CN108418871A - A kind of cloud storage performance optimization method and system - Google Patents
A kind of cloud storage performance optimization method and system Download PDFInfo
- Publication number
- CN108418871A CN108418871A CN201810139125.0A CN201810139125A CN108418871A CN 108418871 A CN108418871 A CN 108418871A CN 201810139125 A CN201810139125 A CN 201810139125A CN 108418871 A CN108418871 A CN 108418871A
- Authority
- CN
- China
- Prior art keywords
- subregion
- file
- result
- calculation
- migration
- 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
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L67/00—Network arrangements or protocols for supporting network services or applications
- H04L67/01—Protocols
- H04L67/10—Protocols in which an application is distributed across nodes in the network
- H04L67/1097—Protocols in which an application is distributed across nodes in the network for distributed storage of data in networks, e.g. transport arrangements for network file system [NFS], storage area networks [SAN] or network attached storage [NAS]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F3/00—Input arrangements for transferring data to be processed into a form capable of being handled by the computer; Output arrangements for transferring data from processing unit to output unit, e.g. interface arrangements
- G06F3/06—Digital input from, or digital output to, record carriers, e.g. RAID, emulated record carriers or networked record carriers
- G06F3/0601—Interfaces specially adapted for storage systems
- G06F3/0602—Interfaces specially adapted for storage systems specifically adapted to achieve a particular effect
- G06F3/061—Improving I/O performance
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F3/00—Input arrangements for transferring data to be processed into a form capable of being handled by the computer; Output arrangements for transferring data from processing unit to output unit, e.g. interface arrangements
- G06F3/06—Digital input from, or digital output to, record carriers, e.g. RAID, emulated record carriers or networked record carriers
- G06F3/0601—Interfaces specially adapted for storage systems
- G06F3/0628—Interfaces specially adapted for storage systems making use of a particular technique
- G06F3/0638—Organizing or formatting or addressing of data
- G06F3/0643—Management of files
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F3/00—Input arrangements for transferring data to be processed into a form capable of being handled by the computer; Output arrangements for transferring data from processing unit to output unit, e.g. interface arrangements
- G06F3/06—Digital input from, or digital output to, record carriers, e.g. RAID, emulated record carriers or networked record carriers
- G06F3/0601—Interfaces specially adapted for storage systems
- G06F3/0628—Interfaces specially adapted for storage systems making use of a particular technique
- G06F3/0638—Organizing or formatting or addressing of data
- G06F3/0644—Management of space entities, e.g. partitions, extents, pools
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F3/00—Input arrangements for transferring data to be processed into a form capable of being handled by the computer; Output arrangements for transferring data from processing unit to output unit, e.g. interface arrangements
- G06F3/06—Digital input from, or digital output to, record carriers, e.g. RAID, emulated record carriers or networked record carriers
- G06F3/0601—Interfaces specially adapted for storage systems
- G06F3/0628—Interfaces specially adapted for storage systems making use of a particular technique
- G06F3/0655—Vertical data movement, i.e. input-output transfer; data movement between one or more hosts and one or more storage devices
- G06F3/0656—Data buffering arrangements
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F3/00—Input arrangements for transferring data to be processed into a form capable of being handled by the computer; Output arrangements for transferring data from processing unit to output unit, e.g. interface arrangements
- G06F3/06—Digital input from, or digital output to, record carriers, e.g. RAID, emulated record carriers or networked record carriers
- G06F3/0601—Interfaces specially adapted for storage systems
- G06F3/0668—Interfaces specially adapted for storage systems adopting a particular infrastructure
- G06F3/067—Distributed or networked storage systems, e.g. storage area networks [SAN], network attached storage [NAS]
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L63/00—Network architectures or network communication protocols for network security
- H04L63/04—Network architectures or network communication protocols for network security for providing a confidential data exchange among entities communicating through data packet networks
- H04L63/0428—Network architectures or network communication protocols for network security for providing a confidential data exchange among entities communicating through data packet networks wherein the data content is protected, e.g. by encrypting or encapsulating the payload
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L63/00—Network architectures or network communication protocols for network security
- H04L63/08—Network architectures or network communication protocols for network security for authentication of entities
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L67/00—Network arrangements or protocols for supporting network services or applications
- H04L67/50—Network services
- H04L67/55—Push-based network services
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L67/00—Network arrangements or protocols for supporting network services or applications
- H04L67/50—Network services
- H04L67/56—Provisioning of proxy services
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L67/00—Network arrangements or protocols for supporting network services or applications
- H04L67/50—Network services
- H04L67/56—Provisioning of proxy services
- H04L67/563—Data redirection of data network streams
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L67/00—Network arrangements or protocols for supporting network services or applications
- H04L67/50—Network services
- H04L67/56—Provisioning of proxy services
- H04L67/568—Storing data temporarily at an intermediate stage, e.g. caching
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L63/00—Network architectures or network communication protocols for network security
- H04L63/16—Implementing security features at a particular protocol layer
- H04L63/168—Implementing security features at a particular protocol layer above the transport layer
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Computer Networks & Wireless Communication (AREA)
- Signal Processing (AREA)
- General Engineering & Computer Science (AREA)
- Human Computer Interaction (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Computer Security & Cryptography (AREA)
- Computer Hardware Design (AREA)
- Computing Systems (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The present invention provides a kind of cloud storage performance optimization method and systems, are related to the technical field of internet, including：Obtain the partition information of each subregion of memory node in Swift cloud storage frameworks；The first subregion and the second subregion are determined in each subregion based on partition information；By in the file migration to agent node in the first subregion, and the file migration in the second subregion is gone out in agent node.In embodiments of the present invention, by the dynamic migration of hot spot data, the access speed for improving cloud storage system has been achieved the purpose that, and then solve the technical issues of cache cache policies traditional in the prior art cannot effectively improve the read or write speed of mass small documents.
Description
Technical field
The present invention relates to the technical fields of internet, more particularly, to a kind of cloud storage performance optimization method and system.
