CN1588299A - Method and relative device for hardware array data migration - Google Patents
Method and relative device for hardware array data migration Download PDFInfo
- Publication number
- CN1588299A CN1588299A CN 200410056276 CN200410056276A CN1588299A CN 1588299 A CN1588299 A CN 1588299A CN 200410056276 CN200410056276 CN 200410056276 CN 200410056276 A CN200410056276 A CN 200410056276A CN 1588299 A CN1588299 A CN 1588299A
- Authority
- CN
- China
- Prior art keywords
- data
- hard disk
- read
- write
- memory storage
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
- 238000000034 method Methods 0.000 title claims abstract description 51
- 230000005012 migration Effects 0.000 title abstract description 13
- 238000013508 migration Methods 0.000 title abstract description 13
- 230000005055 memory storage Effects 0.000 claims description 21
- 238000012545 processing Methods 0.000 claims description 20
- 230000001186 cumulative effect Effects 0.000 abstract 1
- 230000009897 systematic effect Effects 0.000 abstract 1
- 230000008569 process Effects 0.000 description 21
- 238000005516 engineering process Methods 0.000 description 10
- 230000003139 buffering effect Effects 0.000 description 7
- 230000035508 accumulation Effects 0.000 description 5
- 238000009825 accumulation Methods 0.000 description 5
- 238000010586 diagram Methods 0.000 description 3
- 241001269238 Data Species 0.000 description 2
- 238000013500 data storage Methods 0.000 description 2
- 230000005540 biological transmission Effects 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 238000000151 deposition Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 230000006870 function Effects 0.000 description 1
- 238000007726 management method Methods 0.000 description 1
- 230000002093 peripheral effect Effects 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 241000894007 species Species 0.000 description 1
Images
Landscapes
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
This invention relates to a method of data migration and correspondent set, read the data from the original hard disk and write into each hard disk of one hard disk array. This invention first read the data continually from the original hard disk, till the cumulative read data makes the data of each son data no less than one after separating each block of son data and write in the extreme, write the son data in each hard disk. Thus, this invention makes full use of every chance of writing array, makes each time write more data, shorts the systematic resource consume caused by writing control.
Description
Technical field
The invention provides a kind of method and relevant apparatus that carries out the hard disk array data migtation, refer in particular to a kind of can the minimizing and write the instruction control number of times to promote the data migration method and the relevant apparatus of data migtation usefulness.
Background technology
Computer system can be assisted user's management, computing, a large amount of data, data and the audio-video files of access, has become one of most important hardware foundation of advanced information society; And modern information dealer also is devoted to research and develop various technology so that computer system can be quicker, the more correctly more data of access, data.
In recent years,, a plurality of hard disks can be set in the modern computer system, and to utilize various data structure comprehensively be a DISK to Image array with these hard disks along with the cost of hard disk reduces, capacity raises.For instance, at the cheap hard disk array (RAID of redundancy, under the data structure of grade 0,2,3,4,5 (being RAID0, RAID2 to RAID5) Redundant array of inexpensivedisks) etc., deposit the mode that can divide (stripe) to interlock to the data of hard disk array in and be divided into many group subdatas, each is organized subdata and is deposited in (writing) respectively to the different hard disks of hard disk array simultaneously again.In like manner, when this batch data will be read out by hard disk array, then the different hard disks in the hard disk array can be read the subdata of its storage respectively, and recombinant goes out that original batch data.Via the data access that each hard disk to hard disk array walks abreast simultaneously, hard disk array just can provide higher data access usefulness.
But, because it is higher that the hardware of hard disk array is provided with cost, existing computer system all still only is provided with a hard disk mostly when dispatching from the factory, if the user has other needs, the hard disk that just can install other in computer system additional is realized out the hard disk array that usefulness is higher with upgrading.Perhaps, a hard disk array has been arranged in user's the computer system, but with the growth of data volume, the user also may additionally buy more new hard disk in hard disk array, so that the less hard disk array of script hard disk number is upgraded to the more hard disk array of hard disk number.When hard disk array is upgraded, the user just carries out data migtation (dataemigration) possibly, so that the data in the original hard disk (hard disk array) are read,, data are returned again the hard disk array of depositing to the upgrading again according to the data structure of upgrading back hard disk array.
