CN105117282A - Input and output request splitting method and device - Google Patents
Input and output request splitting method and device Download PDFInfo
- Publication number
- CN105117282A CN105117282A CN201510524313.1A CN201510524313A CN105117282A CN 105117282 A CN105117282 A CN 105117282A CN 201510524313 A CN201510524313 A CN 201510524313A CN 105117282 A CN105117282 A CN 105117282A
- Authority
- CN
- China
- Prior art keywords
- request
- data
- current
- length
- bio
- 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
Landscapes
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention discloses an input and output request splitting method and device. The method comprises: receiving an IO request delivered by an application layer; for different service applications, acquiring chunk sizes of service operations; and according to a length of IO data corresponding to the IO request and the chunk size of the service operation, splitting the IO request, so that each IO request resulting from the splitting corresponds to a separate chunk. According to the method and device provided by the invention, an IO splitting mechanism is established above a plug-in layer (namely a common layer), which can free the plug-in layer from caring about whether an IO is across chunks (chunk), wherein visible IO sizes (iosize) of the plug-in layer are all less than or equal to a chunksize, so as to reduce complexity and redundancy of a disaster recovery plug-in layer, and to maintain consistency among plug-ins. Moreover, adding other spaces for buffering or backing up original IO data is not needed. Splitting in all IO modes can be satisfied, and splitting granularity is changeable according to the chunksize.
Description
Technical field
The present invention relates to memory technology, espespecially a kind of method that input output request is split and device.
Background technology
Along with the fast development of computer technology, digital information progressively replaces paper media's data, presents average daily exponential growth.Mass data has expedited the emergence of the development of cloud computing, greatly data and mass storage system (MSS), also more and more higher to the demand of memory device and storage application.Government's industry, banking system, electric power enterprise etc. have higher requirement to the security of data and business continuance.The calamity of high-end storage systems nowadays is also more and more important for characteristic.
Clone, snapshot, remote copy is comprised for plug-in unit in the calamity storing data Layer.Three plug-in units all need to carry out the data processing of corresponding calamity for characteristic to different mode input and output IO.Meanwhile, for different business application (database or file system), it is variable that calamity should maximize adaptation business for the block size (chunksize) of relational operation.
Summary of the invention
In order to solve the problems of the technologies described above, the invention provides a kind of method that input output request is split and device, complicacy and the redundancy of layer plug can be reduced, the consistance of service data.
In order to reach the object of the invention, the invention provides a kind of method that input output request is split, comprising:
Receive the input and output I/O request that application layer issues;
For different service application, obtain the block size of business operation;
The length of I/O data corresponding according to I/O request and the block size of described business operation, split described I/O request, makes each I/O request after fractionation correspond to an independent block.
Further, the length of the described I/O data corresponding according to I/O request and the block size of described business operation, split I/O request, comprising:
If the length of the I/O data that current I/O request is corresponding in the length range that current block holds, does not then split current I/O request.
Further, the length of the described I/O data corresponding according to I/O request and the block size of described business operation, split I/O request, comprising:
If the length of the I/O data that current I/O request is corresponding exceeds length that current block holds and the length of I/O data that indicates of at least one page of vector parameter bio_vec that current block can hold current I/O request, then in units of the length of the I/O data of bio_vec instruction, using the I/O data that the I/O data that current block holds is asked as current I O, the I/O data do not held by current block is as the I/O data of next I/O request.
Further, the length of the described I/O data corresponding according to I/O request and the block size of described business operation, split I/O request, comprising:
If the length of the I/O data that current I/O request is corresponding exceeds length that current block holds and current block only can hold the I/O data of the partial-length of the first bio_vec instruction of current I/O request, the I/O data that the I/O data of the partial-length of the first bio_vec instruction then held by current block is asked as current I O, the I/O data of the residue length of the first bio_vec instruction do not held by current block is as the I/O data of next I/O request, the I/O data that other bio_vec of the current I/O request do not held by current block indicate is as the I/O data of next I/O request again.
