CN105095439A - Picture specified security storage cloud system - Google Patents
Picture specified security storage cloud system Download PDFInfo
- Publication number
- CN105095439A CN105095439A CN201510438198.6A CN201510438198A CN105095439A CN 105095439 A CN105095439 A CN 105095439A CN 201510438198 A CN201510438198 A CN 201510438198A CN 105095439 A CN105095439 A CN 105095439A
- Authority
- CN
- China
- Prior art keywords
- data
- metadata
- picture
- back end
- image data
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/10—File systems; File servers
- G06F16/18—File system types
- G06F16/182—Distributed file systems
- G06F16/1824—Distributed file systems implemented using Network-attached Storage [NAS] architecture
- G06F16/183—Provision of network file services by network file servers, e.g. by using NFS, CIFS
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/10—File systems; File servers
- G06F16/18—File system types
- G06F16/182—Distributed file systems
Abstract
The invention provides a picture specified security storage cloud system which is applied to the storage of picture data generated by a visual monitoring system and includes: two metadata nodes and a plurality of data nodes, wherein the metadata nodes are used to store very little metadata, achieve a distribution mapping between the picture data and the storage space of the data nodes through the meta data, manage and configure a property and real-time status of each data node, and provide an entry accessing the picture data. The data nodes are used to provide a storage space for achieving physical storage and access of the picture data, and feed back the picture data accessed by clients to the clients in the form of picture stream. The picture specified security storage cloud system adapts to the application scenario characteristics, and according to the actual demand of security maintenance of the picture data, an uniform picture specified security storage cloud system architecture and data access management model are built, so that the picture specified security storage cloud system has advantages of security and efficiency.
Description
Technical field
The present invention relates to the areas of information technology relevant to visual control, particularly relate to the exclusive safe storage cloud system of a kind of picture.
Background technology
At present, in the construction of the Visualized Monitoring Systems such as safe city, wisdom traffic, a large amount of image datas can be related to and store.Along with Visualized Monitoring System lays the growth of increasing of forward terminal position and picture-storage duration, the capacity that picture-storage takies sharply rises, 3,000 ten thousand can being reached as some districts and cities bayonet socket of day monitoring institute produces picture, by often opening 1MB calculating, within one month, just needing the storage space of nearly 1000TB.
For the storage of image data, conventional way, is store based on general file system by data, the server of system background directly carries out picture file storage.Some technology adopts ftp server or WEB server mode to carry out data access, and some is the direct access that the mode adopting NFS/CIFS document of agreement to share is implemented to server.Such mode is simply direct, but along with the very fast growth of picture capacity and the requirements at the higher level to picture-storage access performance, reliability, because of technically intrinsic bottleneck, this type of scheme is more and more unable to do what one wishes.
Specifically, the defect that above-mentioned prior art occurs at present comprises: first, and the universal document system adopted is after quantity of documents sharp increase, and access efficiency sharply declines; Meanwhile, along with the continuous additions and deletions of new ancient deed, disk fragments can get more and more, and causes access efficiency to worsen further.Secondly, if there is disorderly closedown situations such as main frame abnormal power-down, after start, the self-inspection duration reaching a few hours of file system also can be a disaster.And, in order to the bottleneck of the capacity of tackling and performance, Normal practice be manually delimit some forward terminal position source picture-storage on the server that certain is predetermined, but the problem brought like this to manage all storage spaces as a whole, cannot dynamic dispatching distribute, cannot to the residual memory space dynamic equalization of this server, cause the waste of storage space and cannot be superfluous standby, be difficult to dynamic elasticity dilatation.
Along with the development of cloud memory technology taking Hadoop as representative, the existing Visualized Monitoring System of part proposes the storage using cloud storage system as background support mass picture data, be intended to the technical advantage utilizing cloud to store, alleviate the pressure on storage space, improve the efficiency of data access application as far as possible, and ensure data security.