Please refer to Fig. 1.Fig. 1 is the instruction flow synoptic diagram when carrying out data migtation in the routine techniques.For instance, suppose that the user of a certain computer system had only single hardware A originally, but the user has increased a hard disk B in this computer system, and will be combined into a hard disk array, as the hard disk array of RAID0 with hard disk A and hard disk B.In this case, the user will carry out data migtation, with the data read that originally is stored in hard disk A is come out (just hard disk A being used as is the Data Source hard disk), again according to the data structure of hard disk array with these data storage to hard disk array.Certainly, as be familiar with known to the technology personage, when computer system will be operated hard disk and carried out access, can come each hard disk is sent the instruction of reading or write data by hard disk controller in the chipset or hard disk array controller, as ATA (advanced interconnection technique, Advanced Technology Attachment) instruction under the specification is read accordingly or is write to control each hard disk.Read or write in the instruction of data at these, can indicate hard disk will arrive the data of how many hytes of access in which zone of its storage space.In general, in the single instruction of reading or write data, it can require the size of data of access that one maximum value is arranged.For instance, read into instruction as one under the ATA specification, the may command hard disks read out the data of 128K hyte in the zone of appointment at most.In like manner, single write instruction then at most may command one hard disks in the zone of appointment, write the data of 128K hyte; In other words, the single data that write instruction write extreme value and are the 128K hyte.And represented among Fig. 1, when carrying out data migtation with routine techniques exactly each hard disk is sent the control flow that reads or write instruction.
As shown in Figure 1, when routine techniques will begin to carry out data migtation, conventional computer system will be sent an instruction of reading 102 to hard disk A earlier, to control hard disk A by the data of reading the 128K hyte among the hard disk A.As aforementioned, in the hard disk array that hard disk A, B form, realize the hard disk array data structure of parallel access, deposit in to the data of this hard disk array and can be divided into two groups of subdatas, be written into respectively again among hard disk A and the hard disk B.So will mark off the subdata of two groups of each 64K hytes in the routine techniques of Fig. 1 according to the data that read out in the instruction 102, and send instruction 104A, 104B respectively to hard disk A, hard disk B simultaneously, write (storage) respectively to hard disk A and hard disk B with subdata with these two groups of 64K hytes.
After instruction 104A, 104B, routine techniques can send instruction 106 to hard disk A again, so that the data of among a hard disk A 128K hyte are read, same in the subdata of carrying out producing after data necessary is divided and handled each 64K hyte, instruct 108A, 108B that the subdata of these two groups of 64K is write to hard disk A and hard disk B respectively with writing of 64K hyte again.Repeat the steering order of above-mentioned " 128K reads, 64K-64K writes, 128K reads, 64K-64K write " constantly, routine techniques just can be finished data migtation to all data to be migrated in regular turn.
Yet above-mentioned routine techniques is the not good shortcoming of available energy also.As aforementioned, the hard disk of existing technology reads, writes instruction can be specified 128K hyte data at most in single instruction access.So routine techniques can be in instruction 102,106 with the reading command of 128K hyte by hard disk A in sense data, read out maximum data with indication hard disk A in single instruction as far as possible.But, after the data that obtain the 128K hyte, in order to meet the data structure of hard disk array, routine techniques can write instruction immediately with two of 64K hyte and come respectively the subdata of 64K hyte is write to hard disk A and hard disk B; But, single write instruction in fact at most may command one hard disks write the data of 128K hyte, and therefore routine techniques also just can't make full use of the single size of data extreme value that instruction can be used that writes.
In general, when will be when writing to hard disk with a file (for example being the file of 1024K hyte), if all indicate hard disk to write more data when writing instruction sending each time, as long as that instruct with fewer writing just can write to this document in the hard disk fully; Relatively, write instruction and the time all can only control each hard disk and write less data, then certainly will will send more a plurality of instructions that write continually, identical file could be write in the hard disk fully if send each time.And computer system will be sent instruction to hard disk and the time all will consume suitable system resource each time, so many more write instruction, will consume many more system resource.And the routine data migrating technology among Fig. 1 will write the specified data of instruction very little and the data that need carry out more repeatedly write because of each, and consumes more system resources, causes efficiency of data migration not good.