Further, the length of the described I/O data corresponding according to I/O request and the block size of described business operation, split I/O request, comprising:
If the length of the I/O data that current I/O request is corresponding exceeds length that current block holds and current block only can hold the I/O data of the partial-length of the first bio_vec instruction of current I/O request, the I/O data that the I/O data of the partial-length of the first bio_vec instruction then held by current block ask as current I O, the I/O data of other bio_vec instruction of the current I/O request that the I/O data of the residue length of the first bio_vec instruction do not held by current block and current block do not hold is as the I/O data of next I/O request.
Further, a kind of device split input output request, comprising: receiving element, acquiring unit and split cells, wherein:
Described receiving element, for receiving the input and output I/O request that application layer issues;
Described acquiring unit, for for different service application, obtains the block size of business operation;
Described split cells, for according to the length of I/O data corresponding to I/O request and the block size of described business operation, splits described I/O request, makes each I/O request after fractionation correspond to an independent block.
Further, described split cells, according to the length of I/O data corresponding to I/O request and the block size of described business operation, splits I/O request, comprising:
If the length of the I/O data that current I/O request is corresponding in the length range that current block holds, does not then split current I/O request.
Further, described split cells, according to the length of I/O data corresponding to I/O request and the block size of described business operation, splits I/O request, comprising:
If the length of the I/O data that current I/O request is corresponding exceeds length that current block holds and the length of I/O data that indicates of at least one page of vector parameter bio_vec that current block can hold current I/O request, then in units of the length of the I/O data of bio_vec instruction, using the I/O data that the I/O data that current block holds is asked as current I O, the I/O data do not held by current block is as the I/O data of next I/O request.
Further, described split cells, according to the length of I/O data corresponding to I/O request and the block size of described business operation, splits I/O request, comprising:
If the length of the I/O data that current I/O request is corresponding exceeds length that current block holds and current block only can hold the I/O data of the partial-length of the first bio_vec instruction of current I/O request, the I/O data that the I/O data of the partial-length of the first bio_vec instruction then held by current block is asked as current I O, the I/O data of the residue length of the first bio_vec instruction do not held by current block is as the I/O data of next I/O request, the I/O data that other bio_vec of the current I/O request do not held by current block indicate is as the I/O data of next I/O request again.
Further, described split cells, according to the length of I/O data corresponding to I/O request and the block size of described business operation, splits I/O request, comprising:
If the length of the I/O data that current I/O request is corresponding exceeds length that current block holds and current block only can hold the I/O data of the partial-length of the first bio_vec instruction of current I/O request, the I/O data that the I/O data of the partial-length of the first bio_vec instruction then held by current block ask as current I O, the I/O data of other bio_vec instruction of the current I/O request that the I/O data of the residue length of the first bio_vec instruction do not held by current block and current block do not hold is as the I/O data of next I/O request.
Compared with prior art, the present invention includes the I/O request receiving application layer and issue; For different service application, obtain the block size of business operation; The length of I/O data corresponding according to I/O request and the block size of described business operation, split described I/O request, makes each I/O request after fractionation correspond to an independent block.The present invention's (i.e. common layer) on layer plug sets up a set of IO and splits mechanism, layer plug can be made not need whether to pay close attention to IO across block (chunk), the visible IO size (iosize) of layer plug is all less than or equal to chunksize, thus reduce calamity for the complicacy of layer plug and redundance, safeguard the consistance between many plug-in units; Further, do not need add other spatial caches or back up former I/O data; Can meet the fractionation under all IO patterns, and it is variable with chunksize to split granularity.
Other features and advantages of the present invention will be set forth in the following description, and, partly become apparent from instructions, or understand by implementing the present invention.Object of the present invention and other advantages realize by structure specifically noted in instructions, claims and accompanying drawing and obtain.
Accompanying drawing explanation
Accompanying drawing is used to provide the further understanding to technical solution of the present invention, and forms a part for instructions, is used from and explains technical scheme of the present invention, do not form the restriction to technical solution of the present invention with the embodiment one of the application.
Fig. 1 is the process flow diagram to the method that input output request splits of the present invention;
Fig. 2 is the schematic diagram that I/O request of the present invention splits principle;
Fig. 3 be in the embodiment of the present invention calamity for the integrated stand composition of module;
Fig. 4 is the Organization Chart to the device that input output request splits of the present invention.
