CN108628765A - Cache implementation methods and device in distributed storage of increasing income software Ceph - Google Patents
Cache implementation methods and device in distributed storage of increasing income software Ceph Download PDFInfo
- Publication number
- CN108628765A CN108628765A CN201810332956.XA CN201810332956A CN108628765A CN 108628765 A CN108628765 A CN 108628765A CN 201810332956 A CN201810332956 A CN 201810332956A CN 108628765 A CN108628765 A CN 108628765A
- Authority
- CN
- China
- Prior art keywords
- cache
- subregions
- osd
- data
- patterns
- 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
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F12/00—Accessing, addressing or allocating within memory systems or architectures
- G06F12/02—Addressing or allocation; Relocation
- G06F12/08—Addressing or allocation; Relocation in hierarchically structured memory systems, e.g. virtual memory systems
- G06F12/0802—Addressing of a memory level in which the access to the desired data or data block requires associative addressing means, e.g. caches
- G06F12/0844—Multiple simultaneous or quasi-simultaneous cache accessing
- G06F12/0846—Cache with multiple tag or data arrays being simultaneously accessible
- G06F12/0848—Partitioned cache, e.g. separate instruction and operand caches
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
This application provides Cache implementation methods and devices in the distributed storage software Ceph that increases income.In the application, by creating Cache in the Cache subregions according to Cache patterns, Cache can be realized in OSD；Further, in the application, based on the excellent specific property for reading and writing flow in Cache, then the application is by that in OSD realization Cache, can improve read-write efficiency；Further, in the application, no matter whether Cache patterns support dirty data, all can realize Cache in OSD, and non-artificial realization, can improve deployment efficiency, can also simplify maintenance.
Description
Technical field
This application involves memory technologies, more particularly to distributed storage of increasing income software (Ceph) high speed buffer memory
(Cache) implementation method and device.
Background technology
Ceph can safeguard portable operating system interface (POSIX as distributed file system：Portable
Operating System Interface of UNIX) compatibility while, also increase replicate and fault tolerance.
Ceph, file system as Linux are one of alternative, have been integrated in main line linux kernel, subsequently can be used for
Solve the demand of mass memory.
Currently, in the storage architecture (abbreviation Ceph storage architectures) of Ceph applications, and Cache is not present.
Invention content
This application provides Cache implementation methods and devices in distributed storage software (Ceph) of increasing income, to be deposited in Ceph
Cache is realized in storage framework.
Technical solution provided by the present application includes：
One kind is increased income distributed storage software Ceph high speed buffer memory Cache implementation methods, and this method is applied to
Back end OSD is stored, when this OSD is activated, this method includes：
Obtain the Cache patterns of generated Cache subregions in this OSD；
If being determined according to the Cache patterns and creating Cache in the Cache subregions, in the Cache subregions
Cache is created, and according to the file type for the data information that generated data Data subregions are written in this OSD by the number
It is believed that the Cache is written in breath；
By the data information carry being written in the Cache to the corresponding OSD ID catalogues of this OSD, according to the OSD ID
Catalogue starts OSD processes.
One kind is increased income distributed storage software Ceph high speed buffer memory Cache realization devices, which is applied to
Back end OSD is stored, including：
Acquiring unit, the Cache patterns for obtaining generated Cache subregions in the OSD；
Determination unit determines whether to create Cache in the Cache subregions according to the Cache patterns；
Cache units are created for being determined according to the Cache patterns in the determination unit in the Cache subregions
When building Cache, Cache is created in the Cache subregions, and be written according to generated data Data subregions in this OSD
The file type of data information the Cache is written into the data information；
Processing unit, data information carry for will be written in the Cache to the corresponding OSD ID catalogues of this OSD,
Start OSD processes according to the OSD ID catalogues.
As can be seen from the above technical solutions, in the application, by being created in the Cache subregions according to Cache patterns
Cache is built, can realize Cache in OSD；
Further, in the application, based on the excellent specific property for reading and writing flow in Cache, then the application in OSD by realizing
Cache can improve read-write efficiency；
Further, in the application, no matter whether Cache patterns support dirty data, all can realize Cache in OSD, and
Non-artificial realization can improve deployment efficiency, can also simplify maintenance.