Cloud storage is the comparatively ripe important branch of current cloud (CLOUD), its core is unified storage space by dispersion and the memory device of isomery in a large number by network interconnection and integration, utilize general external reference interface, operational data storage is provided.
At present, cloud storage system is to the access of mass data and the main Hadoop distributed file system ripe based on the comparison of management.Whole cloud storage system is configured to back end, metadata node and client by Hadoop system.Wherein, a large amount of memory device with disperseing, as back end, take file as the actual storage that unit performs to data, the access application of supported data, and the information that file is relevant is regularly reported metadata node.Metadata node is host node, its function is that the information uploaded according to back end generates and preserves the metadata of described file on the one hand, metadata reflects the mapping of described file to the physical memory space, on the other hand, metadata node is responsible for the request of customer in response end and the operation performed file, wherein the most basic operation is the reading of file, also comprises the deletion of file, rename and other directory operation.Back end can cause the change of file-related information to the aforesaid operations of file, thus needs the metadata that on correspondingly changing metadata node, file involved to operation is relevant.In data security, Hadoop system adopts the mode of duplicate of the document to perform redundancy backup, a file generated is three copies and distributed store on different back end, other copy can be utilized immediately to be recovered when a duplicate of the document goes wrong.
But the mass picture file cloud storage system on Hadoop distributed file system basis being applied to visual control stores, and still there will be problem.First, the image data that visual control produces has the quantity of documents of magnanimity level and the file size of lightweight, such as mentioned above, it is the picture file of about 1M that the Visualized Monitoring System in city may produce 3,000 ten thousand sizes for one day, under the system of Hadoop distributed file system, need the picture file for magnanimity level to generate, preserve and the metadata of more each file of new management, to spreading unchecked and Meta data system complicated of metadata data scale be caused, give and bring difficulty based on the file access of metadata and operation.Secondly, the special application scenarios of visual control causes its incompatible with Hadoop system pattern, and the conventional cloud memory module that Hadoop system is suitable for is: file write-once, seldom deletion, repeatedly read; And picture file needs write-once, circulation deletion, repeatedly reads in the middle of visual control application.Both are the frequency that " deletion " operates at maximum difference, because under the scene of visual control, along with the continuous passing of time, the generation that picture file can be unlimited, therefore must circulate and delete rise time upper forward picture file, thus vacate storage space.But deletion action can bring the great expense incurred of metadata management frequently, this is also that Hadoop system is when processing problems " short slab ".Again, in order to augmentation data safety, the redundancy backup scheme that provides sets up three copies to Hadoop system, expends too much storage space, is difficult to actually to adopt under the data magnitude of visual control.
Summary of the invention
In order to overcome the defect that above-mentioned prior art exists, the invention provides the exclusive safe storage cloud system of a kind of picture.Native system is for magnanimity level image data scale in visual control, adapt to the application scenarios feature of " write-once, circulation are deleted, repeatedly read " that it has, and according to safeguarding the actual demand of image data safety, construct the exclusive cloud storage system framework of unified picture and data access management model, the image data cloud being specially adapted for the visual control such as public security, traffic police stores, and has safety, efficiently advantage.
The exclusive safe storage cloud system of a kind of picture provided by the invention, application and the storage to the image data that visual control generates, is characterized in that, comprising: two metadata node and multiple back end; Wherein
Described metadata node for storing the metadata of minute quantity, and utilizes described metadata to realize the allocation map of described image data to the storage space of back end, the attribute of each back end of administration configuration and real-time status; And, described metadata node provides the entrance of accessing described image data, receive and the request of access of customer in response end, and the storage space corresponding to the image data that described metadata determination client is asked, by the task matching of response request of access to the back end providing this storage space;
Described back end realizes the physical store of image data and the storage space of access for being provided for, obtain image data, and the image data described according to described metadata is to the allocation map store picture data of the storage space of back end, and according to the task that metadata node is distributed, with the image data of the form of picture stream to its request access of client feedback.
Preferably, the image data that at least one data source generates by described metadata node as data stream, for each data stream sets up corresponding metadata; Described metadata is for the allocation map of the storage space of the image data to back end that describe this data stream; Further, described metadata further describes the additional information of this data stream.
Further preferably, described additional information comprises: position, control point corresponding to data stream, picture-storage duration, photo resolution, space hold situation, average discharge, support access protocal type and data stream hierarchical information.
Preferably, described metadata node is that each back end is set up and safeguards storage unit metadata, and described storage unit metadata is used for the attribute of configuration data node and upgrades the state of back end.
Further preferably, storage unit metadata comprises the read-write properties configuration information of back end, and described read-write properties configuration information is for configuring the read-write properties of arbitrary back end, and described read-write properties comprise " readable write " and " read-only " option.
Preferably, described metadata node is used for setting up and maintenance system global metadata; Described system global metadata is used for using the disk of whole cloud storage system, the monitoring state such as data volume is handled up, CPU takies, reading and writing data, space hold, and is configured system node formation and memory allocation.
Preferably, described metadata node re-registers system back end when the online dynamically increase and decrease that back end occurs is formed, and recalculates free memory, and corresponding renewal system global metadata.
Preferably, in the middle of described two metadata node, one is main frame, and another is standby host, and standby host performs 1+1 hot backup to main frame.
Preferably, described metadata node, by the backup register information of the metadata of data stream, records this data stream whether redundancy backup, the copy rank of redundancy backup and Stream Element of redundancy backup.
Preferably, described data stream is after described metadata node carries out the allocation map of storage space to it, by data transmission channel by this data stream transmitting to corresponding back end, and the image data in data stream is divided into the data block of specific size, and the naked dish of disk that data block is write direct.
Native system is specially towards the mass picture data that the Visualized Monitoring System such as public security, traffic is generated by means such as Intelligent candids, provide the exclusive safe storage cloud system of picture, can be used as the picture-storage platform of visual control, realize the business support to multiple client application based on various protocols such as HTTP or FTP.The actual features that native system generates according to image data continuation in visual control process, by the form of data stream, image data is organized, form the data access process of streaming, and be that object builds metadata with data stream, greatly simplify scale and the administration overhead of metadata, more adapt to the feature of image data relative to the Hadoop system of prior art based on the Meta data system of file.Due to not organising data in units of file, the present invention can be that the disk that unit realizes image data is directly write with block under the mode of streaming write, improves access efficiency, decreases disk fragments, reduce system maintenance cost.The present invention carries out 1+1 Redundant backup to the node storing key metadata, and can carry out redundancy backup configuration flexibly according to actual needs for mass picture data, achieves the unification of security and storage space utilization ratio.By the present invention is docked to existing Visualized Monitoring System, effectively can promote the access speed that image data stores, safeguards, applies, save operating cost, reduce failure risk, service data safety.
Accompanying drawing explanation
Fig. 1 is the exclusive safe storage cloud system of the picture described in embodiment of the present invention overall architecture schematic diagram.
Embodiment
Below by embodiment, and by reference to the accompanying drawings, technical scheme of the present invention is done more specifically bright.
Fig. 1 is the picture exclusive safe storage cloud system overall architecture schematic diagram of the embodiment of the present invention.This cloud storage system comprises 2 PSU-Master as metadata node (exclusive safe storage main frame of picture, and multiple PSU as back end (the exclusive secure storage unit of picture, PictureSpecifiedSecurityStorageUnit) PictureSpecifiedSecurityStorageUnit-Master).2 metadata node PSU-Master adopt 1+1 hot backup framework, for storing minute quantity key metadata, realize the allocation map of image data to back end storage space, the attribute managing each back end PSU and turnover dynamically, and by entrance that metadata node PSU-Master provides user to access; Back end PSU provide storage space, realizes physical store and the access of the block mode of image data, and dynamic loads and exits, and accepts the management and running of PSU-Master.
Picture of the present invention exclusive safe storage cloud system is relative to the important improvement of existing Hadoop system the Meta data system that establishes based on data stream and supports the reading and writing data mode of streaming thus instead of in the middle of Hadoop system based on the Meta data system of file and the read-write mode of documentation.
In the middle of Visualized Monitoring System, Intelligent candid equipment is as general data source, can gather continuously according to fixing or revocable frequency and generate the image data of same format and similar size, therefore its image data entirety exported can be managed as a data stream and be applied.As background technology when described in, if according to the model of Hadoop system, each frame picture is generated as picture file and sets up corresponding metadata, then can form the metadata scale of magnanimity, add the metadata updates that picture deletion operation frequently causes, the load being difficult to bear can be caused, until cause the poor efficiency of whole system even to transship to the metadata node of cloud storage system.On the contrary, the present invention, by the Meta data system based on described data stream, significantly reduces the scale of metadata, and reduces the expense of the metadata updates brought by operations such as deletions.This makes PSU-Master only need to store the metadata of minute quantity, and performs the obvious metadata updates operation reduced of relative Hadoop.
As shown in Figure 1, metadata node PSU-Master sets up the management information designated lane between data source, and establishes data transmission channel between back end PSU and data source.The image data that each data source can generate by PSU-Master is overall as a data stream, for each data stream sets up corresponding metadata; The image data merger that some data sources also can generate by PSU-Master is a data stream, and sets up corresponding metadata for the data stream after merger.
Metadata based on data stream describes the allocation map of image data to the storage space of back end PSU of this data stream, thus based on metadata, the image data of data stream can be write the storage space of the back end PSU distributed, and perform based on mapping and read the streaming of image data, change and the operation such as deletion, concrete image data access mode will be introduced hereinafter in detail.Metadata based on data stream further describes the additional information of this data stream, comprises position, control point corresponding to data stream, picture-storage duration, photo resolution, space hold situation, average discharge, support access protocal type (as FTP, HTTP etc.) etc.Native system can support Bedding storage for data stream, be the multilayer image data that this data stream storage has different quality grade, such as same candid photograph picture, the low layer pictures data generated have the quality grades such as lower resolution, its data volume corresponding is less, and high-rise image data has quality grade and the larger data volumes such as higher resolution; Data stream after layering can meet the application scenarios demand of different brackets, and some application scenarios (such as Mobile solution) only can call low layer pictures data in this data stream to save data transfer throughput; Metadata based on data stream can describe this data stream whether Bedding storage and layering configuration information.
Back end PSU obtains the actual picture data of data stream from data source by described data transmission channel, and describe according to the metadata of data stream, the storage to actual picture data is performed in the respective stored space of PSU, thus the physical store provided image data and access.
Native system supports that client is read by the streaming access of real-time performance to image data.As shown in Figure 1, first client initiates request of access to the metadata node PSU-Master as access entrance, native system supports the access under the various procotol such as FTP, HTTP, and can provide image data is searched by requests such as candid photograph monitoring of tools point position, monitor channel, storage and monitoring time segment, real time access and packing download.The request of PSU-Master customer in response end, the data stream additional information described according to metadata and allocation map, determine the storage space corresponding to image data of client-requested, and by the task matching of response request of access to the back end PSU providing this storage space; If the image data that client is asked is stored the storage space in being provided by multiple back end PSU, then the tasks in parallel of response request of access is dispensed to corresponding multiple PSU by PSU-Master.Back end PSU according to distributed task, with the form of the picture stream image data that it is asked to client feedback; If perform feedback by multiple PSU, then form the parallel picture stream transferring to client.Because native system is not organize and access data in units of file according to the mode of Hadoop system, but build based on data stream the tissue that metadata carries out image data, thus also just do not need in units of file to client transmissions image data, but the form read access of picture stream can be supported easily, namely from the data stream that PSU preserves, intercept according to metadata client asks part and send data to client in the form of streaming.