Embodiment
For making the object, technical solutions and advantages of the present invention clearly understand, hereinafter will be described in detail to embodiments of the invention by reference to the accompanying drawings.It should be noted that, when not conflicting, the embodiment in the application and the feature in embodiment can combination in any mutually.
Can perform in the computer system of such as one group of computer executable instructions in the step shown in the process flow diagram of accompanying drawing.Further, although show logical order in flow charts, in some cases, can be different from the step shown or described by order execution herein.
As shown in Figure 1, the method split input output request of the present invention, comprising:
Step 101: receive the I/O request that application layer issues;
Step 102: for different service application, obtains the block size of business operation;
Such as, be exactly that (parameter is right with type-value composition for parameter in the order of traversal establishment relation for storing calamity standby, a corresponding value of type), get block (chunk) this type, value is exactly block size (chunksize).
Step 103: the length of I/O data corresponding according to I/O request and the block size of business operation, split I/O request, makes each I/O request after fractionation correspond to an independent block.
As shown in Figure 2, the data that application layer issues are all be made up of in order I/O request (IO) one by one.I/O data is stored in correspondence position in corresponding page by page vector parameter (bio_vec) one by one by IO in internal memory.In internal memory, be discrete between bio_vec, bio_vec corresponds to the data on physical disk is one by one Coutinuous stores.Wherein, each bio_vec structure comprises three members:
structbio_vec{
Bv_page; The page of // this bio_vec in internal memory
Bv_offset; The start offset of // this bio_vec in internal memory in page
Bv_len; The length of // this bio_vec in internal memory in page
}；
On physical disk, plug-in unit manages data in units of chunk.Public module on plug-in unit splits I/O request according to the length of I/O data corresponding to I/O request and the block size of business operation, namely correspond to the chunk position of physical disk according to the bio_vec information of current I O request, point three kinds of situations split I/O request:
(1) if the length of I/O data corresponding to current I/O request is in the length range that current block holds, then current I/O request is not split.
Current I O afterbody is less than or equal to chunk border, from the reference position of the corresponding chunk of current I O, if the skew sum of all bio_vec does not exceed current chunk remain open ended length, then current I O does not need to carry out any fractionation, is directly issued to layer plug and carries out IO process.
(2) if the length of I/O data corresponding to current I/O request exceeds length that current block holds and the length of I/O data that indicates of at least one bio_vec that current block can hold current I/O request, then in units of the length of the I/O data of bio_vec instruction, using the I/O data that the I/O data that current block holds is asked as current I O, the I/O data do not held by current block is as the I/O data of next I/O request.
One or more bio_vec of current I O are less than or equal to chunk border and not across chunk: from the reference position of the corresponding chunk of current I O, if the skew sum of all bio_vec exceedes current chunk remain open ended length, then calculate the open ended maximum bio_vec of current chunk.Using first of current I O bio_vec to maximum bio_vec as segmentation IO, be issued to layer plug and carry out IO process, simultaneously using the remainder of current I O as next IO, i.e. the bIO_vev attribute of overall IO, IO sector, iosize upgrade.
(3) if the length of I/O data corresponding to current I/O request exceeds length that current block holds and current block only can hold the I/O data of the partial-length of the first bio_vec instruction of current I/O request, the I/O data that the I/O data of the partial-length of the first bio_vec instruction then held by current block is asked as current I O, the I/O data of the residue length of the first bio_vec instruction do not held by current block is as the I/O data of next I/O request, the I/O data that other bio_vec of the current I/O request do not held by current block indicate is as the I/O data of next I/O request again.
First bio_vec of current I O is across chunk: from the reference position of the corresponding chunk of current I O, if the skew sum of all bio_vec exceedes current chunk remain open ended length, then calculate the open ended maximum bio_vec of current chunk.If current chunk first bio_vec does not hold, illustrate that first bio_vec is just across chunk, now need this bio_vec to split according to chunk border.Bio_vec before and after chunk border is handed down to layer plug respectively as two IO and carries out IO process.Then using current I O remainder as next current I O.