Summary of the invention
Therefore, fundamental purpose of the present invention, promptly be to propose the method and the relevant apparatus that can improve data migtation usefulness, it can make full use of and send the chance that writes instruction to each hard disk each time and make hard disk write the data of more (at most), to promote the usefulness of data migtation, overcome the shortcoming of routine techniques.
When carrying out data migtation with technology of the present invention, continue earlier to send reading command to Data Source hard disk/hard disk array, by the time the subdata that the data that accumulation is read divide out can meet the single data that write instruction and write extreme value, and just actual each hard disk is sent writes instruction.So, the present invention just can make full use of and send the chance that writes instruction each time, controls the data that each hard disk writes more (at most), writes the number of times that instruction is sent with minimizing, promotes the usefulness of data migtation.
For instance, when the user will be when the data migtation of a hard disk A be to the RAID0 hard disk array that hard disk A, hard disk B are formed, the reading command that meeting of the present invention is sent twice 128K hyte to hard disk A earlier to be obtaining the data of 256K hyte, and the structure that the data of this 256K hyte just can be divided according to the hard disk array data and the data of these two 128K hytes being write to respectively among hard disk A and the hard disk B with the instruction that writes of two 128K hytes.In other words, equally will be with in the data of 256K hyte the hard disk array by two hard disks of a hard disk migration, the present invention only need divide and sends 128K reading command, 128K reading command and two 128K that send simultaneously for three times in regular turn and write instruction.In comparison, the routine techniques among Fig. 1 then will divide and sends 128K reading command, two 64K for four times in regular turn and write instruction, 128K reading command, two 64K and write six instructions such as instruction, could similarly finish the data migtation of 256K hyte.Since the present invention in data migration process required send write the instruction less, also therefore, the present invention can finish data migtation with higher usefulness.
Description of drawings
The schematic flow sheet of associated hard disk steering order when Fig. 1 carries out data migtation for routine techniques.
Fig. 2 unifies the function block schematic diagram of embodiment for department of computer science of the present invention.
Fig. 3 carries out the schematic flow sheet of data migtation for computer system among Fig. 2.
The schematic flow sheet of associated hard disk steering order when Fig. 4 carries out for flow process among Fig. 3.
Fig. 5 carries out the schematic flow sheet of data migtation in another embodiment for computer system among Fig. 2.
The reference numeral explanation
10: computer system 12: central processing unit
14: chipset 16: controller
18: storer 20: migrator
22A: data read program 22B: BUF
22C: determining program 22D: write-in program
102-108B, 302-306B: instruction 200,400: flow process
202-214,402-414: step HD (1)-HD (N): hard disk
Embodiment
Please refer to Fig. 2; Fig. 2 is the synoptic diagram of computer system 10 1 embodiment of the present invention.Be provided with a central processing unit 12, a chipset 14, a storer 18, a controller 16 and each non-volatile memory storage HD (1) in the computer system 10 to HD (N).Central processing unit 12 is used for the running of host computer system 10, is responsible for the processing and the program implementation of data, data; Storer 18 as system storage can be a dynamic RAM, is used for storing needed data during central processing unit 12 running, data or program.Chipset 14 be electrically connected to central processing unit 12, storer 18 and controller 16 between, be used for managing the exchanges data between these circuit.Memory storage HD (1) to HD (N) can be hard disk, 16 of controllers can be a hard disk controller or a hard disk array controller, by bus (as ATA, serial ATA or the bus of SCSI, wherein ATA is Advanced TechnologyAttachment, SCSI is Small Computer System Interface) formed passage (channel) is electrically connected to each hard disk HD (1) to HD (N), so that send the instruction of reading or writing to each hard disk, control each hard disk and data read is come out or write in the hard disk, and comprehensively be a hard disk array these hard disks.Controller 16 can be the intelligent hard disk electronics (IDE that is incorporated in the chipset, Intelligent Drive Electronic) controller, or passing through periphery interconnection (PCI, Peripheral Communication Interconnect) bus with the auxiliary insert card form is electrically connected to the hard disk array controller of chipset 14.
In computer system 10, Data Transference Technology of the present invention can realize in the software mode with migrator 20.Be provided with a data read program 22A, a BUF 22B, a determining program 22C and a write-in program 22D in the migrator 20.Carry out data migtation as the user and data origin source hard disk will be read out and when writing to each hard disk of hard disk array again according to the data structure of hard disk array, central processing unit 12 just can be carried out each program in the migrator 20 to realize technology of the present invention.Wherein, when data read program 22A was performed, central processing unit 12 can send a plurality of reading command to the source of data migtation hard disk by chipset 14, controller 16; The source hard disk is carried out the data that these reading command read out can be cushioned, store (as being stored in the storer 18) by the execution of BUF 22B, and suitable in addition processing, as reading of data being carried out the computing of parity check sign indicating number and according to the data structure of hard disk array it being divided into each batch subdata etc.And when central processing unit 12 is carried out determining program 22C, just can judge that whether suitable permissible maximum writes each hard disk that extreme value writes to hard disk array in the instruction to write for subdata size that these buffered datas form.If then central processing unit 12 just can write instruction to actual the sending of each hard disk in the hard disk array by to the execution of write-in program 22D, each batch subdata of buffered data is write to respectively in each hard disk.So, the present invention just can make full use of each time and send the chance that writes instruction to hard disk, allow each hard disk in the hard disk array write the fashionable data that can both preferably write more (at most) carrying out data each time as far as possible, write the number of times that instruction is sent with minimizing, promote the usefulness of data migtation.
Situation when carrying out data migtation for further specifying computer system 10 please refer to Fig. 3 (and in the lump with reference to figure 2); Flow process when the flow process 200 among Fig. 3 is exactly the invention process data migtation.The following step is arranged in the flow process 200:
Step 202: begin to carry out data migtation, will reading out in the hard disk of data origin source, and deposit in again to each hard disk of hard disk array according to the data structure of hard disk array.For instance, in the computer system 10 a hard disk HD (1) is only arranged originally, but the user is buying more other hard disks HD (2) to HD (N), wishing to form to HD (N) hard disk array of a RAID0 with hard disk HD (1).After comprehensively going out hard disk array, the user just can carry out data migtation, the data read that originally is stored in hard disk HD (1) is come out and again according to the data structure of RAID0 with data storage to hard disk array.At this moment, hard disk HD (1) just can be considered the Data Source hard disk.
Step 204: each source hard disk is sent a reading command, will read in the hard disk of data origin source.In preferred embodiment of the present invention, when each time a source hard disk being sent a reading command, all can indicate this source hard disk to read out maximum data.The ATA specification instruction of discussing with the front is an example, because of its each reading command that hard disk is sent can make this hard disk read the data of 128K hyte at most, so the present invention in this step, just can indicate the source hard disk to read the data of 128K hyte in each reading command.In addition, if a plurality of sources hard disk is arranged, then can send a reading command respectively to this a plurality of sources hard disk simultaneously in this step.
Step 206: cushion, keep in the step 204 by the data that read out in each source hard disk; If necessary, then in addition suitable processing is calculated corresponding parity check sign indicating number etc. as the data structure at hard disk array, to form corresponding array data, just will deposit the data to the hard disk array in.
Step 208: whether the array data of buffering has been fit to write in each hard disk of hard disk array in the mode of preferable (the best) in the determining step 206.If its data volume of data of buffering not enough for a long time in the step 206, write (storage) each hard disk to the hard disk array with regard at once it being divided into subdata, so each batch subdata will less than single write instruction allowed write extreme value, equal to have wasted each hard disk in the hard disk array sent the chance that writes instruction.Therefore, if the data of buffering not enough for a long time in the step 206, the present invention will proceed to step 204 again, and origin source hard disk relaying is resumed studies and fetched data once again, and continues to accumulate, cushion the data that read in step 206.When the data accumulation of these bufferings is more than enough, just the size of the subdata that marks off when these buffered datas met write instruction write extreme value the time, the present invention just can proceed to step 210.Perhaps, finished, also can proceed to step 210 when data to be migrated have been read in step 204.
Step 210: each hard disk in the hard disk array sent simultaneously write instruction, each batch subdata is write (storage) to each hard disk.Through the running of the present invention in step 204,206 and 208, in this step, the present invention should be able to write the data that each hard disk that makes in the instruction in the hard disk array writes more (at most) at each, to make full use of each time each hard disk is sent the chance that writes instruction.
Step 212:, can get back to step 204 if also have other data to need migration.If not, then can proceed to step 214.
Step 214: end data migration.
For being described more specifically the situation that the present invention carries out data migtation, please continue with reference to figure 4 (and in the lump with reference to figure 3); Fig. 4 is flow process 200 of the present invention when carrying out the data migtation of 256K hyte, sends the situation of instruction to each hard disk.In Fig. 4, suppose to have in the computer system 10 two hard disk HD (1) and HD (2), and with hard disk HD (1) serve as the source hard disk, with the data migtation among the hard disk HD (1) to hard disk HD (1) and the formed RAID0 hard disk array of HD (2).When migration data began, meeting of the present invention was sent instruction 302 to the source hard disk earlier, makes its data of reading the 128K hyte (step 204 of flow process 200 just), and the data buffering of this 128K hyte is kept in (step 206).Because reading of data only had 128K, these data are divided into after two batches of subdatas according to the hard disk array data structure of two hard disks, the every batch of subdata is the 64K hyte only, do not reach as yet write instruction write extreme value (128K hyte just).Therefore, meeting of the present invention is got back to step 204 by step 208 in flow process 200, sends another instruction 304 (Fig. 3) again, reads the data of 128K hyte in the hard disk of origin source again.Thus, reading of data will be accumulate to 256K hyte (step 206).Arrived in the step 208, because the data of 256K hyte can form two batches and respectively be the subdata of 128K hyte after division, what reached that each writes instruction writes extreme value 128K hyte, so flow process 200 just can proceed to step 210, simultaneously two hard disks in hard disk array send instruction 306A and 306B, and the data of each 128K hyte are write in two hard disks of hard disk array.
Compare with the routine techniques among Fig. 1, be will be equally with the data of 256K hyte by migrating in the hard disk array of two hard disks in the source hard disk, the present invention only need divide the instruction of sending four read/writes for three times to each associated hard disk in Fig. 4, and the routine techniques of Fig. 1 just needs to divide the instruction of sending six read/writes for four times to each hard disk.Hence one can see that, and the present invention can reduce the number of read/write instruction really effectively in data migration process, and reduce the read/write instruction transmission, carry out negative effect to computer system performance, and then promote the usefulness of data migtation.
When carrying out data migtation, because the scale of data migtation, if any several sources hard disk, migrate to hard disk array of several hard disks or the like, all fix, so the present invention can be directly according to the scale of data migtation, calculate earlier and will send several times that reading command just can read enough data to the source hard disk, actual again send to instruct carry out data migtation.About this situation, please refer to Fig. 5 (and in the lump with reference to figure 2); Flow process 400 among Fig. 5 is the present invention carries out data migtation in computer system 10 another embodiment.The following step is arranged in the flow process 400:
Step 402: beginning data migtation.
Step 403: the scale according to data migtation determines a default reading times, just will send what reading command to the source hard disk, could accumulate enough data (just being enough to make full use of the size of data that writes extreme value that writes instruction).For instance, if from a source hard disk with data migtation to the RAID0 hard disk array of N hard disk, then should default reading times can be N.Because, N reading command accumulated the data of the N*128K hyte of reading, just can divide the back in data and form the subdata that N criticizes the 128K hyte, the convenient instruction (just meeting the instruction that writes that writes extreme value) that writes with the 128K hyte writes to N hard disk in the hard disk array respectively.
Step 404: be similar to the step 204 in the flow process 200, this step also is to send one simultaneously to each source hard disk to write instruction.
Step 406: be similar to step 206, read the data that obtain in buffering, the temporary step 404.
Step 408: whether the number of times that determining step 404 accumulations are carried out has met the default reading times of calculating in the step 403.If not, then get back to step 404 and repeat, the number of times that carries out up to step 404,406 accumulations has arrived default reading times (or all data to be migrated all are read out), proceeds to step 410 again.
Step 410: be similar to the step 210 among Fig. 3, actual the sending of the hard disk in each hard disk array writes data.After carrying out step 404,406 and 408, the data of buffering should meet the single extreme value that writes that writes in the instruction after being divided into subdata, send the chance that writes instruction to each hard disk each time and can make full use of, make it write the data of more (at most).
Step 412:, then get back in the step 404 and proceed if the data that are still waiting to move need migration; If not, then can proceed to step 414.
Step 414: end data migration.
As seen from the above description, flow process 400 of the present invention basically with flow process 200 equivalences, all be to continue reading of data in the hard disk of first origin source, dividing metapedes to make full use of the extreme value that writes that writes instruction up to the data that read, just reality writes to each hard disk in the hard disk array to write instruction with data.For instance, if will be by in the hard disk array with data migtation to two hard disk in the source hard disk, then the present invention just can calculate earlier in step 403: flow process 400 needs the source hard disk is sent the reading command of twice 128K hyte, just can obtain enough data (256K hyte just) making full use of the instruction that writes of two 128K hytes, these data are write two hard disks in the hard disk array respectively.Calculating default reading times is after two, just can with two 128K hyte reading command the data of 256K hyte be read earlier, again simultaneously with each hard disks that subdata after instruction will be divided writes to hard disk array that writes of two 128K hytes, as the instruction control flow process of Fig. 4.In Fig. 2, the algorithm of calculating default reading times can be implemented among the determining program 22C, can calculate default reading times in the step 403 by the execution of determining program 22C, and carry out step 408 according to this default reading times.
The computer system 10 in applying to Fig. 2, technical spirit of the present invention also can apply to the computer system of other species structures widely.For instance, in some computer system a plurality of hard disk/hard disk array controllers can be arranged, be built in intelligent hard disk electronics (IDE) controller in the chipset in comprising and be connected in the hard disk array controller of chipset, and technical spirit of the present invention is certainly also applicable to this kind computer system in the auxiliary insert card mode.For instance, be to control two hard disk A, B with intelligent hard disk electronic controller and form the hard disk array of RAID0 in certain computer system, and control four disk C, D, E, F with the hard disk array controller of auxiliary insert card form and form the hard disk array of another RAID0.When this computer system will migrate to disk C in the hard disk array of data by hard disk A, B to the hard disk array of F the time with technology of the present invention, the present invention just can send the reading command (just simultaneously sending 128K hyte reading command respectively to hard disk A, B in each time) of 128K hyte earlier at twice simultaneously respectively to source hard disk A, B by intelligent hard disk electronic controller, obtain the data of 4*128K hyte with accumulation.The instruction that writes that next just can utilize the hard disk array controller of auxiliary insert card form simultaneously to send the 128K hyte to disk C to F to make full use of the instruction that writes each time, makes disk C to F write-once more (at most) data as far as possible.In addition, the present invention also can be used to data by migrating in the hard disk array in the CD player (or flash memory devices, flash memory).
Compare with routine techniques, the present invention can make full use of each time and send the chance that writes instruction to each hard disk, thus in data migration process, can reduce as far as possible each hard disk is sent the number of times that writes instruction, and then promote the usefulness of data migtation.For instance, by Fig. 1 and Fig. 4 more as can be known, equally will be with the data of 256K hyte by the hard disk array that migrates to two hard disks in the source hard disk, the present invention only need divide and sends four read/writes instructions (Fig. 4) for three times, routine techniques then will divide and sends six read/writes instructions (Fig. 1) for four times and just can reach identical purpose, so the present invention can save the time and the system resources consumption of about 1/4 (being 1-3/4) in the process of data migtation, finish data migtation with higher usefulness, promote the operational paradigm of computer system.
The above only is preferred embodiment of the present invention, and all equivalences of carrying out according to claim of the present invention change and revise, and all should belong to covering scope of the present invention.
Claims (9)
1. method of between a plurality of memory storages, carrying out data migtation, it includes:
Carry out a data read, from a memory storage, to read one first data;
These first data are write to these a plurality of memory storages before, carry out one time data read more at least, from least one memory storage, to read one second data;
Produce an array data according to these first data and this second data;
This array data is divided into many batches of subdatas, and makes the size of data of each subdata be not less than one to write extreme value; And
Each batch subdata is write to respectively in each memory storage.
2. the method for claim 1 wherein when reading one second data from least one memory storage, reads this second data from the memory storage that stores these first data.
3. the method for claim 1, wherein this writes the size of data that extreme value is equivalent to these first data.
4. method of between a plurality of memory storages, carrying out data migtation, it includes:
Carry out a data read, from least one memory storage, to read data;
Produce an array data according to these data;
This array data is divided into many batches of subdatas; And
Carry out a determining step, whether write extreme value less than one with the size of data of judging each batch subdata; If not, then each batch subdata is write to respectively in each memory storage; If, then carry out this data read once again to read other data, and merge each secondary data and read the data that read and upgrade this array data according to this, be not less than this up to the subdata that this array data split and write extreme value, again each batch subdata is write to respectively in each memory storage.
5. method as claimed in claim 4 when wherein carrying out this data read, reads the fixing data of a size of data respectively at every turn in single memory storage.
6. method as claimed in claim 5, when wherein this writes extreme value and is equivalent to each data read by the size of data that is read in the reading device.
7. computer system, it includes:
A plurality of memory storages;
One central processing unit is used for executive routine to control the running of this computer system; And
One storer, it can store a migrator, and this migrator includes:
One data read program; When this central processing unit was carried out this data read program, this computer system of may command was by reading data at least one memory storage;
One BUF; When this central processing unit was carried out this data read program, this computer system of its may command producing an array data according to the data that read in this data read program, and was divided into many batches of subdatas with this array data; And
One determining program, when this central processing unit was carried out this determining program, this central processing unit can judge whether the size of data of each batch subdata in this BUF writes extreme value less than one; If not, then this each memory storage of central processing unit may command so that each subdata is write to respectively in each memory storage; If, then this central processing unit can carry out this data read program once again to read other data, and in this BUF, merge each secondary data and read the data that read and upgrade this array data according to this, be not less than this up to the subdata that this array data split and write extreme value, control each memory storage again so that each batch subdata is write to respectively in each memory storage.
8. computer system as claimed in claim 7 wherein when this central processing unit is carried out this data read program at every turn, reads the fixing data of a size of data respectively in single memory storage.