Description of the drawings
The drawings herein are incorporated into the specification and forms part of this specification, and shows the implementation for meeting the disclosure
Example, and together with specification for explaining the principles of this disclosure.
Fig. 1 is the Prepare phase flow figures of OSD；
Fig. 2 is the service logic structure chart of OSD；
Fig. 3 is the Activate phase flow figures of OSD；
Fig. 4 is the exemplary embodiment flow chart that Cache is realized in Ceph provided by the present application；
Fig. 5 is the exemplary embodiment flow chart provided by the present application that Cache is created in Cache subregions；
Fig. 6 is the service logic structure chart of OSD provided by the present application；
Fig. 7 is the Prepare phase flow figures of OSD provided by the present application；
Fig. 8 is apparatus structure schematic diagram provided by the present application.
Specific implementation mode
Before describing method provided by the present application, first the service logic of OSD in existing Ceph storage architectures is retouched
It states：
The service logic of OSD is mainly in two stages in Ceph storage architectures：(Prepare) stage of deployment, startup
(Activate) stage.
In the Prepare stages, as shown in Figure 1, including mainly following operation：
Step 101, daily record (Journal) partition size is obtained.
In the application, OSD can be directed to and configure a file (being known as OSD configuration files).OSD configuration files are stored in specified
Under catalogue.As one embodiment, specified directory here can be under the subdirectory such as Ceph.cof catalogues of Ceph catalogues.
In one example, OSD configuration files can record Journal partition sizes.Here OSD configuration files record
Journal partition sizes are configured according to business demand, for example can be 10G, 20G, 30G etc., and the application does not limit specifically.
Based on this, in step 101, obtaining Journal partition sizes can be：Journal points are read from OSD configuration files
Area's size.
Step 102, Journal subregions are generated in OSD according to the Journal partition sizes obtained.
Step 103, it is that corresponding Journal Universally Unique Identifiers (UUID is arranged in Journal subregions：Universally
Unique Identifier), and generate the link of Journal subregions.
In one example, Journal UUID can be determined according to the preset algorithm for calculating UUID.In different OSD
The corresponding Journal UUID of Journal subregions are different.
In one example, the link of Journal subregions is primarily to facilitate quick access Journal subregions.The application answers
For linux system, the link in linux system is somewhat similarly to the shortcut of Windo ws.Link in linux system
There are two types of：One is hard links；Another kind is Symbolic Links (also known as soft link).As one embodiment, this step 103 generates
The links of Journal subregions can be soft link.
Step 104, data (Data) subregion is generated in OSD.
As one embodiment, this step 104 may include：Using the region in OSD in addition to Journal subregions as Data
Subregion.
Journal subregions, Data subregions in OSD has been shown in particular in Fig. 2.
Step 105, it is that corresponding Data UUID are arranged in Data subregions, and generate the link of Data subregions.
In one example, Data UUID can be determined according to the preset algorithm for calculating UUID.In different OSD
The corresponding Data UUID of Data subregions are different.
In one example, the link of Data subregions is primarily to facilitate quick access Data subregions.As one embodiment,
The link for the Data subregions that this step 105 generates and Journal subregions are linked similar or soft are linked.
Step 106, Data subregions are formatted, OSD related data informations are written in Data subregions.
In one example, OSD related data informations may include：The corresponding Journal of Journal subregions in OSD
Corresponding cluster UUID of cluster where corresponding OSD UUID of UUID, OSD, OSD etc..
After Data subregions are written in data information, the file type of data information can be recorded to OSD configuration files.
On schematically illustrate the flow in Prepare stages.
It should be noted that in the Prepare stages, other than above-mentioned steps 101 are to step 106, other are will also include
Many steps only describe by taking step 101 to step 106 as an example, are not offered as covering the entire Prepare stages here.
In the Activate stages, as shown in Figure 3, it may include following steps：
Step 301, the file type for the data information that Data subregions are written is obtained.
As described above, after Data subregions are written in data information, the file type of data information can be recorded to OSD and is matched
Set file.Based on this, in this step 301, the file type for obtaining the data information that Data subregions are written may include：From OSD
The file type for the data information that Data subregions are written is read in configuration file.Here file type may include：Log type
File system (XFS).
