disc is repaired module according to the subregion device number that is write down on the amended hard disk information table, revise the subregion device number of the subregion table of comparisons of Redundant Array of Inexpensive Disc, make corresponding that hard disk and Redundant Array of Inexpensive Disc can be correct, at last, after the subregion device number all finished by the renewal of contrast scratch list, Redundant Array of Inexpensive Disc just can recover normal running.

Description of drawings

Fig. 1 is a system construction drawing of the present invention;

Fig. 2 is a method flow diagram of the present invention;

Fig. 3 is the contrast scratch list of one embodiment of the invention;

Fig. 4 a is the former hard disk information table of one embodiment of the invention;

Fig. 4 b is the new hard disk information table of one embodiment of the invention;

Fig. 5 a is the former subregion table of comparisons of one embodiment of the invention; And

Fig. 5 b is the new subregion table of comparisons of one embodiment of the invention.

Wherein, description of reference numerals is as follows:

100 Redundant Array of Inexpensive Discs are set up module

110 hard disk identification modules

120 contrast scratch list

130 Redundant Array of Inexpensive Discs are repaired module

140 hard disk groups

More than 150 hard disk information table

The 160 subregion tables of comparisons

170 Redundant Array of Inexpensive Discs

180 operating systems

Does step 200 check that the hard disk code name has change?

Step 210 stops the Redundant Array of Inexpensive Disc operation

Step 220 reads the subregion table of comparisons

Step 230 is asked for new subregion device number to operating system

Step 240 is set up the contrast scratch list

Step 250 is revised the hard disk information table according to the contrast scratch list

Step 260 is rebuild the subregion table of comparisons according to the hard disk information table

Embodiment

Please refer to Fig. 1, this is a system construction drawing of the present invention, and hard disk groups 140 can be built into a Redundant Array of Inexpensive Disc 170, and native system comprises mainly: (A) Redundant Array of Inexpensive Disc is set up module 100; (B) the hard disk identification module 110; (C) contrast scratch list 120; Reach (D) Redundant Array of Inexpensive Disc reparation module 130.Now, be described in detail as follows respectively with regard to its function:

(A) Redundant Array of Inexpensive Disc is set up the hard disk groups 140 of module 100 in order to a plurality of hard disk is gathered, be built into a Redundant Array of Inexpensive Disc 170, and respectively within it in the hard disk information table 150 of portion, record is by operating system (O.S., operating system) the 180 subregion device numbers of being given; Wherein, each hard disk information table 150 is stored in the retention data zone of each hard disk, the field information that write down its inside please refer to Fig. 4 a or Fig. 4 b, this is the field composition of hard disk information table 150, comprise: a hard disk code name, one Redundant Array of Inexpensive Disc name, a subregion device number, and the order in the Redundant Array of Inexpensive Disc.

Be that example explains now with the hard disk information table 150 that is write down in the hard disk 1, wherein first the row content " hard disk 1 ", the hard disk code name that is this hard disk is " hard disk 1 ", in order to represent this hard disk.

The content of second row " volume 1 " means that this hard disk belongs to first volume of these Redundant Array of Inexpensive Disc that is integrated into 170 inside; The sign of the third line " 17 " is the subregion device number of just having been given when reading hard disk by operating system, and Redundant Array of Inexpensive Disc 170 can find this subregion according to this subregion device number; " 1 " filled out in last column represents the order that this hard disk is endowed in Redundant Array of Inexpensive Disc 170, do not change because the order of hard disk changes.

(B) the hard disk identification module 110, in order to read the information that is write down in each hard disk information table 150; When Redundant Array of Inexpensive Disc is set up module 100 with hard disk groups 140, when being built into a Redundant Array of Inexpensive Disc 170, can be in each hard disc data storage area, choose a retention data zone and deposit hard disk information table 150, content is put down in writing the subregion device number of each hard disk, and belong in the Redundant Array of Inexpensive Disc 170 which the volume, series arrangement all Redundant Array of Inexpensive Disc partition informations why.