For above-mentioned situation (3), the I/O data that the I/O data of the partial-length of the first bio_vec instruction also current block can held ask as current I O, the I/O data of other bio_vec instruction of the current I/O request that the I/O data of the residue length of the first bio_vec instruction do not held by current block and current block do not hold is as the I/O data of next I/O request
So, according to above three kinds of situations to current I O process, until all bio_vec process.IO is handed down to layer plug after splitting be all drop on one independently in chunk, and plug-in unit does not need whether to pay close attention to IO across chunk, and the visible iosize of layer plug is all less than or equal to chunksize, only needs to carry out data processing according to plug-in unit characteristic in units of chunk.Because this reducing layer plug complicacy and redundancy, maintaining multiple plug-in unit IO and splitting consistance.
Standby for calamity this method being applied to data storage below, method of the present invention is described.
Being illustrated in figure 3 calamity for module whole Organization Chart, is User space and kernel state framework from top to bottom respectively.User space comprises user operation commands (aegis-admin) and monitoring agent (aegis-monitor).Kernel state comprises user interface layer, general framework layer, public library layer, calamity for layer plug and IO adaptation layer.User interface layer is used for mutual between User space and kernel state, log information.General framework layer builds the general frame of calamity for subsystem for providing, and provides plug-in management, IO mapping, the management of IO time-out, IO to split and system configuration management function.Public library layer provides public service for plug-in unit for concrete calamity.Calamity realizes concrete data protection characteristic for layer plug, and the plug-in management interface that each plug-in unit calls general framework layer is registered to calamity in subsystem.IO adaptation layer docking IO manager or bIO generic block layer.By the design of plug-in type, reach calamity maximum multiplexing for subsystem internal, concrete business module is provided in the mode of plug-in unit meanwhile, be convenient to positioning problems and concurrent development.
IO splits and belongs to general framework layer, is positioned at calamity on layer plug.When User space has IO to issue, converting kernel IO to by scst and arrive general framework layer, obtaining chunksize (this value is arranged for during relation by creating calamity) by inquiry.Split by the IO of this layer and current I O is issued to corresponding calamity for layer plug in units of chunk.Layer plug carries out process for characteristic to I/O data according to calamity and returns to IO manager, and IO manager terminates IO.
The IO that please refer again to Fig. 2 splits principle schematic.Each IO of sending down service exists with bio_vec chained list in internal memory, and be discrete between bio_vec, each bio_vec comprises three attributes: place page, page bias internal, length.Multiple bio_vec corresponding I/O operation object (physical medium) the preceding paragraph continuous print data space of current I O.When current I O issues, inside save as it and applied for 3 bio_vec, bv_page1, bv_page2 and bv_page3 in corresponding internal memory, physical medium is one section of continuous space from sec_begin to sec_end respectively.And the standby block size chunksize specifying physical medium data manipulation when tying up to establishment that closes of calamity, so how IO is mapped to corresponding chunk, so that it is that IO splits the thing that will do that plug-in unit carries out calamity for performance data management.Particularly:
All bio_vec that situation 1: current I O comprises all drop in current chunk, and namely the right margin of current chunk is more than or equal to sec_end.Directly issue IO to layer plug process.
All bio_vec that situation 2: current I O comprises all do not drop in current chunk, and namely the right margin of current chunk is less than sec_end.So calculating the open ended maximum bio_vec of current chunk, is the 2nd bio_vec here.The 1st of current I O bio_vec is split to the 2nd bio_vec as an IO, be issued to layer plug and carry out IO process, sec_begin is moved to the reference position of the 3rd piece, the bio_vec chained list of current I O is adjusted to only has block 3 simultaneously, upgrades iosize and IO sector.By remainder as next current I O process.
First bio_vec of situation 3: current I O just cannot all drop in current chunk, and namely this bio_vec is across chunk.Two IO orders are so divided into by first bio_vec to be issued to calamity for plug-in unit.Again using second bio_vec and the 3rd bio_vec as the 3rd I/O request.Or first bio_vec is divided into two parts, and the part held by chunk is as first IO, and the part do not held by chunk and second bio_vec and the 3rd bio_vec are as second IO.