Step 302, according to the file type that obtains by data information carry that Data subregions are written to having created
Data subregion temp directories.
Step 303, check whether the data information of Data subregion temp directory carries includes information of the OSD by certification,
If it is not, step 304 is executed, if so, executing step 305.
Step 304, apply for the corresponding OSD marks (ID) of this OSD and key (Key), recognized by authenticating device in the Key
When card, 305 are thened follow the steps.
In one example, here, OSD is that monitor (Monitor) application being equipped with into Ceph storage architectures corresponds to
OSD ID and Key be responsible for OSD when Monitor receives application and distribute corresponding Key and OSD ID.
As one embodiment, after OSD applies for this OSD corresponding OSD ID and Key, OSD is based on the OSD of application
ID and Key initiates certification to authenticating device, then can be by OSD by recognizing when certifications of the Key of OSD applications by authenticating device
The information of card is recorded to Data subregions, executes step 304 later.In one example, authenticating device here is alternatively above-mentioned
Monitor.
Step 305, cancel the carry relationship between the data information that Data subregions temp directory and Data subregions are written,
The data information carry that Data subregions are written is to the corresponding OSD ID catalogues of this OSD.
Here, why cancel carry relationship, in order to prevent invalid data information from interfering the management of catalogue.
Step 306, start OSD processes according to OSD ID catalogues.
So far, OSD processes just start to start.
The flow in Activate stages described briefly above.
It should be noted that in the Activate stages, other than above-mentioned steps 301 are to step 306, other are will also include
Many steps only describe by taking step 301 to step 306 as an example, are not offered as covering the entire Activate stages here.
The Activate ranks of the OSD described from flow shown in the Prepare stages of the OSD of the description of flow shown in Fig. 2 and Fig. 3
Section is not as can be seen that OSD disposes Cache, this also means that Cache is not implemented in Ceph storage architectures.
And if realizing Cache in Ceph storage architectures, it is often necessary to it is more complicated in OSD deployment Cache manually, and
And it is less susceptible to safeguard.
In order to which automatic deployment realizes Cache in Ceph storage architectures, this application provides flows as shown in Figure 4：
Referring to Fig. 4, Fig. 4 is the exemplary embodiment flow chart that Cache is realized in Ceph provided by the present application.This method is answered
For OSD, when OSD is activated, as shown in figure 4, this approach includes the following steps：
Step 401, the Cache patterns of generated Cache subregions in this OSD are obtained.
Here, generated Cache subregions are described below in this OSD, wouldn't describe here.
Step 402, Cache is created in Cache subregions if being determined according to the Cache patterns, at the Cache points
Create Cache in area, and according to the file type for the data information that generated data Data subregions are written in this OSD by institute
State data information write-in Cache.
In one example, determine that Cache is created in Cache subregions may include in step 402 according to Cache patterns：
Judge whether the Cache patterns are to be used to indicate the first specified Cache patterns for supporting dirty data, if so, determining in institute
It states in Cache subregions and creates Cache.
Here the first specified Cache patterns can support load, create function, specified as one embodiment, first
Cache patterns can be write-back (Write-back) pattern in Cache mechanism.
Be the first specified Cache patterns such as Write-back patterns based on Cache patterns, then in step 402,
Cache is created in Cache subregions to may include：By loading Cache partition creatings Cache.
In another example, determine that Cache is created in Cache subregions to be wrapped in step 402 according to Cache patterns
It includes：Judge whether the Cache patterns are to be used to indicate the second specified Cache patterns for not supporting dirty data；If so, determining
Cache is created in the Cache subregions.
Here the second specified Cache patterns do not support load function, as one embodiment, the second specified Cache moulds
Formula can be direct write (Write-through) pattern in Cache mechanism or be that data are directly write around solid state disk (SSD)
To common hard disc (Write-around) pattern etc..
It is the second specified Cache patterns such as Write-through patterns or Write-around moulds based on Cache patterns
Formula etc., in step 402, establishment Cache may include flow shown in Fig. 5 in Cache subregions：
As shown in figure 5, the flow may include：
Step 501, the file type for the data information being written according to the Data subregions is by the data information carry
To the Data subregion temp directories created.