(C) contrast scratch list 120 when hard disk identification module 110 finds that occurring in sequence of hard disk changes, just begins to set up this table, with the change on the subregion device number that stores each hard disk; Hard disk identification module 110 is write fashionable carrying out reading of data by the subregion table of comparisons 160 of Redundant Array of Inexpensive Disc to this hard disk respectively, after discovery does not meet with the hard disk information table, its subregion device number of hard disk journal of pressing the subregion table of comparisons 160 is in this table, and ask for after changing in the hard disk order to operating system 180, give the new subregion device number of each hard disk again, and insert in this table by hard disk code name and Redundant Array of Inexpensive Disc name, with as of the corresponding modification of hard disk information table 150 with the subregion table of comparisons 160.

Please refer to Fig. 3, this is contrast scratch list 120 and composition thereof, comprise a hard disk code name, one Redundant Array of Inexpensive Disc name, belong to which volume in the Redundant Array of Inexpensive Disc 170 in order to write down this fdisk, one former subregion device number, one new subregion device number, before and after the change of hard disk order, give by operating system, allow Redundant Array of Inexpensive Disc 170 can find desired data by the subregion device number, and the order in the Redundant Array of Inexpensive Disc, represent the order of this disk in Redundant Array of Inexpensive Disc 170.

(D) Redundant Array of Inexpensive Disc is repaired module 130, in order to be established when contrast scratch list 120, and after making Redundant Array of Inexpensive Disc set up module 100 to revise the subregion device number of finding in the hard disk information tables 150 and finish according to contrast scratch list 120, according to remodifying the Redundant Array of Inexpensive Disc partition information of finishing on each hard disk, set up the subregion table of comparisons 160 on the Redundant Array of Inexpensive Disc once more.

Disclosed method has comprised following steps:

Please refer to Fig. 2, at first, can hard disk identification module 110 check that the hard disk code name of hard disk information table 150 has change? (step 200) to avoid the situation of data read write error; If hard disk groups 140 changes because of increasing newly, remove, change to make to put in order, and reading of causing that computing machine can't be correct writes data, then stop the operation (step 210) of Redundant Array of Inexpensive Disc 170, in cheap redundant arrays of inexpensive disks 170, read the hard disk order and the subregion device number (step 220) thereof of the subregion table of comparisons 160 by hard disk identification module 110.

Then, by ask for it gives each hard disk again because the order of hard disk changes new subregion device number (step 230) to operating system 180; According to the new and old subregion device number of gained, set up the contrast scratch list 120 (step 240) of a new and old subregion device number; Then, by hard disk code name of being put down in writing in the contrast scratch list 120 and corresponding new subregion device number thereof, revise the hard disk information table 150 (step 250) of each hard disk; At last, utilize and revise the hard disk information table of finishing 150, repair the subregion table of comparisons 160 (step 260) that module 130 is rebuild Redundant Array of Inexpensive Disc 170 by Redundant Array of Inexpensive Disc, it is one correct corresponding that hard disk and Redundant Array of Inexpensive Disc 170 are obtained, and the unification of subregion device number also makes reading that data can be correct.

Please refer to Fig. 3, Fig. 4 a, Fig. 4 b, Fig. 5 a and Fig. 5 b, to do the embodiment explanation of Redundant Array of Inexpensive Disc 170 for one or three hard disks, when hard disk identification module 110 is wanted reading of data according to the subregion table of comparisons 160, the subregion device number of finding hard disk groups 140 changes, stop the System Operation of Redundant Array of Inexpensive Disc 170 immediately, read the hard disk code name " hard disk 1 " of the former subregion table of comparisons 160 of Fig. 5 a, " hard disk 2 ", and " hard disk 3 ", and corresponding former subregion device number " 17 ", " 33 ", and " 49 ", and ask for corresponding new subregion device number " 49 " to operating system, " 17 ", and " 33 ", in order to set up contrast scratch list 120 as shown in Figure 3, after foundation is finished, then revise the hard disk information table 150 of each hard disk by contrast scratch list 120, be depicted as the former hard disk information table 150 of each hard disk as Fig. 4 a, contrast becomes Fig. 4 b after revising; At last, rebulid the new subregion table of comparisons 160 shown in Fig. 5 b according to the hard disk information table 150 of each hard disk again, then Redundant Array of Inexpensive Disc 170 can resume operation again.