As shown in Figure 4, present invention also offers a kind of device that input output request is split, comprising: receiving element, acquiring unit and split cells, wherein:
Described receiving element, for receiving the input and output I/O request that application layer issues;
Described acquiring unit, for for different service application, obtains the block size of business operation;
Described split cells, for according to the length of I/O data corresponding to I/O request and the block size of described business operation, splits described I/O request, makes each I/O request after fractionation correspond to an independent block.
Further, described split cells, according to the length of I/O data corresponding to I/O request and the block size of described business operation, splits I/O request, comprising:
If the length of the I/O data that current I/O request is corresponding in the length range that current block holds, does not then split current I/O request.
Further, described split cells, according to the length of I/O data corresponding to I/O request and the block size of described business operation, splits I/O request, comprising:
If the length of the I/O data that current I/O request is corresponding exceeds length that current block holds and the length of I/O data that indicates of at least one page of vector parameter bio_vec that current block can hold current I/O request, then in units of the length of the I/O data of bio_vec instruction, using the I/O data that the I/O data that current block holds is asked as current I O, the I/O data do not held by current block is as the I/O data of next I/O request.
Further, described split cells, according to the length of I/O data corresponding to I/O request and the block size of described business operation, splits I/O request, comprising:
If the length of the I/O data that current I/O request is corresponding exceeds length that current block holds and current block only can hold the I/O data of the partial-length of the first bio_vec instruction of current I/O request, the I/O data that the I/O data of the partial-length of the first bio_vec instruction then held by current block is asked as current I O, the I/O data of the residue length of the first bio_vec instruction do not held by current block is as the I/O data of next I/O request, the I/O data that other bio_vec of the current I/O request do not held by current block indicate is as the I/O data of next I/O request again.
Further, described split cells, according to the length of I/O data corresponding to I/O request and the block size of described business operation, splits I/O request, comprising:
If the length of the I/O data that current I/O request is corresponding exceeds length that current block holds and current block only can hold the I/O data of the partial-length of the first bio_vec instruction of current I/O request, the I/O data that the I/O data of the partial-length of the first bio_vec instruction then held by current block ask as current I O, the I/O data of other bio_vec instruction of the current I/O request that the I/O data of the residue length of the first bio_vec instruction do not held by current block and current block do not hold is as the I/O data of next I/O request.
Although the embodiment disclosed by the present invention is as above, the embodiment that described content only adopts for ease of understanding the present invention, and be not used to limit the present invention.Those of skill in the art belonging to any the present invention; under the prerequisite not departing from the spirit and scope disclosed by the present invention; any amendment and change can be carried out in the form implemented and details; but scope of patent protection of the present invention, the scope that still must define with appending claims is as the criterion.
Claims (10)
1. to the method that input output request splits, it is characterized in that, comprising:
Receive the input and output I/O request that application layer issues;
For different service application, obtain the block size of business operation;
The length of I/O data corresponding according to I/O request and the block size of described business operation, split described I/O request, makes each I/O request after fractionation correspond to an independent block.
2. method according to claim 1, is characterized in that, the length of the described I/O data corresponding according to I/O request and the block size of described business operation, split I/O request, comprising:
If the length of the I/O data that current I/O request is corresponding in the length range that current block holds, does not then split current I/O request.
3. method according to claim 1, is characterized in that, the length of the described I/O data corresponding according to I/O request and the block size of described business operation, split I/O request, comprising:
If the length of the I/O data that current I/O request is corresponding exceeds length that current block holds and the length of I/O data that indicates of at least one page of vector parameter bio_vec that current block can hold current I/O request, then in units of the length of the I/O data of bio_vec instruction, using the I/O data that the I/O data that current block holds is asked as current I O, the I/O data do not held by current block is as the I/O data of next I/O request.