As described above, after Data subregions are written in data information, the file type of data information can be recorded to OSD and is matched
Set file.Based on this, in this step 501, the file type for obtaining the data information that Data subregions are written may include：From OSD
The file type for the data information that Data subregions are written is read in configuration file.Here file type may include：Log type
File system (XFS).
Step 502, check whether the data information of the Data subregions temp directory carry including this OSD passes through certification
Information；If so, step 503 is executed, if it is not, executing step 504.
Step 503, Cache is created in Cache subregions.
Step 504, apply for the corresponding OSD ID and Key of this OSD, when the Key is by authenticating device certification,
Cache is created in Cache subregions.
Here, OSD is that the monitor (Monitor) being equipped with into Ceph storage architectures applies for corresponding OSD ID and Key,
When Monitor receives application, it is responsible for OSD and distributes corresponding Key and OSD ID.
As one embodiment, after OSD applies for this OSD corresponding OSD ID and Key, OSD is based on the OSD of application
ID and Key initiates certification to authenticating device, then can be by OSD by recognizing when certifications of the Key of OSD applications by authenticating device
The information of card is recorded to Data subregions.In one example, authenticating device here is alternatively above-mentioned Monitor.
So far, it is realized by flow shown in Fig. 5 and creates Cache in Cache subregions.Based on flow shown in Fig. 5, then step
In 402, the data information is written the file type according to the data information that generated Data subregions are written in this OSD
The Cache includes：Cancel the carry relationship between the data information that Data subregions temp directory and Data subregions are written, according to
It will be described in data information write-in according to the file type for the data information that generated data Data subregions are written in this OSD
Cache。
It should be noted that when executing this step 402, it is also possible to not generate Cache subregions in OSD, or
The pattern of Cache subregions is default value such as sky or designated value in OSD, can be executed at this time according to flow shown in above-mentioned Fig. 3, this
In repeat no more,
Step 403, by the data information carry being written in Cache to the corresponding OSD ID catalogues of this OSD, according to described in
OSD ID catalogues start OSD processes.
In one example, before the data information carry that will be written in Cache to OSD ID catalogues, Data can be cancelled
Carry relationship between the data information being written in subregion temp directory and Cache, so that data information no longer carry in Cache
To Data subregion temp directories, prevent invalid data information from interfering the management of catalogue.
So far, flow shown in Fig. 4 is completed.
Cache can be realized by flow shown in Fig. 4 in OSD.Based on the excellent specific property for reading and writing flow in Cache, then originally
Application can improve read-write efficiency by realizing Cache in OSD；
Further, in the application, no matter whether Cache patterns support dirty data, all can realize Cache in OSD, rather than
It is artificial to realize, deployment efficiency can be improved, maintenance can be also simplified.
In one example, flow shown in Fig. 4 is to realize in the Prepare stages based on following OSD.Here, OSD
Improvement of the Prepare stages compared to the Prepare stages shown in Fig. 3 include mainly：The realization of Cache subregions is increased in OSD.
Service logic structure compared to existing OSD increases Cache subregions in the service logic structure of OSD provided by the present application.Figure
6 show the service logic structure of OSD provided by the present application.
Based on improvement described above, the Prepare ranks of OSD provided by the present application are described below by flow shown in Fig. 7
Section：
As shown in fig. 7, including mainly following operation：
Step 701, Journal partition sizes, Cache partition sizes are obtained.
In the application, OSD can be directed to and configures an OSD configuration file.OSD configuration files are stored under specified directory.Make
For one embodiment, specified directory here can be under the subdirectory such as Ceph.cof catalogues of Ceph catalogues.
In one example, OSD configuration files can record Journal partition sizes, Cache partition sizes.Based on this, walk
In rapid 701, Journal partition sizes are obtained, Cache partition sizes may include：Journal points are read from OSD configuration files
Area's size, Cache partition sizes.
Step 702, Journal subregions are generated in OSD according to the Journal partition sizes obtained, according to the Cache obtained
Partition size generates Cache subregions in OSD.
Step 703, it is that Journal UUID are arranged in Journal subregions, and generate the link of Journal subregions, is Cache
Cache UUID and Cache patterns are arranged in subregion, and generate the link of Cache subregions.
In one example, Cache UUID being arranged for Cache subregions, Cache mode records can be configured text in OSD
In part.
Step 704, Data subregions are generated in OSD.
As one embodiment, this step 704 may include：By the area in OSD in addition to Journal subregions, Cache subregions
Domain is as Data subregions.So far, an OSD includes at least Journal subregions, Cache subregions, Data subregions, specifically such as Fig. 6 institutes
Show.
Step 705, it is that corresponding Data UUID are arranged in Data subregions, and generate the link of Data subregions.
Step 706, Data subregions are formatted, OSD related data informations are written in Data subregions.
In one example, OSD related data informations may include：The corresponding Journal of Journal subregions in OSD
The corresponding cluster UUID of cluster, Cache UUID, Cache patterns where corresponding OSD UUID of UUID, OSD, OSD etc..
After Data subregions are written in data information, the file type of data information can be recorded to OSD configuration files.
On schematically illustrate OSD provided by the present application the Prepare stages flow.
It should be noted that in the Prepare stages, other than above-mentioned steps 701 are to step 706, other are will also include
Many steps only describe by taking step 701 to step 706 as an example, are not offered as covering the entire Prepare stages here.
It should be noted that in the application, the Cache subregions generated in OSD not immobilize, can be according to business need
It asks and changes.
When generated Cache subregions are changed in OSD, can further perform the step of：
First Cache UUID of Data subregions the are written and carry in the data information of the OSD ID catalogues is repaiied
The corresponding 2nd Cache UUID of Cache subregions being changed to after change；The first Cache UUID and Cache points before change
Area corresponds to；Return to flow shown in Fig. 4.
Method provided by the present application is described above.Device provided by the present application is described below：
Referring to Fig. 8, Fig. 8 is device exemplary embodiment structure chart provided by the present application.The device is applied to OSD, including：
Acquiring unit, the Cache patterns for obtaining generated Cache subregions in the OSD；
Determination unit determines whether to create Cache in the Cache subregions according to the Cache patterns；
Cache units are created for being determined according to the Cache patterns in the determination unit in the Cache subregions
When building Cache, Cache is created in the Cache subregions, and be written according to generated data Data subregions in this OSD
The file type of data information the Cache is written into the data information；
Processing unit, data information carry for will be written in the Cache to the corresponding OSD ID catalogues of this OSD,
Start OSD processes according to the OSD ID catalogues.
In one example, the Cache units create Cache according to the determination of Cache patterns in the Cache subregions
Including：Judge whether the Cache patterns are to be used to indicate the first specified Cache patterns for supporting dirty data；If so, determining
Cache is created in the Cache subregions；
Based on this, the Cache units create Cache in Cache subregions and include：By loading the Cache subregions
Create Cache.
In another example, the Cache units are determined according to the Cache patterns and are created in the Cache subregions
Building Cache includes：Judge whether the Cache patterns are to be used to indicate the second specified Cache patterns for not supporting dirty data；If
It is that determination creates Cache in the Cache subregions；
Based on this, the Cache that created in Cache subregions includes：The data letter being written according to the Data subregions
The file type of breath is by the data information carry to the Data subregion temp directories created；Check that the Data subregions are interim
Whether the data information of catalogue carry includes information of this OSD by certification；If so, creating Cache in Cache subregions；If
It is no, apply for the corresponding OSD marks ID and key Key of this OSD, when the Key is by authenticating device certification, in Cache subregions
Middle establishment Cache.
As one embodiment, the processing unit further change in the OSD by generated Cache subregions
When, by Data subregions write-in and the first Cache UUID that carry is in the data information of the OSD ID catalogues repair
The corresponding 2nd Cache UUID of Cache subregions being changed to after change；The first Cache UUID and Cache points before change
Area corresponds to, and triggers the operation that the acquiring unit executes.
So far, the structure description of Fig. 8 shown devices is completed.