Though the present invention discloses as above with aforesaid preferred embodiment; right its is not in order to limit the present invention; any those of ordinary skill in the art; without departing from the spirit and scope of the present invention; therefore can do a little change and retouching, scope of patent protection of the present invention must be looked this instructions appending claims and defined and be as the criterion.

Claims

1. hard disk recognition system again in the executable Redundant Array of Inexpensive Disc of computing machine is characterized in that this system has:

One contrast scratch list has stored the contrast that a plurality of subregion device numbers of a plurality of hard disks before order changes and after the change change;

One hard disk identification module in order to reading a plurality of Redundant Array of Inexpensive Disc partition informations that write down in a plurality of hard disk information tables, and when the hard disk order changes, begins to set up this contrast scratch list;

One Redundant Array of Inexpensive Disc is set up module, in order to a plurality of hard disks are built into a Redundant Array of Inexpensive Disc, and respectively writing down each described hard disk information table in this hard disk, and occur in sequence when changing at hard disk, revise Redundant Array of Inexpensive Disc partition information in the hard disk information table of each described hard disk by this contrast scratch list; And

One Redundant Array of Inexpensive Disc is repaired module, in order to when this hard disk identification module finds that hard disk occurs in sequence change, rebuilds the subregion table of comparisons according to each the described subregion device number in amended each described hard disk information table.

2. hard disk recognition system again in the executable Redundant Array of Inexpensive Disc of computing machine as claimed in claim 1 is characterized in that this contrast scratch list just sets up in an internal memory when each described hard disk occurs in sequence change.

3. hard disk recognition system again in the executable Redundant Array of Inexpensive Disc of computing machine as claimed in claim 1, the field that it is characterized in that this contrast scratch list comprises a hard disk code name, one Redundant Array of Inexpensive Disc name, one former subregion device number, one new subregion device number, and the order in the Redundant Array of Inexpensive Disc.

4. hard disk recognition system again in the executable Redundant Array of Inexpensive Disc of computing machine as claimed in claim 1 is characterized in that this contrast scratch list revises the respectively subregion device number of this hard disk information table in order to contrast.

5. hard disk recognition methods again in the executable Redundant Array of Inexpensive Disc of computing machine is characterized in that this method comprises the following step:

Confirm in a plurality of hard disk information tables have more than one hard disk code name that change is arranged;

Stop the operation of this Redundant Array of Inexpensive Disc;

Read in a subregion table of comparisons of this Redundant Array of Inexpensive Disc;

Ask for a plurality of new subregion device number of described hard disk to an operating system;

Set up a contrast scratch list;

Revise respectively this hard disk information table according to this contrast scratch list; And

Rebuild a subregion table of comparisons according to this hard disk information table respectively.

6. hard disk recognition methods again in the executable Redundant Array of Inexpensive Disc of computing machine as claimed in claim 5, it is characterized in that this affirmation respectively this hard disk information table be to be responsible for inspection by a hard disk identification module.

7. hard disk recognition methods again in the executable Redundant Array of Inexpensive Disc of computing machine as claimed in claim 5 is characterized in that this contrast scratch list is just to set up in an internal memory when each described hard disk occurs in sequence change.

8. hard disk recognition methods again in the executable Redundant Array of Inexpensive Disc of computing machine as claimed in claim 7, the field that it is characterized in that this contrast scratch list comprises a hard disk code name, one Redundant Array of Inexpensive Disc name, one former subregion device number, one new subregion device number, and the order in the Redundant Array of Inexpensive Disc.

9. hard disk recognition methods again in the executable Redundant Array of Inexpensive Disc of computing machine as claimed in claim 5 is characterized in that this contrast scratch list is in order to contrast the respectively subregion device number of this hard disk information table of modification.