Metadata node PSU-Master builds and management of metadata based on data stream, develops and manages the mode of metadata, can significantly reduce metadata scale relative to Hadoop system object oriented file, reduces the expense applied metadata, upgrade and manage.Although Visualized Monitoring System constantly produces the image data of magnanimity level, but the quantity integrating the data stream that image data is formed according to modes such as positions, control point obviously reduces and kept stable, thus PSU-Master only needs construct towards each data stream and preserve the key metadata of few quantity, is used for managing the memory allocation to each data stream and real time status monitoring.In addition, image data circulation in data stream is deleted, because data stream during this still keeps stable, so file associated metadata frequent updating can't be there is in the present invention as Hadoop system, do not need to relate to redistributing of storage space yet, but only need to change the minority additional information item in the middle of data stream metadata, so administration overhead significantly reduces.
Also the responsible attribute to each back end PSU and real-time status carry out necessary management to the PSU-Master of metadata node.The PSU-Master of metadata node is that each back end PSU sets up and safeguards storage unit metadata, storage unit metadata back-up system management function, and monitoring is preserved and the attribute of each back end PSU of real-time update and state.Storage unit metadata comprises the read-write properties configuration information of back end PSU, by safeguarding this read-write properties configuration information, can the read-write properties of arbitrary PSU in configuration-system, comprise options such as " readable write " and " read-only ", PSU-Master can change the read-write properties of specific PSU where necessary, such as, the read-write properties configuration information of storage unit metadata is set to " read-only " to ensure global data safety.PSU-Master is also responsible for setting up and maintenance system global metadata, by this system global metadata, the disk of whole cloud storage system is used, the monitoring state such as data volume is handled up, CPU takies, reading and writing data, space hold, system node to be formed and memory allocation is configured.
In order to adjust and expand cloud storage space, the online dynamically increase and decrease of native system supported data node PSU, thus allow the elasticity dilatation of cloud storage space.After back end PSU integrated in system changes, the PSU back end that PSU-Master can re-register system is formed, and recalculates free memory, corresponding renewal system global metadata.Further, PSU-Master can according to free memory situation, to the storage space self-poise of multiple stage PSU in task assignment procedure.
In order to guarantee data security and strengthen abnormal adaptibility to response, and carry out equilibrium between safety and efficiency, native system have employed redundancy backup mechanism.Wherein, as shown in Figure 1, Operation system setting 2 is for storing the PSU-Master of key metadata, in the middle of 2 PSU-Master, one is main frame, another is standby host, and standby host performs 1+1 hot backup to main frame, ensures the PSU-Master that can switch to when main frame is abnormal as standby host at any time, maintenance system is normally run, and ensures that the key metadata of system is not lost.
For the image data that visual control produces, native system support creates the redundancy backup of 1 to 5 copy rank to it.PSU-Master is by this data stream of backup register information record of data stream metadata whether redundancy backup, the copy rank of redundancy backup and Stream Element of redundancy backup.Because native system is not access based on file system and manage to image data, but carry out accessing based on data stream and manage, therefore the redundancy backup management realized by data stream metadata is also fairly simple, and the fineness of redundancy backup is higher, redundancy backup can be performed for the specific time period part in picture stream.For the Stream Element needing redundancy backup, PSU-Master utilizes the allocation map of storage space that this part and at least one copy thereof are stored in different PSU respectively.And, PSU-Master is distributed by the hard drive space of each PSU node of storage unit metadata configurations, specify predetermined memory space as redundancy backup space, to ensure sufficient redundancy backup space, the redundancy backup pattern of multiple copy physical separation can be realized at any time.