The foregoing is merely the preferred embodiments of the application, not limiting the application, all essences in the application
With within principle, any modification, equivalent substitution, improvement and etc. done should be included within the scope of the application protection god.
Claims (10)
- The distributed storage software Ceph high speeds buffer memory Cache implementation methods 1. one kind is increased income, which is characterized in that the party Method is applied to storage back end OSD, and when this OSD is activated, this method includes：Obtain the Cache patterns of generated Cache subregions in this OSD；If being determined according to the Cache patterns and creating Cache in the Cache subregions, created in the Cache subregions Cache, and the file type according to the data information that generated data Data subregions are written in this OSD believes the data The Cache is written in breath；By the data information carry being written in the Cache to the corresponding OSD ID catalogues of this OSD, according to the OSD ID catalogues Start OSD processes.
- 2. according to the method described in claim 1, it is characterized in that, described determine according to Cache patterns in the Cache subregions Middle establishment Cache includes：Judge whether the Cache patterns are to be used to indicate the first specified Cache patterns for supporting dirty data；If so, determination creates Cache in the Cache subregions.
- 3. according to the method described in claim 2, it is characterized in that, the Cache that created in Cache subregions includes：By loading the Cache partition creatings Cache.
- 4. according to the method described in claim 1, it is characterized in that, described determine according to the Cache patterns in the Cache Cache is created in subregion includes：Judge whether the Cache patterns are to be used to indicate the second specified Cache patterns for not supporting dirty data；If so, determination creates Cache in the Cache subregions.
- 5. according to the method described in claim 4, it is characterized in that, the Cache that created in Cache subregions includes：The file type of the data information being written according to the Data subregions is by the data information carry to having created Data subregion temp directories；Check whether the data information of the Data subregions temp directory carry includes information of this OSD by certification；If so, Cache is created in Cache subregions；If it is not, applying for the corresponding OSD marks ID and key Key of this OSD, pass through certification in the Key When device authentication, Cache is created in Cache subregions.
- 6. according to the method described in claim 1, being changed it is characterized in that, working as generated Cache subregions in this OSD When, this method further comprises：By Data subregions write-in and the first Cache that carry is in the data information of the OSD ID catalogues it is general only One identification code UUID is revised as the corresponding 2nd Cache UUID of Cache subregions after change；The first Cache UUID with Cache subregions before change correspond to；Return to the operation for activating the OSD.
- The distributed storage software Ceph high speeds buffer memory Cache realization devices 7. one kind is increased income, which is characterized in that the dress It sets applied to storage back end OSD, including：Acquiring unit, the Cache patterns for obtaining generated Cache subregions in the OSD；Determination unit determines whether to create Cache in the Cache subregions according to the Cache patterns；Cache units are created for being determined according to the Cache patterns in the determination unit in the Cache subregions When Cache, Cache is created in the Cache subregions, and be written according to generated data Data subregions in this OSD The Cache is written in the data information by the file type of data information；Processing unit, data information carry for will be written in the Cache to the corresponding OSD ID catalogues of this OSD, foundation The OSD ID catalogues start OSD processes.
- 8. device according to claim 7, which is characterized in that the Cache units are determined according to Cache patterns described Cache is created in Cache subregions includes：Judge whether the Cache patterns are to be used to indicate to support the first of dirty data to specify Cache patterns；If so, determination creates Cache in the Cache subregions；The Cache units create Cache in Cache subregions：By loading the Cache partition creatings Cache.
- 9. device according to claim 8, which is characterized in that the Cache units are determined according to the Cache patterns Cache is created in the Cache subregions includes：Judge whether the Cache patterns are to be used to indicate do not support dirty data Two specified Cache patterns；If so, determination creates Cache in the Cache subregions；It is described in Cache subregions create Cache include：File type according to the data information that the Data subregions are written By the data information carry to the Data subregion temp directories created；Check the number of the Data subregions temp directory carry It is believed that whether breath includes information of this OSD by certification；If so, creating Cache in Cache subregions；If it is not, applying for this OSD Corresponding OSD marks ID and key Key creates Cache when the Key is by authenticating device certification in Cache subregions.