9. computer system as claimed in claim 8, when wherein this writes extreme value and is equivalent to each data read program and is performed by the size of data that is read in the reading device.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CNB2004100562768A CN1312570C (en) | 2004-08-06 | 2004-08-06 | Method and relative device for hardware array data migration |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CNB2004100562768A CN1312570C (en) | 2004-08-06 | 2004-08-06 | Method and relative device for hardware array data migration |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN1588299A true CN1588299A (en) | 2005-03-02 |
| CN1312570C CN1312570C (en) | 2007-04-25 |
Family
ID=34603214
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CNB2004100562768A Expired - Lifetime CN1312570C (en) | 2004-08-06 | 2004-08-06 | Method and relative device for hardware array data migration |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN1312570C (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN100342319C (en) * | 2005-09-29 | 2007-10-10 | 威盛电子股份有限公司 | Magnetic disk array instruction processing method |
| CN103336672A (en) * | 2013-06-28 | 2013-10-02 | 华为技术有限公司 | Data reading method, device and computer equipment |
| WO2016110004A1 (en) * | 2015-01-05 | 2016-07-14 | 中兴通讯股份有限公司 | Management method and apparatus for disk storage space, and storage device |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102520885B (en) * | 2011-12-15 | 2015-04-15 | 清华大学 | Data management system for hybrid hard disk |
Family Cites Families (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JPH06332623A (en) * | 1993-05-24 | 1994-12-02 | Mitsubishi Electric Corp | Array type recorder and recorder |
| US6058455A (en) * | 1997-07-02 | 2000-05-02 | International Business Corporation | RAID system having a selectable unattended mode of operation with conditional and hierarchical automatic re-configuration |
| US6347359B1 (en) * | 1998-02-27 | 2002-02-12 | Aiwa Raid Technology, Inc. | Method for reconfiguration of RAID data storage systems |
| US7096327B2 (en) * | 2000-03-24 | 2006-08-22 | Koninklijke Philips Electronics N.V. | System for and method of accessing blocks on a storage medium |
| JP3820999B2 (en) * | 2002-01-25 | 2006-09-13 | ソニー株式会社 | Proximity communication system and proximity communication method, data management apparatus and data management method, storage medium, and computer program |
-
2004
- 2004-08-06 CN CNB2004100562768A patent/CN1312570C/en not_active Expired - Lifetime
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN100342319C (en) * | 2005-09-29 | 2007-10-10 | 威盛电子股份有限公司 | Magnetic disk array instruction processing method |
| CN103336672A (en) * | 2013-06-28 | 2013-10-02 | 华为技术有限公司 | Data reading method, device and computer equipment |
| CN103336672B (en) * | 2013-06-28 | 2016-10-05 | 华为技术有限公司 | Method for reading data, device and computing device |
| WO2016110004A1 (en) * | 2015-01-05 | 2016-07-14 | 中兴通讯股份有限公司 | Management method and apparatus for disk storage space, and storage device |
Also Published As
| Publication number | Publication date |
|---|---|
| CN1312570C (en) | 2007-04-25 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN104794070B (en) | Solid state flash memory write buffer system and method based on dynamic non-covered RAID technique | |
| US7254668B1 (en) | Method and apparatus for grouping pages within a block | |
| US8316178B2 (en) | Buffering of data transfers for direct access block devices | |
| US8131969B2 (en) | Updating system configuration information | |
| CN1148658C (en) | Method and system for high speed buffer storage management RAID storage system | |
| CN102122235B (en) | RAID4 (redundant array of independent disks) system and data reading and writing method thereof | |
| CN1258713C (en) | Data distribution dynamic mapping method based on magnetic disc characteristic | |
| CN107678685A (en) | The key assignments memory management method of store path optimization based on flash memory | |
| CN101034375A (en) | Computer memory system | |
| CN1635580A (en) | Flash memory medium data writing method | |
| CN1282067C (en) | Device and relative method for hardware array appositive operation | |
| TW202418087A (en) | Apparatus and method for distributing and storing write data in plural memory regions | |
| CN1945519A (en) | Storage control apparatus and storage control method | |
| CN1588299A (en) | Method and relative device for hardware array data migration | |
| CN1093964C (en) | I/O device having check recover function | |
| US7313651B2 (en) | Method and related apparatus for data migration of disk array | |
| CN1862479A (en) | Magnetic disk array data configuration structure and data acces method thereof | |
| CN1290019C (en) | Method and apparatus for hard disk array data transfer | |
| Pang et al. | PcGC: A Parity-Check Garbage Collection for Boosting 3-D NAND Flash Performance | |
| CN1238783C (en) | Disc damage managing method | |
| CN1238788C (en) | First-in first-out register quenue arrangement capable of processing variable-length data and its control method | |
| CN1776648A (en) | Method and device for getting residual space for FAT file system | |
| CN1276360C (en) | Management method of reconstructing memory | |
| CN113868149A (en) | Data storage method, device and equipment and readable storage medium | |
| JP2009080549A (en) | Disk array controller and data layout 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 | ||
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant | ||
| CX01 | Expiry of patent term |
Granted publication date: 20070425 |
|
| CX01 | Expiry of patent term |