Background technology
With the arrival in internet+epoch, power marketing information system new business develops rapidly, such as：Mobile phone visitor
Family end, online business hall, Online Payment, 95598 telephone services etc. so that the quantity of isomery, non-structured data is in finger
Number property increases, such as：Work order, picture, video, and with word, pdf, ppt etc. be the document files of representative.Internet service
Hot spot data meets Zipf distributions, and similar, there is also hot spot-effect, i.e., certain numbers for the data access of power marketing business
According to access frequency considerably beyond other data.In this context, cloud storage framework becomes the carrier of magnanimity isomeric data,
Various data type, quick stream compression have contained huge commercial value in valuable data resource.
Conventional storage technologies generally use cache cachings cache hot spot data, pass through the raising of cache hit rate
To promote the access performance of hot spot data.However as the arriving in big data epoch, traditional cache cache policies can not have
Effect improves the read or write speed of mass small documents, and cloud storage framework also has no specifically for mass small documents hot spot number currently popular
According to carrying out analysis with storage optimization.
In view of the above problems, not proposing effective solution also.
Invention content
In view of this, the purpose of the present invention is to provide a kind of cloud storage performance optimization method and system, it is existing to solve
The technical issues of traditional cache cache policies cannot effectively improve the read or write speed of mass small documents in technology.
An embodiment of the present invention provides a kind of cloud storage performance optimization method, this method includes：Obtain Swift cloud storage framves
The partition information of each subregion of memory node in structure, wherein each subregion is for storing belonging same type of small text
Part, the small documents are the file less than preset value, and the partition information includes：Partition size information and/or regional addressing time
Number information；The first subregion and the second subregion are determined in each subregion based on the partition information, wherein described first point
The access frequency of file, which is greater than, in area presets access frequency, and the access frequency of file is less than pre- in second subregion
If access frequency；By in the file migration to agent node in first subregion, and the file in second subregion is moved
It removes in the agent node；Wherein, after by the file migration to the agent node in first subregion, user is just
The file in first subregion can be directly accessed in the agent node.
Further, determine that the first subregion and the second subregion include in each subregion based on the partition information：
The hot value of each subregion is calculated based on the partition information；The subregion of the hot value divided by each subregion is big
Small information, to obtain the result of calculation of each subregion, wherein the result of calculation is for characterizing in each subregion
The access frequency of file；First subregion and second subregion are determined based on the result of calculation.
Further, determine that first subregion and second subregion include based on the result of calculation：By the meter
It calculates result to be ranked up, obtains collating sequence；Subregion corresponding to first object result of calculation in the collating sequence is made
For first subregion, wherein the first object result of calculation is the result more than or equal to default access frequency；By institute
The subregion corresponding to the second target result of calculation in collating sequence is stated as second subregion, wherein second target
Result of calculation is the result less than the default access frequency.
Further, include by file migration to the agent node in first subregion：Based on hill-climbing algorithm by institute
It states in the first zoned migration to the agent node；File migration in second subregion is gone out in the agent node to wrap
It includes：Second zoned migration is gone out into the agent node using hungry value strategy.
Further, the method further includes：It is Ring files based on the file generated format in first subregion；It will
It is preserved in the Ring file push to the memory node.
According to embodiments of the present invention, a kind of cloud storage performance optimization system is additionally provided, the system comprises：Data acquire
Module, temperature computing module, subregion dynamic migration module and Swift cloud storage frameworks；The data acquisition module, for obtaining
The partition information of each subregion of memory node in Swift cloud storage frameworks, wherein each subregion is belonging same for storing
The small documents of one type, the small documents are the file less than preset value, and the partition information includes：Partition size information and/
Or regional addressing number information；The temperature computing module, for being determined in each subregion based on the partition information
First subregion and the second subregion, wherein the access frequency of file is greater than default access frequency in first subregion,
The access frequency of file is less than default access frequency in second subregion；The subregion dynamic migration module, being used for will be described
In file migration to agent node in first subregion, and the file migration in second subregion is gone out into the agent node
In；Wherein, after by the file migration to the agent node in first subregion, user can be directly in the generation
The file in first subregion is accessed in reason node.
Further, the temperature computing module includes：Hot statistics algorithm unit, by based on based on the partition information
Calculate the hot value of each subregion；By the hot value divided by the partition size information of each subregion, to obtain
State the result of calculation of each subregion, wherein the result of calculation is used to characterize the access frequency of file in each subregion.
Further, the subregion dynamic migration module includes：Subregion temperature sequencing unit；The subregion temperature sequence is single
Member obtains collating sequence for the result of calculation of the temperature computing module to be ranked up；By in the collating sequence
Subregion corresponding to one target result of calculation is as first subregion, wherein the first object result of calculation be more than or
Person is equal to the result of default access frequency；Using the subregion corresponding to the second target result of calculation in the collating sequence as institute
State the second subregion, wherein the second target result of calculation is the result less than the default access frequency.
Further, the subregion dynamic migration module further includes：Zoned migration policy unit, for being based on hill-climbing algorithm
First zoned migration to the agent node is neutralized, second zoned migration is gone out into the generation using hungry value strategy
Manage node.
Further, the subregion dynamic migration module further includes：Ring file generating units, for being based on described first
File generated format in subregion is Ring files；Ring file encryption push units are used for the Ring file push extremely
It is preserved in the memory node.
In embodiments of the present invention, the partition information of each subregion of memory node in Swift cloud storage frameworks is obtained first；
Then, determine that the first subregion and the second subregion finally move the file in the first subregion in each subregion based on partition information
It moves in agent node, and the file migration in the second subregion is gone out in agent node.In embodiments of the present invention, pass through hot spot
The dynamic migration of data has achieved the purpose that the access speed for improving cloud storage system, has solved traditional in the prior art
Cache cache policies cannot effectively improve the technical issues of read or write speed of mass small documents, realize inhomogeneity under cloud framework
The service dynamic of type prediction data access temperature, and effectively improve the technique effect of the read or write speed of mass small documents.
Other features and advantages of the present invention will illustrate in the following description, also, partly become from specification
It obtains it is clear that understand through the implementation of the invention.The purpose of the present invention and other advantages are in specification, claims
And specifically noted structure is realized and is obtained in attached drawing.
To enable the above objects, features and advantages of the present invention to be clearer and more comprehensible, preferred embodiment cited below particularly, and coordinate
Appended attached drawing, is described in detail below.
Description of the drawings
It, below will be to specific in order to illustrate more clearly of the specific embodiment of the invention or technical solution in the prior art
Embodiment or attached drawing needed to be used in the description of the prior art are briefly described, it should be apparent that, in being described below
Attached drawing is some embodiments of the present invention, for those of ordinary skill in the art, before not making the creative labor
It puts, other drawings may also be obtained based on these drawings.
Fig. 1 is a kind of flow chart of the cloud storage performance optimization method provided according to embodiments of the present invention；
Fig. 2 is the integral frame structure figure of the SSH cryptographic protocols provided according to embodiments of the present invention；
Fig. 3 is a kind of structure chart of the cloud storage performance optimization system provided according to embodiments of the present invention；
Fig. 4 is the interaction figure of the data collecting module collected partition size information provided according to embodiments of the present invention.
Specific implementation mode
In order to make the object, technical scheme and advantages of the embodiment of the invention clearer, below in conjunction with attached drawing to the present invention
Technical solution be clearly and completely described, it is clear that described embodiments are some of the embodiments of the present invention, rather than
Whole embodiments.Based on the embodiments of the present invention, those of ordinary skill in the art are not making creative work premise
Lower obtained every other embodiment, shall fall within the protection scope of the present invention.
Embodiment one：
Fig. 1 is a kind of flow chart of the cloud storage performance optimization method provided according to embodiments of the present invention, as shown in Figure 1,
This method comprises the following steps：
Step S102 obtains the partition information of each subregion of memory node in Swift cloud storage frameworks, wherein Ge Gefen
Area is the file less than preset value for storing belonging same type of small documents, small documents, and partition information includes：Subregion is big
Small information and/or regional addressing number information；
Step S104 determines the first subregion and the second subregion, wherein the first subregion based on partition information in each subregion
The access frequency of middle file, which is greater than, presets access frequency, and the access frequency of file is accessed less than default in the second subregion
Frequency；
Step S106, by the file migration to agent node in the first subregion, and by the file migration in the second subregion
Go out in agent node；Wherein, after by the file migration in the first subregion to the agent node, user can directly exist
The file in first subregion is accessed in agent node.
In embodiments of the present invention, the partition information of each subregion of memory node in Swift cloud storage frameworks is obtained first；
Then, determine that the first subregion and the second subregion finally move the file in the first subregion in each subregion based on partition information
It moves in agent node, and the file migration in the second subregion is gone out in agent node.In embodiments of the present invention, pass through hot spot
The dynamic migration of data has achieved the purpose that the access speed for improving cloud storage system, has solved traditional in the prior art
Cache cache policies cannot effectively improve the technical issues of read or write speed of mass small documents, realize inhomogeneity under cloud framework
The service dynamic of type prediction data access temperature, and effectively improve the technique effect of the read or write speed of mass small documents.
In an alternative embodiment, step S104, based on partition information in each subregion determine the first subregion and
Second subregion includes the following steps：
Step S1041 calculates the hot value of each subregion based on partition information；
Step S1042, by hot value divided by the partition size information of each subregion, to obtain the calculating knot of each subregion
Fruit, wherein result of calculation is used to characterize the access frequency of file in each subregion；
Step S1043 determines the first subregion and the second subregion based on result of calculation.
In embodiments of the present invention, different hot statistics methods may be used to calculate the hot value of each subregion, example
Such as, the partition information of each subregion is calculated by different hot statistics methods, to obtain the temperature of each subregion
Value.Wherein, different hot statistics methods can refer to that the hot statistics method corresponding to each subregion is different, can also be
In multiple subregions partial-partition use identical hot statistics method, specifically can according to the relevance between each subregion come into
Row determines.
In an optional embodiment, when hot statistics method is a kind of focus statistics method based on prediction,
The hot value that the focus statistics algorithm based on prediction may be used periodically to calculate each subregion.
After the hot value for determining each subregion, so that it may with the hot value of each subregion is corresponding divided by each subregion
Partition size message, obtain result of calculation, then the first subregion and the second subregion are determined based on result of calculation, wherein calculate knot
Represented by fruit is exactly the access frequency of file in each subregion.
In an alternative embodiment, step S1043 determines that the first subregion and the second subregion include based on result of calculation
Following steps：
Step S1, result of calculation is ranked up, and obtains collating sequence；
Step S2, using the subregion corresponding to the first object result of calculation in collating sequence as the first subregion, wherein the
One target result of calculation is the result more than or equal to default access frequency；
Step S3, using the subregion corresponding to the second target result of calculation in collating sequence as the second subregion, wherein the
Two target result of calculations are the result less than default access frequency.
In embodiments of the present invention, above-mentioned result of calculation corresponds to the access frequency of file in each subregion in memory node.
Therefore, in embodiments of the present invention, determined respectively in the partition size information of hot value and each subregion based on each subregion
After the result of calculation of a subregion, above-mentioned result of calculation can be ranked up, obtain sorted lists.Obtain sorted lists it
Afterwards, so that it may to determine the first subregion (that is, hot spot subregion) and the second subregion (that is, non-hot subregion) based on sorted lists.Specifically
Result of calculation in sorted lists can be compared by ground with default access frequency, if result of calculation is greater than or equal in advance
If access frequency, then the result of calculation is first object result of calculation, at this point, determining point corresponding to first object result of calculation
Area is as the first subregion (that is, hot spot subregion)；If result of calculation is less than default access frequency, which is the second mesh
Result of calculation is marked, at this point, the subregion corresponding to determining second target result of calculation is as the second subregion (that is, non-hot subregion).
In embodiments of the present invention, by the dynamic migration of hot spot data, the access speed for improving cloud storage system has been reached
The purpose of degree solves the read or write speed that cache cache policies traditional in the prior art cannot effectively improve mass small documents
The technical issues of, prediction data accesses temperature with realizing under cloud framework different types of service dynamic, and effectively improves sea
Measure the technique effect of the read or write speed of small documents.
In an optional embodiment, step S106 will be wrapped in the file migration to agent node in the first subregion
Include following steps：
It will be in the first zoned migration to agent node based on hill-climbing algorithm；
File migration in second subregion is gone out in agent node to include the following steps by step S106：
The second zoned migration is gone out into agent node using hungry value strategy.
In embodiments of the present invention, hill-climbing algorithm may be used and move hot spot subregion (that is, first subregion) is as much as possible
It moves in agent node, until reaching the 80% of agent node memory capacity；Later, tactful by non-hot subregion using starvation value
(that is, second subregion) agent node of moving out realizes the partition data dynamic migration between memory node and agent node.
Specifically, it using heuristic is to depth-first search that hill-climbing algorithm, which is a kind of method of part preferentially,
A kind of improvement, that is, since current data, be compared with the value of the neighbor data of surrounding.If current data is most
Big, then current data is returned to, as maximum value (i.e. mountain peak peak)；Otherwise just worked as to replace with highest neighbor data
Preceding data, to realize the purpose climbed to the eminence on mountain peak；So cycle is until peaking.It is carried out using this algorithm
The migration of hot spot subregion, it is ensured that the hot spot data stored in agent node the always highest file of real time access frequency, it is real
The technique effect that dynamically prediction data accesses temperature is showed.
In embodiments of the present invention, by the dynamic migration of hot spot data, the access speed for improving cloud storage system has been reached
The purpose of degree solves the read or write speed that cache cache policies traditional in the prior art cannot effectively improve mass small documents
The technical issues of, prediction data accesses temperature with realizing under cloud framework different types of service dynamic, and effectively improves sea
Measure the technique effect of the read or write speed of small documents.
In another optional embodiment, which further includes following steps：
It is Ring files based on the file generated format in the first subregion；And by Ring file push to memory node
In preserved.
It in embodiments of the present invention, can be by SSH cryptographic protocols by the storage in Ring file push to Swift frameworks
Node.
Fig. 2 is the integral frame structure figure of the SSH cryptographic protocols provided according to embodiments of the present invention.
Specifically, SSH is the abbreviation of Secure Shell, i.e. safety shell protocol, by the network group of IETF
(Network Working Group) is formulated；SSH is to establish security protocol on the basis of application layer, it is at present compared with can
It leans on, aims at telnet session and other network services provide the agreement of safety；It can be effectively prevent remotely using SSH agreements
Information leakage problem in management process.Can not only data be encrypted transmission in SSH cryptographic protocols, but also can be carried out to data
Compression, improves efficiency and the safety of data transmission.Due to its high safety and stability, just substitutes pass extensively at present
Telnet, ftp etc. of system.
As shown in Fig. 2, SSH agreements basic framework is respectively connection protocol, user authentication protocol and transmission association from top to bottom
View, general frame further includes various SSH upper layer networks security protocols.Wherein transport layer protocol is by server authentication, data encryption
And information integrity verification composition.User authentication protocol is used for subscriber authentication, and connection protocol is responsible for establishing encryption tunnel,
And encryption tunnel is supplied to upper layer application.
In embodiments of the present invention, Ring files are transmitted using SCP encrypted transmissions, SCP is a kind of based on SSH
Public key, imported into other memory nodes by the encrypted transmission tool of cryptographic protocol first, can be verified first when being transmitted
Whether public key is correct, and encrypted tunnel is established after verification is correct, and Ring files are carried out compression transmission.
In embodiments of the present invention, the partition information of each subregion of memory node in Swift cloud storage frameworks is obtained first；
Then, determine that the first subregion and the second subregion finally move the file in the first subregion in each subregion based on partition information
It moves in agent node, and the file migration in the second subregion is gone out in agent node.In embodiments of the present invention, pass through hot spot
The dynamic migration of data has achieved the purpose that the access speed for improving cloud storage system, has solved traditional in the prior art
Cache cache policies cannot effectively improve the technical issues of read or write speed of mass small documents, realize inhomogeneity under cloud framework
The service dynamic of type prediction data access temperature, and effectively improve the technique effect of the read or write speed of mass small documents.
Embodiment two：
The embodiment of the present invention additionally provides a kind of cloud storage performance optimization system, which is mainly used for executing of the invention real
The cloud storage performance optimization method that the above is provided is applied, cloud storage performance provided in an embodiment of the present invention is optimized below
System does specific introduction.
Fig. 3 is a kind of structure chart of the cloud storage performance optimization system provided according to embodiments of the present invention, as shown in figure 3,
The system includes：Data acquisition module 10, temperature computing module 20, subregion dynamic migration module 30 and Swift cloud storage frameworks
40, wherein：
Data acquisition module 10, the partition information for obtaining each subregion of memory node in Swift cloud storages framework 40,
Wherein, each subregion is the file less than preset value, partition information for storing belonging same type of small documents, small documents
Including：Partition size information 101 and/or regional addressing number information 102；
Temperature computing module 20, for determining the first subregion and the second subregion in each subregion based on partition information,
In, the access frequency of file, which is greater than, in the first subregion presets access frequency, the access frequency of file in the second subregion
Less than default access frequency；
Subregion dynamic migration module 30, for by the file migration to agent node in the first subregion, and by second point
File migration in area goes out in agent node；Wherein, after by the file migration to agent node in the first subregion, user is just
The file in the first subregion can be directly accessed in agent node.
In embodiments of the present invention, data acquisition module 10 is the basis of whole system, is collected using Socket communications
The statistical information of subregion in memory node in Swift cloud storages framework 40, including：It acquires partition size information 101 and adopts
Collect access times information 102, on this basis, data acquisition module 10 is by collected information preservation hereof for upper one
Layer temperature computing module 20 uses.
In embodiments of the present invention, acquisition partition size information 101 refers to storing section in acquisition Swift cloud storage frameworks
The size information of each subregion in point.Since Swift cloud storage frameworks are using multinode, more copy structures, partition size letter
The subregion that the acquisition of breath cannot obtain repetition can not omit subregion.
In embodiments of the present invention, each by memory node in data acquisition module acquisition Swift cloud storage frameworks first
The partition information of a subregion；Then, by temperature computing module based on partition information in each subregion determine the first subregion and
Second subregion, finally, by subregion dynamic migration module by the file migration to agent node in the first subregion, and by second
File migration in subregion goes out in agent node.In embodiments of the present invention, by the dynamic migration of hot spot data, reached and carried
The purpose of the access speed of high cloud storage system, solving cache cache policies traditional in the prior art cannot effectively improve
The technical issues of read or write speed of mass small documents, with realizing under cloud framework different types of service dynamic prediction data access
Temperature, and effectively improve the technique effect of the read or write speed of mass small documents.
Fig. 4 is the interaction figure of the data collecting module collected partition size information provided according to embodiments of the present invention.
As shown in figure 4, data acquisition module 10 is divided into client and server end to obtain by the way of socket communications
It takes the partition size of memory node and merges, finally make entire partition size information preservation for subsequent module hereof
With.
Specifically, client sends the request for obtaining partition size information；At this time server end to
Swift agency services send the request for obtaining district location information, and then Swift agency services return to district location to server-side
Information；Server end initiates to calculate the instruction of partition size to Swift storage services after getting district location information,
Swift storage services return to partition size information after getting the instruction, and executing the rear of the instruction to server end；
Finally, server end returns to the partition size information to client；Partition size information collection finishes.
In embodiments of the present invention, acquisition access times information 102 is since Swift cloud storage frameworks use agent node
With the mode of memory node interaction, therefore, agent node, the record partitioning in agent node are passed through in the access of all objects
Access times.But the not download time of conservation object or subregion in agent node, it is therefore desirable to change Swift cloud storage framves
Structure realizes that regional addressing number information collects task.
Specifically, object is accessed using GETorHead modes in Swift, positions proxy/controllers/
The function is finally passed through in the access of GETorHead functions in obj.py files, all objects, therefore is added in the function
Data aggregation service.When receiving the access request of object, the subregion where the object is calculated first, reads subregion statistics
Data simultaneously lock, the subsequent number of partitions carry out it is cumulative after write back in subregion statistical data, unlock and wait for and calculate next time.Core
Thought is schematically as follows：
Since object accesses use multithreading in Swift cloud storage frameworks, it is therefore necessary to add to subregion statistics file
For lock to ensure the accuracy counted, it is very low negligible that experiment shows that file locking influences system performance.So far, subregion
Access times collecting work is completed, and the file of subregion download time is generated, and this document can be automatically updated with the arrival of access.
In embodiments of the present invention, the partition information of each subregion of memory node in Swift cloud storage frameworks is obtained first；
Then, determine that the first subregion and the second subregion finally move the file in the first subregion in each subregion based on partition information
It moves in agent node, and the file migration in the second subregion is gone out in agent node.In embodiments of the present invention, pass through hot spot
The dynamic migration of data has achieved the purpose that the access speed for improving cloud storage system, has solved traditional in the prior art
Cache cache policies cannot effectively improve the technical issues of read or write speed of mass small documents, realize inhomogeneity under cloud framework
The service dynamic of type prediction data access temperature, and effectively improve the technique effect of the read or write speed of mass small documents.
In an optional embodiment, as shown in figure 3, temperature computing module 20 includes：Hot statistics algorithm unit
201, the hot value for calculating each subregion based on partition information；Hot value divided by the partition size of each subregion are believed
Breath, to obtain the result of calculation of each subregion, wherein result of calculation is used to characterize the access frequency of file in each subregion.
In embodiments of the present invention, hot statistics algorithm unit 201 is the core of temperature computing module 20, the module according to
Partition size information 101 and regional addressing number information 102, according to the difference in configuration selection hot statistics algorithm unit 201
Hot statistics algorithm is calculated, such as：Memory node is periodically calculated using the focus statistics algorithm based on prediction
Hot value divided by partition size is used in combination in the hot value of middle subregion, obtains the result of calculation of each subregion, specific calculating process is such as
Under：
(1) Regress Forecast algorithm needs to obtain the access situation of a subzone every 1 minute, and is remembered
Record keeps regional addressing number in nearest 10 minutes.
(2) hot statistics algorithm unit 201 needs to sample the access situation of data in different cycles, in the embodiment of the present invention
In, set T1 to a hour, T2 is set as half an hour, and T3 is set as 10 minutes.A T3 was updated every 10 minutes
The access times in period update the access times of T2 every half an hour, and the access times of T1 are updated every a hour.
(3) the dynamic migration period is that the time is arranged in embodiments of the present invention for the time interval of progress dynamic migration
It is 15 minutes, i.e., every 15 minutes, calculates the hot value of subregion in next period, and zoned migration is carried out with this hot value.
Specifically, the pseudocode of temperature computing module 20 is described as follows：
It should be noted that a variety of methods may be used in hot statistics algorithm unit 201, for example, using binary linearity
It returns to predict the access situation of next period subregion, and as the input of multicycle hot statistics algorithm, it is final to predict
Go out the hot value of next period subregion.
In another optional embodiment, as shown in figure 3, subregion dynamic migration module 30 includes：Subregion temperature is arranged
Sequence unit 301；Wherein, subregion temperature sequencing unit 301 is for arranging the result of calculation of hot statistics algorithm unit 201
Sequence obtains collating sequence；Using the subregion corresponding to the first object result of calculation in collating sequence as the first subregion, wherein
First object result of calculation is the result more than or equal to default access frequency；The second target in collating sequence is calculated into knot
Subregion corresponding to fruit is as the second subregion, wherein the second target result of calculation is the result less than the default access frequency.