- 10. device according to claim 7, which is characterized in that the processing unit has further generated in the OSD Cache subregions when changing, by Data subregions write-in and carry in the data information of the OSD ID catalogues The first Cache Universally Unique Identifiers UUID be revised as change after the corresponding 2nd Cache UUID of Cache subregions；It is described First Cache UUID are corresponding with the Cache subregions before change, and trigger the operation that the acquiring unit executes.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810332956.XA CN108628765B (en) | 2018-04-13 | 2018-04-13 | Method and device for realizing Cache in open-source distributed storage software Ceph |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810332956.XA CN108628765B (en) | 2018-04-13 | 2018-04-13 | Method and device for realizing Cache in open-source distributed storage software Ceph |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN108628765A true CN108628765A (en) | 2018-10-09 |
| CN108628765B CN108628765B (en) | 2021-03-23 |
Family
ID=63705392
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810332956.XA Active CN108628765B (en) | 2018-04-13 | 2018-04-13 | Method and device for realizing Cache in open-source distributed storage software Ceph |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108628765B (en) |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109408429A (en) * | 2018-11-01 | 2019-03-01 | 郑州云海信息技术有限公司 | A kind of caching method and device of low-speed interface |
| CN109799954A (en) * | 2018-12-19 | 2019-05-24 | 创新科存储技术（深圳）有限公司 | Realize the method and OSD of CEPH cluster cache |
| CN109814946A (en) * | 2018-12-29 | 2019-05-28 | 创新科存储技术（深圳）有限公司 | A kind of method and apparatus for realizing distributed storage in XenServer |
| CN111026681A (en) * | 2019-10-30 | 2020-04-17 | 烽火通信科技股份有限公司 | Caching method, caching system and caching medium based on Ceph |
| CN115314718A (en) * | 2021-05-07 | 2022-11-08 | 北京字节跳动网络技术有限公司 | Live broadcast data processing method, device, equipment and medium |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106649141A (en) * | 2016-11-02 | 2017-05-10 | 郑州云海信息技术有限公司 | Storage interaction device and storage system based on ceph |
| CN107046563A (en) * | 2017-01-19 | 2017-08-15 | 无锡华云数据技术服务有限公司 | A kind of implementation method, system and the cloud platform of distribution type high efficient cloud disk |
| US20170251073A1 (en) * | 2016-02-29 | 2017-08-31 | Intel Corporation | Technologies for managing replica caching in a distributed storage system |
| CN107861693A (en) * | 2017-11-27 | 2018-03-30 | 郑州云海信息技术有限公司 | The OSD procedure deployments method, apparatus and medium of a kind of distributed block storage system |
-
2018
- 2018-04-13 CN CN201810332956.XA patent/CN108628765B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20170251073A1 (en) * | 2016-02-29 | 2017-08-31 | Intel Corporation | Technologies for managing replica caching in a distributed storage system |
| CN106649141A (en) * | 2016-11-02 | 2017-05-10 | 郑州云海信息技术有限公司 | Storage interaction device and storage system based on ceph |
| CN107046563A (en) * | 2017-01-19 | 2017-08-15 | 无锡华云数据技术服务有限公司 | A kind of implementation method, system and the cloud platform of distribution type high efficient cloud disk |
| CN107861693A (en) * | 2017-11-27 | 2018-03-30 | 郑州云海信息技术有限公司 | The OSD procedure deployments method, apparatus and medium of a kind of distributed block storage system |
Non-Patent Citations (6)
| Title |
|---|
| HELLO_WORLDW6: ""ceph的数据存储之路(12)----cache tier"", 《HTTP://WWW.360DOC.COM/CONTENT/17/0814/11/46248428_679071890.SHTML》 * |
| INTEL: ""Ceph Cache Tiering Introduction"", 《HTTPS://SOFTWARE.INTEL.COM/CONTENT/WWW/US/EN/DEVELOP/BLOGS/CEPH-CACHE-TIERING-INTRODUCTION.HTML》 * |
| 世民谈云计算: ""理解OpenStack+Ceph（1）：Ceph+OpenStack集群部署和配置"", 《HTTPS://WWW.CNBLOGS.COM/SAMMYLIU/P/4804037.HTML》 * |
| 世民谈云计算: ""理解QEMU/KVM和Ceph（1）：QEMU-KVM和CephRBD的缓存机制总结"", 《HTTPS://WWW.CNBLOGS.COM/SAMMYLIU/P/5066895.HTML》 * |