4. method according to claim 1, is characterized in that, the length of the described I/O data corresponding according to I/O request and the block size of described business operation, split I/O request, comprising:
If the length of the I/O data that current I/O request is corresponding exceeds length that current block holds and current block only can hold the I/O data of the partial-length of the first bio_vec instruction of current I/O request, the I/O data that the I/O data of the partial-length of the first bio_vec instruction then held by current block is asked as current I O, the I/O data of the residue length of the first bio_vec instruction do not held by current block is as the I/O data of next I/O request, the I/O data that other bio_vec of the current I/O request do not held by current block indicate is as the I/O data of next I/O request again.
5. method according to claim 1, is characterized in that, the length of the described I/O data corresponding according to I/O request and the block size of described business operation, split I/O request, comprising:
If the length of the I/O data that current I/O request is corresponding exceeds length that current block holds and current block only can hold the I/O data of the partial-length of the first bio_vec instruction of current I/O request, the I/O data that the I/O data of the partial-length of the first bio_vec instruction then held by current block ask as current I O, the I/O data of other bio_vec instruction of the current I/O request that the I/O data of the residue length of the first bio_vec instruction do not held by current block and current block do not hold is as the I/O data of next I/O request.
6. to the device that input output request splits, it is characterized in that, comprising: receiving element, acquiring unit and split cells, wherein:
Described receiving element, for receiving the input and output I/O request that application layer issues;
Described acquiring unit, for for different service application, obtains the block size of business operation;
Described split cells, for according to the length of I/O data corresponding to I/O request and the block size of described business operation, splits described I/O request, makes each I/O request after fractionation correspond to an independent block.
7. device according to claim 6, is characterized in that, described split cells, according to the length of I/O data corresponding to I/O request and the block size of described business operation, splits I/O request, comprising:
If the length of the I/O data that current I/O request is corresponding in the length range that current block holds, does not then split current I/O request.
8. device according to claim 6, is characterized in that, described split cells, according to the length of I/O data corresponding to I/O request and the block size of described business operation, splits I/O request, comprising:
If the length of the I/O data that current I/O request is corresponding exceeds length that current block holds and the length of I/O data that indicates of at least one page of vector parameter bio_vec that current block can hold current I/O request, then in units of the length of the I/O data of bio_vec instruction, using the I/O data that the I/O data that current block holds is asked as current I O, the I/O data do not held by current block is as the I/O data of next I/O request.
9. device according to claim 6, is characterized in that, described split cells, according to the length of I/O data corresponding to I/O request and the block size of described business operation, splits I/O request, comprising:
If the length of the I/O data that current I/O request is corresponding exceeds length that current block holds and current block only can hold the I/O data of the partial-length of the first bio_vec instruction of current I/O request, the I/O data that the I/O data of the partial-length of the first bio_vec instruction then held by current block is asked as current I O, the I/O data of the residue length of the first bio_vec instruction do not held by current block is as the I/O data of next I/O request, the I/O data that other bio_vec of the current I/O request do not held by current block indicate is as the I/O data of next I/O request again.