Optionally, as shown in figure 3, subregion dynamic migration module 30 further includes：Ring file generating units 302 are used for base
File generated format in the first subregion is Ring files；Ring file encryptions push unit 303, for pushing away Ring files
It send and is preserved into memory node.
Optionally, as shown in figure 3, subregion dynamic migration module 30 further includes：Zoned migration policy unit 304 is used for base
The first zoned migration to agent node is neutralized in hill-climbing algorithm, the second zoned migration is gone out into agent node using hungry value strategy.
In embodiments of the present invention, by the dynamic migration of hot spot data, the access speed for improving cloud storage system has been reached
The purpose of degree solves the read or write speed that cache cache policies traditional in the prior art cannot effectively improve mass small documents
The technical issues of, prediction data accesses temperature with realizing under cloud framework different types of service dynamic, and effectively improves sea
Measure the technique effect of the read or write speed of small documents.
Below with a specific embodiment come to the present invention hot spot subregion dynamic migration explain.Specifically, hot
The work miscarriage of point subregion dynamic migration is as follows：
(1) timing ga(u)ge point counting area hot value H, partition size Size sort from high to low according to H/Size.
(2) CurrentProxy lists are set, in scanning object.ring.gz files _ replica2part2dev_
The corresponding list of tri- copies of id, the subregion id for belonging to agent node is added in CurrentProxy lists.
(3) WantToProxy lists are set, it will according to the sequence of subregion hot value/partition size using hill-climbing algorithm
Subregion id is added to as possible in the list, until accumulative capacity summation reaches agent node total capacity 80%.
(4) MoveInProxy lists are set, indicate the subregion for wishing to move into agent node, but subregion is not being acted on behalf of at this time
In node.WantToProxy lists are scanned in order, if subregion id belongs to CurrentProxy lists, by subregion id
It deletes from CurrentProxy lists, otherwise subregion id is added in MoveInProxy lists.At this time
That preserved in CurrentProxy lists is the subregion id that move out, and CurrentProxy will be assigned a value of at this time
MoveOutProxy lists.
(5) NoProxyGreedy dictionaries are set, which preserves the id of non-proxy equipment and corresponding hungry value, initially
Value is set as 0.
(6) MoveInProxy lists are scanned, the corresponding equipment id of first copy in record partitioning id first sets this
Standby id starvation values add one, i.e. NoProxyGreedy [id] plus one.The corresponding equipment id of the subregion is then revised as agent node
Id completes moving into for hot spot subregion.
(7) to NoProxyGreedyGreedy dictionaries, it is worth descending sequence by hungry.
(8) MoveOutProxy lists are scanned, subregion id is revised as starvation corresponding to the copy of agent equipment id originally
It is worth maximum equipment id, if hungry be worth maximum equipment id in the corresponding list of devices of subregion id, selection time
Big, until the third-largest, it can centainly find a hungry equipment id.Then the hungry value of equipment is subtracted one, resequenced
NoProxyGreedy dictionaries.Step (6) is repeated until all subregion id all complete by modification, non-hot subregion has been moved out at this time
At.
(9) objec.ring.gz regenerated is preserved.
(10) newly-generated object.ring.gz files are pushed to using scp in each memory node.scp new_
object_ring.gz openstackwift0@192.168.1.10:/etc/swift/object.ring.gz。
So far, hot spot subregion dynamic migration is completed, and the more new demand servicing of each memory node carries out subregion according to Ring files
Data Migration.
In embodiments of the present invention, the partition information of each subregion of memory node in Swift cloud storage frameworks is obtained first；
Then, determine that the first subregion and the second subregion finally move the file in the first subregion in each subregion based on partition information
It moves in agent node, and the file migration in the second subregion is gone out in agent node.In embodiments of the present invention, pass through hot spot
The dynamic migration of data has achieved the purpose that the access speed for improving cloud storage system, has solved traditional in the prior art
Cache cache policies cannot effectively improve the technical issues of read or write speed of mass small documents, realize inhomogeneity under cloud framework
The service dynamic of type prediction data access temperature, and effectively improve the technique effect of the read or write speed of mass small documents.
The technique effect and preceding method embodiment phase of the device that the embodiment of the present invention is provided, realization principle and generation
Together, to briefly describe, device embodiment part does not refer to place, can refer to corresponding contents in preceding method embodiment.
In addition, in the description of the embodiment of the present invention unless specifically defined or limited otherwise, term " installation ", " phase
Even ", " connection " shall be understood in a broad sense, for example, it may be being fixedly connected, may be a detachable connection, or be integrally connected；It can
Can also be electrical connection to be mechanical connection；It can be directly connected, can also indirectly connected through an intermediary, Ke Yishi
Connection inside two elements.For the ordinary skill in the art, above-mentioned term can be understood at this with concrete condition
Concrete meaning in invention.
In the description of the present invention, it should be noted that term "center", "upper", "lower", "left", "right", "vertical",
The orientation or positional relationship of the instructions such as "horizontal", "inner", "outside" be based on the orientation or positional relationship shown in the drawings, merely to
Convenient for the description present invention and simplify description, do not indicate or imply the indicated device or element must have a particular orientation,
With specific azimuth configuration and operation, therefore it is not considered as limiting the invention.In addition, term " first ", " second ",
" third " is used for description purposes only, and is not understood to indicate or imply relative importance.
It is apparent to those skilled in the art that for convenience and simplicity of description, the system of foregoing description,
The specific work process of device and unit, can refer to corresponding processes in the foregoing method embodiment, and details are not described herein.
In several embodiments provided herein, it should be understood that disclosed systems, devices and methods, it can be with
It realizes by another way.The apparatus embodiments described above are merely exemplary, for example, the division of the unit,
Only a kind of division of logic function, formula that in actual implementation, there may be another division manner, in another example, multiple units or component can
To combine or be desirably integrated into another system, or some features can be ignored or not executed.Another point, it is shown or beg for
The mutual coupling, direct-coupling or communication connection of opinion can be by some communication interfaces, device or unit it is indirect
Coupling or communication connection can be electrical, machinery or other forms.
The unit illustrated as separating component may or may not be physically separated, aobvious as unit
The component shown may or may not be physical unit, you can be located at a place, or may be distributed over multiple
In network element.Some or all of unit therein can be selected according to the actual needs to realize the mesh of this embodiment scheme
's.
In addition, each functional unit in each embodiment of the present invention can be integrated in a processing unit, it can also
It is that each unit physically exists alone, it can also be during two or more units be integrated in one unit.
It, can be with if the function is realized in the form of SFU software functional unit and when sold or used as an independent product
It is stored in the executable non-volatile computer read/write memory medium of a processor.Based on this understanding, of the invention
Technical solution substantially the part of the part that contributes to existing technology or the technical solution can be with software in other words
The form of product embodies, which is stored in a storage medium, including some instructions use so that
One computer equipment (can be personal computer, server or the network equipment etc.) executes each embodiment institute of the present invention
State all or part of step of method.And storage medium above-mentioned includes：USB flash disk, mobile hard disk, read-only memory (ROM, Read-
Only Memory), random access memory (RAM, Random Access Memory), magnetic disc or CD etc. are various can be with
Store the medium of program code.
Finally it should be noted that：Embodiment described above, only specific implementation mode of the invention, to illustrate the present invention
Technical solution, rather than its limitations, scope of protection of the present invention is not limited thereto, although with reference to the foregoing embodiments to this hair
It is bright to be described in detail, it will be understood by those of ordinary skill in the art that：Any one skilled in the art
In the technical scope disclosed by the present invention, it can still modify to the technical solution recorded in previous embodiment or can be light
It is readily conceivable that variation or equivalent replacement of some of the technical features；And these modifications, variation or replacement, do not make
The essence of corresponding technical solution is detached from the spirit and scope of technical solution of the embodiment of the present invention, should all cover the protection in the present invention
Within the scope of.Therefore, the protection scope of the present invention shall be subject to the protection scope of the claims.
Claims (10)
1. a kind of cloud storage performance optimization method, which is characterized in that including：
Obtain the partition information of each subregion of memory node in Swift cloud storage frameworks, wherein each subregion is for storing
Belonging same type of small documents, the small documents are the file less than preset value, and the partition information includes：Partition size
Information and/or regional addressing number information；
The first subregion and the second subregion are determined in each subregion based on the partition information, wherein first subregion
The access frequency of middle file, which is greater than, presets access frequency, and the access frequency of file is less than default in second subregion
Access frequency；
By in the file migration to agent node in first subregion, and the file migration in second subregion gone out described
In agent node；Wherein, after by the file migration to the agent node in first subregion, user can be direct
The file in first subregion is accessed in the agent node.
2. according to the method described in claim 1, it is characterized in that, being determined in each subregion based on the partition information
First subregion and the second subregion include：
The hot value of each subregion is calculated based on the partition information；
By the hot value divided by the partition size information of each subregion, to obtain the calculating knot of each subregion
Fruit, wherein the result of calculation is used to characterize the access frequency of file in each subregion；
First subregion and second subregion are determined based on the result of calculation.
3. according to the method described in claim 2, it is characterized in that, determining first subregion and institute based on the result of calculation
Stating the second subregion includes：
The result of calculation is ranked up, collating sequence is obtained；
Using the subregion corresponding to the first object result of calculation in the collating sequence as first subregion, wherein described
First object result of calculation is the result more than or equal to default access frequency；
Using the subregion corresponding to the second target result of calculation in the collating sequence as second subregion, wherein described
Second target result of calculation is the result less than the default access frequency.
4. according to the method described in claim 1, it is characterized in that,
Include by file migration to the agent node in first subregion：Based on hill-climbing algorithm by first zoned migration
To in the agent node；
File migration in second subregion, which is gone out the agent node, includes：Using starvation value strategy by described second point
Area migrates out the agent node.
5. according to the method described in claim 1, it is characterized in that, the method further includes：
It is Ring files based on the file generated format in first subregion；
It will be preserved in the Ring file push to the memory node.
6. a kind of cloud storage performance optimization system, which is characterized in that including：Data acquisition module, temperature computing module, subregion are dynamic
State transferring module and Swift cloud storage frameworks；
The data acquisition module, the partition information for obtaining each subregion of memory node in Swift cloud storage frameworks,
In, each subregion is the file less than preset value, institute for storing belonging same type of small documents, the small documents
Stating partition information includes：Partition size information and/or regional addressing number information；
The temperature computing module, for determining the first subregion and second point in each subregion based on the partition information
Area, wherein the access frequency of file is greater than default access frequency, the second subregion Chinese in first subregion
The access frequency of part is less than default access frequency；
The subregion dynamic migration module, for by the file migration to agent node in first subregion, and will be described
File migration in second subregion goes out in the agent node；Wherein, by the file migration in first subregion to described
After agent node, user can directly access the file in first subregion in the agent node.
7. system according to claim 6, which is characterized in that the temperature computing module includes：Hot statistics algorithm list
Member, the hot value for calculating each subregion based on the partition information；By the hot value divided by each subregion
Partition size information, to obtain the result of calculation of each subregion, wherein the result of calculation is described each for characterizing
The access frequency of file in a subregion.
8. system according to claim 6, which is characterized in that the subregion dynamic migration module includes：Subregion temperature is arranged
Sequence unit；
The subregion temperature sequencing unit obtains sequence sequence for the result of calculation of the temperature computing module to be ranked up
Row；
Using the subregion corresponding to the first object result of calculation in the collating sequence as first subregion, wherein described
First object result of calculation is the result more than or equal to default access frequency；
Using the subregion corresponding to the second target result of calculation in the collating sequence as second subregion, wherein described
Second target result of calculation is the result less than the default access frequency.
9. system according to claim 6, which is characterized in that the subregion dynamic migration module further includes：
Zoned migration policy unit is utilized for being neutralized first zoned migration to the agent node based on hill-climbing algorithm
Second zoned migration is gone out the agent node by hungry value strategy.
10. system according to claim 6, which is characterized in that the subregion dynamic migration module further includes：
Ring file generating units, for being Ring files based on the file generated format in first subregion；
Ring file encryption push units, for will be preserved in the Ring file push to the memory node.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810139125.0A CN108418871B (en) | 2018-02-09 | 2018-02-09 | Cloud storage performance optimization method and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810139125.0A CN108418871B (en) | 2018-02-09 | 2018-02-09 | Cloud storage performance optimization method and system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN108418871A true CN108418871A (en) | 2018-08-17 |
| CN108418871B CN108418871B (en) | 2022-02-11 |
Family
ID=63128343
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810139125.0A Active CN108418871B (en) | 2018-02-09 | 2018-02-09 | Cloud storage performance optimization method and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108418871B (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111158807A (en) * | 2019-11-29 | 2020-05-15 | 华为技术有限公司 | Data access method and device based on cloud virtual machine |
| CN112256204A (en) * | 2020-10-28 | 2021-01-22 | 重庆紫光华山智安科技有限公司 | Storage resource allocation method and device, storage node and storage medium |
| CN114185866A (en) * | 2022-02-16 | 2022-03-15 | 安翰科技(武汉)股份有限公司 | Data local storage control method and medical data storage system |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103118133A (en) * | 2013-02-28 | 2013-05-22 | 浙江大学 | Mixed cloud storage method based on file access frequency |
| CN104142953A (en) * | 2013-05-10 | 2014-11-12 | 中国电信股份有限公司 | Metadata dynamic management method and system |
| CN104462240A (en) * | 2014-11-18 | 2015-03-25 | 浪潮(北京)电子信息产业有限公司 | Method and system for realizing hierarchical storage and management in cloud storage |
| CN106210015A (en) * | 2016-07-05 | 2016-12-07 | 福州大学 | A kind of cloud storage method of temperature data buffer storage in mixed cloud structure |
| US20170245280A1 (en) * | 2015-11-25 | 2017-08-24 | Network Performance Research Group Llc | System, method, and apparatus for setting device geolocation via location proxies |
-
2018
- 2018-02-09 CN CN201810139125.0A patent/CN108418871B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103118133A (en) * | 2013-02-28 | 2013-05-22 | 浙江大学 | Mixed cloud storage method based on file access frequency |
| CN104142953A (en) * | 2013-05-10 | 2014-11-12 | 中国电信股份有限公司 | Metadata dynamic management method and system |
| CN104462240A (en) * | 2014-11-18 | 2015-03-25 | 浪潮(北京)电子信息产业有限公司 | Method and system for realizing hierarchical storage and management in cloud storage |
| US20170245280A1 (en) * | 2015-11-25 | 2017-08-24 | Network Performance Research Group Llc | System, method, and apparatus for setting device geolocation via location proxies |
| CN106210015A (en) * | 2016-07-05 | 2016-12-07 | 福州大学 | A kind of cloud storage method of temperature data buffer storage in mixed cloud structure |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111158807A (en) * | 2019-11-29 | 2020-05-15 | 华为技术有限公司 | Data access method and device based on cloud virtual machine |
| CN112256204A (en) * | 2020-10-28 | 2021-01-22 | 重庆紫光华山智安科技有限公司 | Storage resource allocation method and device, storage node and storage medium |
| CN114185866A (en) * | 2022-02-16 | 2022-03-15 | 安翰科技(武汉)股份有限公司 | Data local storage control method and medical data storage system |
Also Published As
| Publication number | Publication date |
|---|---|
| CN108418871B (en) | 2022-02-11 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Xu et al. | Am I eclipsed? A smart detector of eclipse attacks for Ethereum | |
| CN104424229B (en) | A kind of calculation method and system that various dimensions are split | |
| Yi et al. | Building a network highway for big data: architecture and challenges | |
| CN104065568B (en) | Web server cluster routing method | |
| Bocchi et al. | Personal cloud storage benchmarks and comparison | |
| CN105208059B (en) | A kind of content distribution method, terminal, server and system | |
| Qiao et al. | Characterizing flow, application, and user behavior in mobile networks: A framework for mobile big data | |
| Erlandsson et al. | Crawling online social networks | |
| CN102968591B (en) | Malicious-software characteristic clustering analysis method and system based on behavior segment sharing | |
| Hefeeda et al. | Design and evaluation of a proxy cache for peer-to-peer traffic | |
| Feng et al. | A multi-tier data reduction mechanism for IoT sensors | |
| Tudoran et al. | Jetstream: Enabling high performance event streaming across cloud data-centers | |
| Li et al. | MIN: Co-governing multi-identifier network architecture and its prototype on operator’s network | |
| CN108418871A (en) | A kind of cloud storage performance optimization method and system | |
| US20160299971A1 (en) | Identifying Search Engine Crawlers | |
| CN103414608A (en) | Rapid web flow collection statistical system and method | |
| Yang et al. | A novel collaborative task offloading scheme for secure and sustainable mobile cloudlet networks | |
| Zhang et al. | Leveraging QoE heterogenity for large-scale livecaset scheduling | |
| Li et al. | Challenges, designs, and performances of large-scale open-P2SP content distribution | |
| Soleimanzadeh et al. | SD‐WLB: An SDN‐aided mechanism for web load balancing based on server statistics | |
| Canali et al. | Data acquisition in social networks: Issues and proposals | |
| Kumar et al. | An optimal and secure resource searching algorithm for unstructured mobile peer-to-peer network using particle swarm optimization | |
| CN105184559B (en) | A kind of payment system and method | |
| CN104503983A (en) | Method and device for providing website certification data for search engine | |
| Wang et al. | Enhancing internet-scale video service deployment using microblog-based prediction |
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 |