| 世民谈云计算: "《https://www.cnblogs.com/sammyliu/p/4836014.html》", 29 September 2015 * |
| 郭建楠: ""SSD在分布式文件系统中的应用场景及方案对比"", 《通信世界》 * |
Cited By (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109408429A (en) * | 2018-11-01 | 2019-03-01 | 郑州云海信息技术有限公司 | A kind of caching method and device of low-speed interface |
| CN109408429B (en) * | 2018-11-01 | 2020-10-16 | 苏州浪潮智能科技有限公司 | Caching method and device for low-speed interface |
| CN109799954A (en) * | 2018-12-19 | 2019-05-24 | 创新科存储技术（深圳）有限公司 | Realize the method and OSD of CEPH cluster cache |
| CN109799954B (en) * | 2018-12-19 | 2022-02-22 | 深圳创新科技术有限公司 | Method for realizing CEPH cluster cache and OSD |
| CN109814946A (en) * | 2018-12-29 | 2019-05-28 | 创新科存储技术（深圳）有限公司 | A kind of method and apparatus for realizing distributed storage in XenServer |
| CN109814946B (en) * | 2018-12-29 | 2022-03-25 | 深圳创新科技术有限公司 | Method and device for realizing distributed storage in XenServer |
| CN111026681A (en) * | 2019-10-30 | 2020-04-17 | 烽火通信科技股份有限公司 | Caching method, caching system and caching medium based on Ceph |
| CN115314718A (en) * | 2021-05-07 | 2022-11-08 | 北京字节跳动网络技术有限公司 | Live broadcast data processing method, device, equipment and medium |
| CN115314718B (en) * | 2021-05-07 | 2023-07-14 | 北京字节跳动网络技术有限公司 | Live broadcast data processing method, device, equipment and medium |
Also Published As
| Publication number | Publication date |
|---|---|
| CN108628765B (en) | 2021-03-23 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108628765A (en) | Cache implementation methods and device in distributed storage of increasing income software Ceph | |
| KR101643022B1 (en) | Catalog-based software component management | |
| US10067835B2 (en) | System reset | |
| JP5781736B2 (en) | Method, system and computer readable storage medium for integrating virtualization sublayers | |
| CN106598667A (en) | Method and device used for restoring kernel vulnerability | |
| CN107710215A (en) | The method and apparatus of mobile computing device safety in test facilities | |
| CN107329741A (en) | A kind of software distribution upgrade method and device based on fingerprint recognition | |
| CN102662741A (en) | Method, device and system for realizing virtual desktop | |
| CN107562483A (en) | A kind of method and device of carry guiding file system | |
| US20070159974A1 (en) | Method, sysem and computer program for backing-up data with centralized control | |
| KR101624005B1 (en) | Controlling access to software component state | |
| CN112162925A (en) | Test method and system based on service virtualization | |
| CN103685259B (en) | The method and its device of Account Logon | |
| CN105302501A (en) | Control method and device for disk sectors | |
| CN107479923A (en) | Application program updating method, apparatus and display terminal | |
| CN109684026A (en) | APP replacing skins method, apparatus, electronic equipment and storage medium | |
| CN101714092A (en) | Method for controlling boot sequence of server | |
| CN104750579B (en) | The method and apparatus for making memory mirror | |
| CN110837442B (en) | KVM virtual machine backup system based on dirty data bitmap and network block equipment | |
| JP3703708B2 (en) | How to use a shared computer | |
| CN108629197A (en) | File access control method and system for the integration environment | |
| JP6098392B2 (en) | Software correction patch extraction program, software correction patch extraction method, and information processing apparatus | |
| JP4156388B2 (en) | AP addition / AP personalization method, implementation apparatus thereof, and processing program thereof | |
| CN110096285A (en) | Dispositions method, device, computer equipment and the storage medium of software systems | |
| JP2003150321A (en) | Device, method and program for managing virtual storage device, and computer readable recording medium recorded with virtual storage device |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |