CN110209602A - Region division and space allocation method in cross-domain virtual data space - Google Patents
Region division and space allocation method in cross-domain virtual data space Download PDFInfo
- Publication number
- CN110209602A CN110209602A CN201910412026.XA CN201910412026A CN110209602A CN 110209602 A CN110209602 A CN 110209602A CN 201910412026 A CN201910412026 A CN 201910412026A CN 110209602 A CN110209602 A CN 110209602A
- Authority
- CN
- China
- Prior art keywords
- space
- data
- virtual
- global
- region
- 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
- 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
-
- 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
- G06F12/10—Address translation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/22—Indexing; Data structures therefor; Storage structures
- G06F16/2282—Tablespace storage structures; Management thereof
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Software Systems (AREA)
- Data Mining & Analysis (AREA)
- Databases & Information Systems (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The present invention proposes region division and space allocation method in a kind of cross-domain virtual data space, based on the global namespace space aggregated by Decentralized Autonomous storage resource, the unified management and access of global data are realized by cross-domain uniform areas division methods, and the space distribution that optimization global data is placed is realized based on virtual view.Firstly, formulating the modelling statement of virtual data region and virtual memory space, provide fundamental basis for region division and space distribution；Then data region division mechanism is realized and is recorded the data area customized in global namespace space nodes and divide, to support the unified management and access of global data；Finally, research virtual memory space distribution method, realizes the distribution of the virtual memory space of data area mapping, by the area maps customized into the actual storage resource of bottom.To the mapping relations established between data area and actual physical resources in logic, with the global United Dispatching for supporting mass computing to apply.
Description
Technical field:
The invention discloses region division and space allocation methods in a kind of cross-domain virtual data space, are related to the high property of wide area
Facing challenges can be calculated, field of computer technology is belonged to.
Background technique:
Currently, the dispersion of storage resource wide area and isolation autonomy, mass computing application are urgent in national high performance computing environment
Need to support the global data space of cross-domain unified access.On this basis, realize cross-domain uniform areas division methods and
Space allocation method based on virtual view is distributed according to need and is laid out storage resource and realizes reflecting for virtual region and physical resource
Penetrate conversion, with shield underlying resource isomerism and cross-domain the dispersion, thus effectively support storage resource physically disperse
Global data management and access of the autonomy to unified polymerization in logic.Currently towards high-performance computing sector region division and
In the research field of space allocation method, many scholars have carried out a large amount of research, and summary has following two class: Grid-oriented calculates
Region division and space allocation method, the region division towards cloud computing and space allocation method.
Cloud computing, which refers to, provides the resources such as calculating and storage of dynamic virtualization by internet for user, since its is convenient
Property and availability, at present internet area be widely used and develop as Windows Azure Storage, SPANStore,
OceanStore etc..On the basis of cloud computing system, there has also been significant progress, cloud storage systems to pass through wide for cloud storage technology
The storage resource of a variety of isomeries is aggregating by domain net, provides a kind of convenient available data storage and access function to user jointly
Energy.
Windows Azure Storage (WAS) provides a kind of cloud storage system that the overall situation is expansible, inexpensive.
WAS provides a kind of name space that the overall situation is expansible, by the workload of many different users and different resource requirement groups
It is combined, the storage demand provided at any point in time in this way is all than running these services on themselves dedicated hardware
Want much less.In region division and space allocation method, WAS provides multiple cross-domain in wide scope
Back mechanism is provided between StorageStamp, each StorageStamp, and WAS is that each user mentions on StorageStamp
It is divided into three parts for one or more store name spaces, and by store name space: account title, partition name and object name
Claim.Wherein, name on account is for realizing the isolation in store name space between different accounts, and partition name is for providing across storage
The data access of node, object oriented provide the physical address of real resource simultaneously, and DNS has been incorporated store name space by WAS,
Therefore, all data can be accessed by being stored in the URI of store name space list in WAS.WAS system passes through above-mentioned side
Method realizes cross-domain storage and access to data, and the isolation of the memory space based on user and safety.
SPANStore system is the cloud storage system of a kind of low latency, low cost, and SPANStore has combined multiple clouds and deposited
Service provider is stored up, a kind of data storage service across multiple clouds, including Amazon S3, Windows Azure, GCS are provided
Deng.Each application program uses an individual SPANStore deployment.In each data center of disposing application program, answer
The PUT and GET request to object are issued with the library SPANStore that program is all linked to application program.In SPANStore, often
A data center is all an access collection, and each data center stores metadata respectively.SPANStore is realized by the above method
The low cost and low latency of cloud storage system.
OceanStore system is intended to provide a kind of cloud storage system for having better availability, Performance And Reliability,
Core is the Pool of cross-domain distribution one by one, and data can mutually circulate between each Pool, a user and one or more
Pool is connected.OceanStore system is placed excellent using a kind of Global Algorithm progress data of wide-area distribution type Data Position
Change, each addressable entity (such as floating copy, Archive Clip or client computer) in OceanStore is by one or more
GUID mark.The identical entity of function is identified such as the different copies of same object by identical GUID, and client uses a system
Column protocol message is interacted with these entities.OceanStore is using cross-domain distribution and interconnected Pool is core, with
Pool connection lower layer storage resource forms a kind of cloud storage system of High Availabitity with GUID mark data resource.
MetaStorage system is a kind of Highly Scalable, High Availabitity, the cloud storage system based on distributed hashtable.
MetaStorage system is connected to actual storage resource, a Distributor with the Distributor of multiple cross-domain distributions
Many node, Amazon S3 memory node etc. can be locally stored with being connected to.In storage file data, MetaStorage
System is realized by Distributor based on hash and is distributed the position of file data, while MetaStorage system is with one
Hash key assignments preference list based on arbitration replaces and eliminates partitioning algorithm, makees load balancing.For guarantee it is multiple across
The consistency of preference list between the Distributor of domain distribution, MetaStorage system is on the basis of Distributor
Coordinator is constructed to manage the last state of the configuration of bottom Distributor and member.
Compared to cloud computing, its object is to the gridding resources for dispersing wide area to be aggregating for grid computing, big to support
The calculating application of type.Grid computing shields the isomerism of underlying resource by middleware, while passing through grid for the resource of isomery
Integration utilizes, and supports Virtual Organization, provides high-level service, and provide huge processing capacity.What Grid-oriented calculated deposits
The global union file system (GFFS) of storage system such as XSEDE system, Onedata of EGI system etc. require to provide a kind of screen
The storage method of bottom storage resource isomerism is covered, meanwhile, to meet the needs of high-performance calculation, also need to provide global height
Effect, security access mechanism.
U.S.'s XSEDE project is intended to integrate distributed resource and tool in R&E field and utilize, to study work
Make to provide extreme a science and engineering and find environment, with to more comprehensively, more there is the digital service system of cohesiveness to stride forward one
Step.The storage system GFFS of XSEDE project development, which is based on RNS, realizes global namespace space, has been connected to resource supplier
Storage resource, the storage resource of university's cluster, the storage resource of national Supercomputer Center, and support to import personal directory as storage
Resource, the container based on Genesis 2 realize the management of storage resource, with support storage resource addition, registration and
It exits.Meanwhile GFFS supports the access mould of multiple scenes such as personal computer, campus memory node, Supercomputer Center's node
Formula is supported under various scenes, global namespace space is based on, to the unified access and management of a variety of storage resources.Pass through unification
Name space, GFFS, which is realized, to be accessed, manage and shares to the unified of cross-domain isomerism storage resources.
European EGI project provides data by Onedata and metadata management service, Onedata are absorbed in distribution
Instant, the transparent access of data set allow directly to visit from local computer or working node without unnecessary temporary and migration
Ask data.In Onedata, all data will be all organized into corresponding Space, and Space is by storage resource provider
Provider provides storage capacity.Space can be considered as virtual folder or volume, it is under the jurisdiction of some user or some group,
Any catalogue and file hierarchy can be may include in shared among users, multiple Provider can also be distributed in simultaneously
In.Provider can be considered as the storage cluster of Supercomputer Center, they directly provide data storage clothes for the owner of Space
Business.If multiple Provider support the same Space, the metadata and data of Space will these Provider it
Between it is synchronous.
The Gfarm grid file system of Japanese University of tsukuba's research and development is also a kind of storage system of Grid-oriented, and Gfarm can
To combine the local file system of multiple PC clusters, the catalogue of existing local file system is integrated, this polymerization methods are not right
The additional hsrdware requirements of bottom storage resource, but data access performance may be inferior to object storage or block storage.Each PC cluster
File, catalogue are managed respectively by each PC cluster, and when storing data, Gfarm is based on each PC cluster cpu load in top layer, can be with holding
Amount, data access round-trip delay (RTT) select file storage location.
Traditional region division and space allocation method there are the problem of be mainly:
Region division and space allocation method towards cloud computing are loose to storage resource management, mainly with data center
Form provides the use of underlying resource, with the application of the resource of Relatively centralized operation dispersion, recognizes the isomerism of resource and not strong
The Virtual Organization on upper layer is adjusted, there are certain limitations for global access and unified management for data, are not suitable for current country
High-performance computing environment.The region division and space allocation method that Grid-oriented calculates can be effectively using the global moneys of polymerization
Source, but need dedicated network to support improving performance, and still loose to the management of data, it is unfavorable for the overall situation of large-scale application
United Dispatching, data Placement Strategy are optimized as unit of data center mostly, and it is unified to lack the overall situation based on virtual view
Placement Strategy.Therefore, research region division and space distribution suitable for the virtual data space of national high performance computing environment
The virtual data space that decentralized resource is polymerized, is divided into the data area of user's customization by method on demand, and realizes area
Domain separation and safety, to support storage resource physically global data management and visit of the Decentralized Autonomous to unified polymerization in logic
It asks, while realizing the transparent distribution of the unification of Decentralized Autonomous physical memory resources, optimize the global Placement Strategy of data, it is big to support
Type calculates the global United Dispatching of application.
Summary of the invention:
The main object of the present invention is to provide region division and space allocation method in a kind of cross-domain virtual data space, needle
To mass computing using demand the characteristics of resource and to data safety, the region division of virtual data space is realized
Virtual data space is divided into the data area of user's customization by method on demand, and real based on interregional access control mechanisms
Existing zone isolation and safety, to support storage resource physically global data management of the Decentralized Autonomous to unified polymerization in logic
And access；For the overall scheduling demand of data and calculating task, memory allocation method in virtual data space is realized,
It realizes the transparent distribution of unification of Decentralized Autonomous physical memory resources, while optimizing the global Placement Strategy of data, to support large size
Calculate the global United Dispatching of application.
The technical scheme is that
Region division and space allocation method in a kind of cross-domain virtual data space, which is characterized in that based on by dispersion from
The global namespace space that storage resource aggregates into is controlled, realizes that the unified of global data is managed by cross-domain uniform areas division methods
Reason and access, and the space distribution that optimization global data is placed is realized based on virtual view.Firstly, formulate virtual data region and
The modelling of virtual memory space is stated, and is provided fundamental basis for region division and space distribution；Then data region is drawn
Extension set system is realized and is recorded the data area customized in global namespace space nodes and divide, to support the system of global data
One management and access；Finally, research virtual memory space distribution method, realizes point of the virtual memory space of data area mapping
Match, by the area maps customized into the actual storage resource of bottom.To establish data area and practical object in logic
The mapping relations between resource are managed, with the global United Dispatching for supporting mass computing to apply.
It is specific as follows including 3 steps:
1) the modelling statement of virtual data region and virtual memory space is formulated；
2) data region division mechanism realizes and records the data area customized in global namespace space nodes
It divides；
3) it is divided based on data area, studies virtual memory space distribution method, realize virtually depositing for data area mapping
The distribution for storing up space, by the area maps customized into the actual storage resource of bottom.
Wherein, step 1) the following steps are included:
Step (1.1) proposes a kind of data area model of customizable, with quintuple form record area information, for number
It provides fundamental basis according to the customizable of region division.Five-tuple for portraying data region information may be designed as AREA
(zoneID,onename,owner,{member},{spaceID})；
Step (1.2) proposes a kind of modelling statement in space, records spatial information with quadruple form, is virtual memory
Space distribution is provided fundamental basis.Four-tuple for recording spatial information may be designed as SPACE (spaceID, name, size,
spacemetadata{hostcenter,storagesrcname,spacepath})；
Step (1.3) based on this, believe by the auxiliary records region such as design section information table, spatial information table and space
Breath.On the basis of wide area storage resource polymerize completion, database is constructed on regional space server for storing above-mentioned table
Middle data.
Wherein, step 2) the following steps are included:
Step (2.1) virtual data space provides a kind of customizable data region division machine for virtual data space user
System, enables a user to on-demand self-defining data region；Such as: definition can only by individual subscriber access data area, as with
The home subregion at family, stores the private data of individual subscriber；For another example: defining chemical subregion, be made of as one multiple users
Chemical work group data area, the user in working group may have access to partition data, shares experimental data etc.；In addition, also
Can define common data subregion, for storing some data for total user share and access, such as: certain common softwares and its
Service manual etc., for all user sharings；
The Decentralized Autonomous storage resource of polymerization is divided into cross-domain, Yong Huke by region division mechanism by step (2.2)
The region of unified management and access；
Step (2.3) is mapped to one or more based on region and spatial model, by a data area in logic
A virtual memory space realizes the mapping of logic region to physical memory resources, and combines security management and control mechanism, realizes data field
Isolation and safety between domain.
Wherein, step 3) the following steps are included:
Step (3.1) realizes a kind of spatial position selection strategy, is based on user customized information, comprehensively considers virtual memory sky
Between performance distance, Supercomputer Center's physical memory space surplus, the factors such as allocation capacity information, combined data and calculating task
Global collaborative dispatching requirement selects the position of allocation space according to spatial position selection algorithm；
Step (3.2) is divided based on the data area of customization, a kind of effective space allocation method is proposed, in physical store
Virtual memory space is constructed in resource, and establishes the mapping relations between data area；
Virtual memory space and actual storage resource are tied by step (3.3), support user storage, it is shared,
The operation such as reading and writing of files, realizes the transparent distribution of unification of Decentralized Autonomous physical memory resources, while optimizing the global of data and placing
Strategy, with the global United Dispatching for supporting mass computing to apply.
The invention has the advantages that
Region division and space allocation method in a kind of cross-domain virtual data space provided by the invention, with existing data
Administrative skill is compared, and major advantage is:
By the mapping relations in virtual data region and data space, the unification to cross-domain dispersion data may be implemented
Scheduling and management can more effectively meet the medium-and-large-sized resource requirement for calculating application of high-performance computing environment；Based on virtual view
The space allocation method of figure, can balance the load of dispersion storage resource, and realize the global data Placement Strategy of optimization, with excellent
Change the storage and computing resource cooperative scheduling efficiency during computation.
Detailed description of the invention:
Fig. 1 is region division and space allocation method implementation flow chart in cross-domain virtual data space
Fig. 2 is region and Spatial infrastructure figure in virtual data space
Fig. 3 is data area division and virtual memory space allocation process diagram
Fig. 4 is district-share operating procedure
Fig. 5 is regional space location information browsing process figure
Specific embodiment:
Below in conjunction with attached drawing, the present invention is described in further detail.
As shown in Figure 1, being region division and space allocation method implementing procedure in cross-domain virtual data space of the invention
Figure.Region division and space allocation method in cross-domain virtual data space, comprising the following steps:
1) the modelling statement of virtual data region and virtual memory space is formulated；
2) data region division mechanism realizes and records the data area customized in global namespace space nodes
It divides；
3) it is divided based on data area, studies virtual memory space distribution method, realize virtually depositing for data area mapping
The distribution for storing up space, by the area maps customized into the actual storage resource of bottom.
The framework in region and space in virtual data space is as shown in Fig. 2, first by data area and virtual memory space
Modelling statement is carried out, the information of data-base recording data area and virtual memory space is constructed；Secondly by region division machine
System by the global data space being polymerized by cross-domain Decentralized Autonomous storage resource be divided into it is customized be managed collectively and
The region of data is accessed, the mapping of logic region to physical memory resources is realized, and combine security management and control mechanism, realizes data field
Isolation and safety between domain；It then is that data area constructs void on physical memory resources by virtual memory space distribution method
Quasi- memory space, and mapping relations between the two are established, realize the transparent distribution of unification of Decentralized Autonomous physical memory resources, together
When optimize the global Placement Strategies of data, improve the global collaborative scheduling of mass computing application and data and calculating task
The access performance of strategy.
Data area divides and the process of virtual memory space distribution is as shown in Figure 3, comprising the following steps:
1) it when user uses virtual data space software, needs to apply for a customized region first, be sent out by client
Application region request and inclusion region customized information out；
2) application region request is sent to the virtual data space region division module of management node, and by region division mould
After block receives the application of user, according to the customized information of user, selection region partition strategy；
3) region division module calls the space distribution module of virtual data space to carry out virtual memory sky in management node
Between distribution, data area is mapped to virtual memory space；
4) spatial position selection algorithm is based on by spatial position selecting module, selection area corresponds to space and to create at which
In storage resource；
5) by space distribution module, the request to create directory is sent to the storage cluster client of selected Supercomputer Center
In, using catalogue created as virtual memory space, and initial rights are set；
6) virtual memory space of creation is mapped with data area, and spatial information is remembered in the form of four-tuple
Record is in the spatial information table on regional space server, by the authority information (reading and writing of such as user, execution permission) in region
It is recorded in the authority list of global area；
7) zone information table being recorded in user data area domain information in the form of five-tuple on regional space server
In, the weak consistency in region and spatial information between each supercomputer node is ensured by database, so far, region division and space distribution
It completes.
On the basis of the data area of building and virtual memory space, district-share operating procedure is as shown in figure 4, include
Following steps:
1) user carries out district-share application, including increase/deletion group region members by client；
2) client selection Optimal Management node is attached, and sends district-share request, the region in management node
Sharing module receives request, and the corresponding whole spaces of query region and spatial orientation information；
3) district-share module calls permission increase/permission removing module, according to space path information in area maps
User right is changed in storage cluster corresponding to whole spaces；
4) after the completion of permission modification, corresponding region members' information is updated in the database, and it is total to return client area
Enjoy operating result.
It is divided based on data area and virtual memory space distribution, regional space location information browsing process is as shown in Figure 5.
The following steps are included:
1) client, which is sent, obtains area locating information request；
2) zone location module receives request in management node, and the space of area maps is obtained to regional space server
Information；
3) zone location module sends the request for obtaining spatial orientation information to space orientation module；
4) space orientation module obtains the location information in whole spaces of area maps to regional space server；
5) zone location module is to client restoring to normal position information result.
It should be noted last that: the present invention can also have other plurality of application scenes, without departing substantially from spirit of that invention and its
In the case where essence, those skilled in the art make various corresponding changes and modifications, but this in accordance with the present invention
A little corresponding changes and modifications all should fall within the scope of protection of the appended claims of the present invention.
Claims (5)
1. region division and space allocation method in cross-domain virtual data space, which is characterized in that based on being stored by Decentralized Autonomous
Resource polymerization at global namespace space, pass through unified management and visit that cross-domain uniform areas division methods realize global data
It asks, and the space distribution that optimization global data is placed is realized based on virtual view；Firstly, formulating virtual data region and virtually depositing
The modelling statement for storing up space, provides fundamental basis for region division and space distribution；Then data region division mechanism,
In global namespace space nodes realize and record customize data area divide, with support global data unified management and
Access；Finally, research virtual memory space distribution method, realizes the distribution of the virtual memory space of data area mapping, it will
The area maps of customization are into the actual storage resource of bottom.To establish between data area and actual physical resources in logic
Mapping relations, with support mass computing apply global United Dispatching.
2. the method according to claim 1, wherein the following steps are included:
1) the modelling statement of virtual data region and virtual memory space is formulated；
2) data region division mechanism is realized and is recorded the data area customized in global namespace space nodes and draws
Point；
3) it is divided based on data area, studies virtual memory space distribution method, realize that the virtual memory of data area mapping is empty
Between distribution, by the area maps customized into the actual storage resource of bottom.
3. according to the method described in claim 2, it is characterized in that, step 1) the following steps are included:
Step (1.1) proposes a kind of data area model of customizable, is data field with quintuple form record area information
The customizable that domain divides is provided fundamental basis.Five-tuple for portraying data region information may be designed as AREA (zoneID,
onename,owner,{member},{spaceID})；
Step (1.2) proposes a kind of modelling statement in space, records spatial information with quadruple form, is virtual memory space
Distribution is provided fundamental basis.Four-tuple for recording spatial information may be designed as SPACE (spaceID, name, size,
spacemetadata{hostcenter,storagesrcname,spacepath})。
Step (1.3) based on this, auxiliary records region and the spatial information such as design section information table, spatial information table.?
On the basis of the polymerization of wide area storage resource is completed, database is constructed on regional space server for storing number in above-mentioned table
According to.
4. according to the method described in claim 2, it is characterized in that, step 2) the following steps are included:
Step (2.1) virtual data space provides a kind of customizable data region division mechanism for virtual data space user,
Enable a user to on-demand self-defining data region；
By region division mechanism, the Decentralized Autonomous storage resource of polymerization is divided into cross-domain, user to be unified step (2.2)
The region of management and access；
Step (2.3) is mapped to one or more void based on region and spatial model, by a data area in logic
Quasi- memory space realizes the mapping of logic region to physical memory resources, and combines security management and control mechanism, between realization data area
Isolation and safety.
5. according to the method described in claim 2, it is characterized in that, step 3) the following steps are included:
Step (3.1) realizes a kind of spatial position selection strategy, is based on user customized information, comprehensively considers virtual memory space
Energy distance, Supercomputer Center's physical memory space surplus, the factors such as allocation capacity information, the overall situation of combined data and calculating task
Cooperative scheduling demand selects the position of allocation space according to spatial position selection algorithm；
Step (3.2) is divided based on the data area of customization, a kind of effective space allocation method is proposed, in physical memory resources
Upper building virtual memory space, and establish the mapping relations between data area；
Virtual memory space and actual storage resource are tied by step (3.3), support storage, the shared, read-write of user
The operation such as file, realizes the transparent distribution of unification of Decentralized Autonomous physical memory resources, while optimizing the global Placement Strategy of data,
With the global United Dispatching for supporting mass computing to apply.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910412026.XA CN110209602A (en) | 2019-05-17 | 2019-05-17 | Region division and space allocation method in cross-domain virtual data space |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910412026.XA CN110209602A (en) | 2019-05-17 | 2019-05-17 | Region division and space allocation method in cross-domain virtual data space |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN110209602A true CN110209602A (en) | 2019-09-06 |
Family
ID=67787527
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910412026.XA Pending CN110209602A (en) | 2019-05-17 | 2019-05-17 | Region division and space allocation method in cross-domain virtual data space |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110209602A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112019495A (en) * | 2020-05-28 | 2020-12-01 | 北京航空航天大学 | Dynamic mapping mechanism and data security control method for wide-area virtual data space account |
| CN113138717A (en) * | 2021-04-09 | 2021-07-20 | 锐捷网络股份有限公司 | Node deployment method, device and storage medium |
Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1737780A (en) * | 2004-08-19 | 2006-02-22 | 国际商业机器公司 | System and method for transmitting information from a device drive program to the other |
| US8677354B2 (en) * | 2010-07-12 | 2014-03-18 | International Business Machines Corporation | Controlling kernel symbol visibility and accessibility across operating system linkage spaces |
| US9245140B2 (en) * | 2013-11-15 | 2016-01-26 | Kabushiki Kaisha Toshiba | Secure data encryption in shared storage using namespaces |
| CN105700826A (en) * | 2015-12-31 | 2016-06-22 | 华为技术有限公司 | Virtualization method and device |
| CN106030552A (en) * | 2014-04-21 | 2016-10-12 | 株式会社日立制作所 | Computer system |
| CN107168884A (en) * | 2016-03-08 | 2017-09-15 | 东芝存储器株式会社 | The control method of storage system, information processing system and nonvolatile memory |
| CN107463402A (en) * | 2017-07-31 | 2017-12-12 | 腾讯科技（深圳）有限公司 | The operation method and device of virtual opetrating system |
| CN109144888A (en) * | 2017-06-28 | 2019-01-04 | 东芝存储器株式会社 | Storage system |
-
2019
- 2019-05-17 CN CN201910412026.XA patent/CN110209602A/en active Pending
Patent Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1737780A (en) * | 2004-08-19 | 2006-02-22 | 国际商业机器公司 | System and method for transmitting information from a device drive program to the other |
| US8677354B2 (en) * | 2010-07-12 | 2014-03-18 | International Business Machines Corporation | Controlling kernel symbol visibility and accessibility across operating system linkage spaces |
| US9245140B2 (en) * | 2013-11-15 | 2016-01-26 | Kabushiki Kaisha Toshiba | Secure data encryption in shared storage using namespaces |
| CN106030552A (en) * | 2014-04-21 | 2016-10-12 | 株式会社日立制作所 | Computer system |
| CN105700826A (en) * | 2015-12-31 | 2016-06-22 | 华为技术有限公司 | Virtualization method and device |
| CN107168884A (en) * | 2016-03-08 | 2017-09-15 | 东芝存储器株式会社 | The control method of storage system, information processing system and nonvolatile memory |
| CN109144888A (en) * | 2017-06-28 | 2019-01-04 | 东芝存储器株式会社 | Storage system |
| CN107463402A (en) * | 2017-07-31 | 2017-12-12 | 腾讯科技（深圳）有限公司 | The operation method and device of virtual opetrating system |
Non-Patent Citations (2)
| Title |
|---|
| 刘忠麟,肖利民,郝沁汾,祝明发: "一种机群上的处理器虚拟化方法", 《北京航空航天大学学报》 * |
| 阮利; 蔺波; 肖利民: "基于虚拟机的僵尸网络仿真系统可扩展性优化", 《电信科学》 * |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112019495A (en) * | 2020-05-28 | 2020-12-01 | 北京航空航天大学 | Dynamic mapping mechanism and data security control method for wide-area virtual data space account |
| CN112019495B (en) * | 2020-05-28 | 2021-11-19 | 北京航空航天大学 | Dynamic mapping mechanism and data security control method for wide-area virtual data space account |
| CN113138717A (en) * | 2021-04-09 | 2021-07-20 | 锐捷网络股份有限公司 | Node deployment method, device and storage medium |
| CN113138717B (en) * | 2021-04-09 | 2022-11-11 | 锐捷网络股份有限公司 | Node deployment method, device and storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US10129333B2 (en) | Optimization of computer system logical partition migrations in a multiple computer system environment | |
| CN110213352B (en) | Method for aggregating dispersed autonomous storage resources with uniform name space | |
| Deka | A survey of cloud database systems | |
| CN104050042B (en) | The resource allocation methods and device of ETL operations | |
| CN103310000B (en) | Metadata management method | |
| Hongchao et al. | Distributed data organization and parallel data retrieval methods for huge laser scanner point clouds | |
| CN108268614B (en) | Distributed management method for forest resource spatial data | |
| CN102833331B (en) | Cloud storage system and metadata writing-in method and metadata reading method thereof | |
| Dong et al. | Towards a fast and secure design for enterprise‐oriented cloud storage systems | |
| CN103957270A (en) | Cloud atomic unit delivery and deployment method and device | |
| CN109314721A (en) | The management of multiple clusters of distributed file system | |
| CN104270412A (en) | Three-level caching method based on Hadoop distributed file system | |
| CN106453618A (en) | Remote sensing image processing service cloud platform system based on G-Cloud cloud computing | |
| Xiahou et al. | Multi-datacenter cloud storage service selection strategy based on AHP and backward cloud generator model | |
| CN106570113A (en) | Cloud storage method and system for mass vector slice data | |
| CN110209602A (en) | Region division and space allocation method in cross-domain virtual data space | |
| CN102158533B (en) | Distributed web service selection method based on QoS (Quality of Service) | |
| Lu et al. | An adaptive multi-level caching strategy for distributed database system | |
| CN110377399A (en) | HBase containerization method, apparatus, equipment and readable storage medium storing program for executing | |
| Dongsheng et al. | Distributed cache memory data migration strategy based on cloud computing | |
| Zhou et al. | Information intelligent management system based on Hadoop | |
| Li et al. | Research on the architecture of Cloud GNSS based on Hadoop | |
| CA1252906A (en) | Generalized directory data model | |
| Wei et al. | PIVOT: An adaptive information discovery framework for computational grids | |
| Wu et al. | DMFSsim: a distributed metadata file system simulator |
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 | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20190906 |
|
| RJ01 | Rejection of invention patent application after publication |