In native system, the data source output stream such as Intelligent candid equipment, after PSU-Master carries out the allocation map of storage space to it, is stored this data stream transmitting to corresponding PSU by data transmission channel.Because the present invention does not adopt universal document system, the image data therefore in data stream also just stores unlike Hadoop system in units of single picture file.On the contrary, image data in data stream is divided into the data block of specific size, the write that it arrives disk is realized by block formula tissue, owing to not needing to be organized as file, therefore data block can be supported to write direct the naked dish of disk, realize the direct writing technology of data block, accelerate writing speed, reduce the risk of data congestion and loss.And, in the circulation delete procedure of image data, owing to have employed the data structure of streaming, file fragmentation problem can be overcome completely; Can read-write be optimized, realize the read-write of disk I/O order as far as possible.
The back end PSU that native system adopts have employed the network hard disc with ideal performance and provides cloud storage space, ensure the read-write operation ACID of each data, namely ensure atomicity (Atomicity), consistance (Consistency), isolation (Isolation), persistence (Durability).Have desirable abnormality processing performance, after equipment abnormal electrical power supply interrupts recovering, can restart voluntarily and enter can service state, and the self-inspection time is less than 2 minutes.
To sum up, native system is specially towards the mass picture data that the Visualized Monitoring System such as public security, traffic is generated by means such as Intelligent candids, provide the exclusive safe storage cloud system of picture, can be used as the picture-storage platform of visual control, realize the business support to multiple client application based on various protocols such as HTTP or FTP.The actual features that native system generates according to image data continuation in visual control process, by the form of data stream, image data is organized, form the data access process of streaming, and be that object builds metadata with data stream, greatly simplify scale and the administration overhead of metadata, more adapt to the feature of image data relative to the Hadoop system of prior art based on the Meta data system of file.Due to not organising data in units of file, the present invention can be that the disk that unit realizes image data is directly write with block under the mode of streaming write, improves access efficiency, decreases disk fragments, reduce system maintenance cost.The present invention carries out 1+1 Redundant backup to the node storing key metadata, and can carry out redundancy backup configuration flexibly according to actual needs for mass picture data, achieves the unification of security and storage space utilization ratio.By the present invention is docked to existing Visualized Monitoring System, effectively can promote the access speed that image data stores, safeguards, applies, save operating cost, reduce failure risk, service data safety.
Above embodiment is only for illustration of the present invention; and be not limitation of the present invention; the those of ordinary skill of relevant technical field; without departing from the spirit and scope of the present invention; can also make a variety of changes and modification; therefore all equivalent technical schemes also belong to category of the present invention, and scope of patent protection of the present invention should be defined by the claims.
Claims (10)
1. the exclusive safe storage cloud system of picture, application and the storage to the image data that visual control generates, is characterized in that, comprising: two metadata node and multiple back end; Wherein
Described metadata node for storing the metadata of minute quantity, and utilizes described metadata to realize the allocation map of described image data to the storage space of back end, the attribute of each back end of administration configuration and real-time status; And, described metadata node provides the entrance of accessing described image data, receive and the request of access of customer in response end, and the storage space corresponding to the image data that described metadata determination client is asked, by the task matching of response request of access to the back end providing this storage space;
Described back end realizes the physical store of image data and the storage space of access for being provided for, obtain image data, and the image data described according to described metadata is to the allocation map store picture data of the storage space of back end, and according to the task that metadata node is distributed, with the image data of the form of picture stream to its request access of client feedback.
2. the exclusive safe storage cloud system of picture according to claim 1, is characterized in that, the image data that at least one data source generates by described metadata node as data stream, for each data stream sets up corresponding metadata; Described metadata is for the allocation map of the storage space of the image data to back end that describe this data stream; Further, described metadata further describes the additional information of this data stream.
3. the exclusive safe storage cloud system of picture according to claim 2, it is characterized in that, described additional information comprises: position, control point corresponding to data stream, picture-storage duration, photo resolution, space hold situation, average discharge, support access protocal type and data stream hierarchical information.
4. the exclusive safe storage cloud system of picture according to claim 1, it is characterized in that, described metadata node is that each back end is set up and safeguards storage unit metadata, and described storage unit metadata is used for the attribute of configuration data node and upgrades the state of back end.
5. the exclusive safe storage cloud system of picture according to claim 4, it is characterized in that, storage unit metadata comprises the read-write properties configuration information of back end, described read-write properties configuration information is for configuring the read-write properties of arbitrary back end, and described read-write properties comprise " readable write " and " read-only " option.
6. the exclusive safe storage cloud system of picture according to claim 1, is characterized in that, described metadata node is used for setting up and maintenance system global metadata; Described system global metadata is used for using the disk of whole cloud storage system, the monitoring state such as data volume is handled up, CPU takies, reading and writing data, space hold, and is configured system node formation and memory allocation.
7. the exclusive safe storage cloud system of picture according to claim 1, it is characterized in that, described metadata node re-registers system back end when the online dynamically increase and decrease that back end occurs is formed, recalculate free memory, and corresponding renewal system global metadata.
8. the exclusive safe storage cloud system of picture according to claim 1, is characterized in that, in the middle of described two metadata node, one is main frame, and another is standby host, and standby host performs 1+1 hot backup to main frame.
9. the exclusive safe storage cloud system of picture according to claim 1, it is characterized in that, described metadata node, by the backup register information of the metadata of data stream, records this data stream whether redundancy backup, the copy rank of redundancy backup and Stream Element of redundancy backup.
10. the exclusive safe storage cloud system of picture according to claim 1, it is characterized in that, described data stream is after described metadata node carries out the allocation map of storage space to it, by data transmission channel by this data stream transmitting to corresponding back end, and the image data in data stream is divided into the data block of specific size, and the naked dish of disk that data block is write direct.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510438198.6A CN105095439A (en) | 2015-07-23 | 2015-07-23 | Picture specified security storage cloud system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510438198.6A CN105095439A (en) | 2015-07-23 | 2015-07-23 | Picture specified security storage cloud system |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN105095439A true CN105095439A (en) | 2015-11-25 |
Family
ID=54575876
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201510438198.6A Pending CN105095439A (en) | 2015-07-23 | 2015-07-23 | Picture specified security storage cloud system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN105095439A (en) |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2017219678A1 (en) * | 2016-06-22 | 2017-12-28 | 杭州海康威视数字技术股份有限公司 | Data recovery method and device, and cloud storage system |
| CN109229147A (en) * | 2018-07-23 | 2019-01-18 | 河源市宝腾软件科技有限公司 | A kind of passenger information system and control method |
| CN109716759A (en) * | 2016-09-02 | 2019-05-03 | 联发科技股份有限公司 | Promote quality delivery and synthesis processing |
| CN110413809A (en) * | 2019-07-31 | 2019-11-05 | 福建南威软件有限公司 | A kind of picture management method and system based on Cloud Server |
| CN112888017A (en) * | 2021-01-12 | 2021-06-01 | 杨飞 | Data fast transmission method |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| EP1441470A1 (en) * | 2003-01-21 | 2004-07-28 | Hewlett-Packard Company | Network attached storage method and system |
| CN102332029A (en) * | 2011-10-15 | 2012-01-25 | 西安交通大学 | Hadoop-based mass classifiable small file association storage method |
| CN102904917A (en) * | 2011-07-28 | 2013-01-30 | 北京邮电大学 | Mass image processing system and method thereof |
-
2015
- 2015-07-23 CN CN201510438198.6A patent/CN105095439A/en active Pending
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| EP1441470A1 (en) * | 2003-01-21 | 2004-07-28 | Hewlett-Packard Company | Network attached storage method and system |
| CN102904917A (en) * | 2011-07-28 | 2013-01-30 | 北京邮电大学 | Mass image processing system and method thereof |
| CN102332029A (en) * | 2011-10-15 | 2012-01-25 | 西安交通大学 | Hadoop-based mass classifiable small file association storage method |
Non-Patent Citations (5)
| Title |
|---|
| BYANE: ""tfs文件系统笔记"", 《HTTPS://BLOG.CSDN.NET/BYANE/ARTICLE/DETAILS/7881620?UTM_SOURCE=BLOGXGWZ9》 * |
| CHUYU: ""intro-Taocode"", 《HTTP://CODE.TAOBAO.ORG/P/TFS/WIKI/INTRO》 * |
| CONCEDE: ""TFS(Taobao FileSystem)"", 《HTTPS://WWW.XUEBUYUAN.COM/1938344.HTML》 * |
| KISIMPLE: ""TFS与Tair的对照表"", 《HTTPS://BLOG.CSDN.NET/KISIMPLE/ARTICLE/DETAILS/42023495》 * |
| 刘小俊等: ""一种结合RDBMS和Hadoop的海量小文件存储方法"", 《武汉大学学报（信息科学版）》 * |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2017219678A1 (en) * | 2016-06-22 | 2017-12-28 | 杭州海康威视数字技术股份有限公司 | Data recovery method and device, and cloud storage system |
| US10824372B2 (en) | 2016-06-22 | 2020-11-03 | Hangzhou Hikvision Digital Technology Co., Ltd. | Data recovery method and device, and cloud storage system |
| CN109716759A (en) * | 2016-09-02 | 2019-05-03 | 联发科技股份有限公司 | Promote quality delivery and synthesis processing |
| CN109229147A (en) * | 2018-07-23 | 2019-01-18 | 河源市宝腾软件科技有限公司 | A kind of passenger information system and control method |
| CN110413809A (en) * | 2019-07-31 | 2019-11-05 | 福建南威软件有限公司 | A kind of picture management method and system based on Cloud Server |
| CN112888017A (en) * | 2021-01-12 | 2021-06-01 | 杨飞 | Data fast transmission method |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| KR102025934B1 (en) | Partitioning and moving ranges in a distributed system | |
| US9430321B2 (en) | Reconstructing data stored across archival data storage devices | |
| US20120173596A1 (en) | Relational objects for the optimized management of fixed-content storage systems | |
| US7590672B2 (en) | Identification of fixed content objects in a distributed fixed content storage system | |
| US8341118B2 (en) | Method and system for dynamically replicating data within a distributed storage system | |
| US9436571B2 (en) | Estimating data storage device lifespan | |
| US7546486B2 (en) | Scalable distributed object management in a distributed fixed content storage system | |
| CN105095439A (en) | Picture specified security storage cloud system | |
| US8131783B2 (en) | Storage system, volume management method, and management computer | |
| CN104506632A (en) | Resource sharing system and method based on distributed multi-center | |
| US9424156B2 (en) | Identifying a potential failure event for a data storage device | |
| CN102541990A (en) | Database redistribution method and system utilizing virtual partitions | |
| CN104813321A (en) | Decoupled content and metadata in a distributed object storage ecosystem | |
| CN104541252A (en) | Server-based hierarchical mass storage system | |
| JP2011034164A (en) | Remote copy system and remote copy control method | |
| CN104050248A (en) | File storage system and storage method | |
| CN102708165A (en) | Method and device for processing files in distributed file system | |
| CN103067461A (en) | Metadata management system of document and metadata management method thereof | |
| US9557938B2 (en) | Data retrieval based on storage device activation schedules | |
| CN105339903A (en) | Restoring a file system object | |
| CN108197260A (en) | A kind of document file management system | |
| CN102982182A (en) | Data storage planning method and device | |
| KR20100048130A (en) | Distributed storage system based on metadata cluster and method thereof | |
| CN103023996A (en) | Cloud data storage system | |
| CN102281312B (en) | Data loading method and system and data processing method and system |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20151125 |