10. device according to claim 6, is characterized in that, described split cells, according to the length of I/O data corresponding to I/O request and the block size of described business operation, splits I/O request, comprising:
If the length of the I/O data that current I/O request is corresponding exceeds length that current block holds and current block only can hold the I/O data of the partial-length of the first bio_vec instruction of current I/O request, the I/O data that the I/O data of the partial-length of the first bio_vec instruction then held by current block ask as current I O, the I/O data of other bio_vec instruction of the current I/O request that the I/O data of the residue length of the first bio_vec instruction do not held by current block and current block do not hold is as the I/O data of next I/O request.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510524313.1A CN105117282B (en) | 2015-08-24 | 2015-08-24 | A kind of method and device that input output request is split |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510524313.1A CN105117282B (en) | 2015-08-24 | 2015-08-24 | A kind of method and device that input output request is split |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN105117282A true CN105117282A (en) | 2015-12-02 |
| CN105117282B CN105117282B (en) | 2018-09-14 |
Family
ID=54665283
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201510524313.1A Active CN105117282B (en) | 2015-08-24 | 2015-08-24 | A kind of method and device that input output request is split |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN105117282B (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107562651A (en) * | 2017-07-20 | 2018-01-09 | 郑州云海信息技术有限公司 | A kind of data cache method and Data Buffer Memory |
| CN110377536A (en) * | 2019-07-12 | 2019-10-25 | 深信服科技股份有限公司 | Data processing method, device, equipment and computer readable storage medium |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20080104590A1 (en) * | 2006-11-01 | 2008-05-01 | Mccrory Dave Dennis | Adaptive, Scalable I/O Request Handling Architecture in Virtualized Computer Systems and Networks |
| CN103425439A (en) * | 2013-07-16 | 2013-12-04 | 记忆科技(深圳)有限公司 | Method for reading and writing solid-state disk and solid-state disk thereof |
| CN103617195A (en) * | 2013-11-05 | 2014-03-05 | 杭州沃趣网络科技有限公司 | Data synchronization system and data synchronization method used for MySQL database |
-
2015
- 2015-08-24 CN CN201510524313.1A patent/CN105117282B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20080104590A1 (en) * | 2006-11-01 | 2008-05-01 | Mccrory Dave Dennis | Adaptive, Scalable I/O Request Handling Architecture in Virtualized Computer Systems and Networks |
| CN103425439A (en) * | 2013-07-16 | 2013-12-04 | 记忆科技(深圳)有限公司 | Method for reading and writing solid-state disk and solid-state disk thereof |
| CN103617195A (en) * | 2013-11-05 | 2014-03-05 | 杭州沃趣网络科技有限公司 | Data synchronization system and data synchronization method used for MySQL database |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107562651A (en) * | 2017-07-20 | 2018-01-09 | 郑州云海信息技术有限公司 | A kind of data cache method and Data Buffer Memory |
| CN110377536A (en) * | 2019-07-12 | 2019-10-25 | 深信服科技股份有限公司 | Data processing method, device, equipment and computer readable storage medium |
| CN110377536B (en) * | 2019-07-12 | 2023-05-12 | 深信服科技股份有限公司 | Data processing method, device, equipment and computer readable storage medium |
Also Published As
| Publication number | Publication date |
|---|---|
| CN105117282B (en) | 2018-09-14 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US11983075B2 (en) | Migrating data and metadata from a backup system | |
| US20220019577A1 (en) | Preview generation operations for files in a network-accessible system | |
| US11321188B2 (en) | Platform-agnostic containerized application data protection | |
| US11429305B2 (en) | Performing backup operations using replicas | |
| US20190188196A1 (en) | Organically managing primary and secondary storage of a data object based on an expiry timeframe supplied by a user of the data object | |
| AU2012347883B2 (en) | System and method for restoring application data | |
| US8856080B2 (en) | Backup using metadata virtual hard drive and differential virtual hard drive | |
| US9348827B1 (en) | File-based snapshots for block-based backups | |
| US12056384B2 (en) | Skipping data backed up in prior backup operations | |
| US11042503B1 (en) | Continuous data protection and restoration | |
| US20180285207A1 (en) | Retrieval operations for files in a network-accessible system | |
| CN112800019A (en) | Data backup method and system based on Hadoop distributed file system | |
| US20160092131A1 (en) | Storage system, storage system control method, and recording medium storing virtual tape device control program | |
| US9251020B1 (en) | Systems and methods for file-level replication | |
| CN103176867A (en) | Fast file differential backup method | |
| US20150026143A1 (en) | Data handling | |
| US8595271B1 (en) | Systems and methods for performing file system checks | |
| CN105117282A (en) | Input and output request splitting method and device | |
| CN103207916A (en) | Metadata processing method and device | |
| US20240338281A1 (en) | Enforcing a recovery point objective (rpo) for a production database without generating secondary copies of the production database | |
| US10324807B1 (en) | Fast native file system creation for backup files on deduplication systems | |
| US20180285430A1 (en) | Preview generation operations for files in a network-accessible system | |
| WO2018019310A1 (en) | Big data system data backup and recovery methods and devices, and computer storage medium | |
| US10976952B2 (en) | System and method for orchestrated application protection | |
| US11907079B2 (en) | System and method for redundant backup of datasets |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |