CN117931513A - Mixed cloud data backup management method based on multi-objective optimal copy management strategy - Google Patents
Mixed cloud data backup management method based on multi-objective optimal copy management strategy Download PDFInfo
- Publication number
- CN117931513A CN117931513A CN202311701095.5A CN202311701095A CN117931513A CN 117931513 A CN117931513 A CN 117931513A CN 202311701095 A CN202311701095 A CN 202311701095A CN 117931513 A CN117931513 A CN 117931513A
- Authority
- CN
- China
- Prior art keywords
- backup
- objective function
- objective
- data
- optimal
- 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
- 238000007726 management method Methods 0.000 title claims abstract description 77
- 239000011159 matrix material Substances 0.000 claims abstract description 41
- 230000005540 biological transmission Effects 0.000 claims abstract description 19
- 238000005457 optimization Methods 0.000 claims abstract description 18
- 238000000354 decomposition reaction Methods 0.000 claims abstract description 17
- 230000006870 function Effects 0.000 claims description 56
- 239000013598 vector Substances 0.000 claims description 24
- 238000000034 method Methods 0.000 claims description 21
- 230000008569 process Effects 0.000 claims description 9
- 238000004590 computer program Methods 0.000 claims description 4
- PCHJSUWPFVWCPO-UHFFFAOYSA-N gold Chemical compound [Au] PCHJSUWPFVWCPO-UHFFFAOYSA-N 0.000 abstract description 3
- 239000010931 gold Substances 0.000 abstract description 3
- 229910052737 gold Inorganic materials 0.000 abstract description 3
- 238000011084 recovery Methods 0.000 description 33
- 230000008901 benefit Effects 0.000 description 6
- 238000013461 design Methods 0.000 description 4
- 238000010586 diagram Methods 0.000 description 3
- 238000012544 monitoring process Methods 0.000 description 3
- 238000001514 detection method Methods 0.000 description 2
- 210000001503 joint Anatomy 0.000 description 2
- 238000012423 maintenance Methods 0.000 description 2
- 238000013523 data management Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 230000014759 maintenance of location Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000012545 processing Methods 0.000 description 1
- 230000004044 response Effects 0.000 description 1
- 238000012795 verification Methods 0.000 description 1
Landscapes
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The invention provides a data backup management method and system based on a multi-objective optimal copy management strategy, comprising the following steps: respectively defining a reliability objective function, a transmission delay objective function and a load balancing objective function for carrying out complete backup on data; acquiring an optimized objective function of data backup management based on the reliability objective function, the transmission delay objective function and the load balancing objective function; and acquiring an optimal placement scheme matrix meeting the optimization objective function by adopting a multi-objective evolution algorithm based on decomposition, and carrying out backup placement allocation on each node on the data based on the optimal placement scheme matrix. The invention comprehensively considers the reliability of backup data, the time delay of the backup system and the stability of the backup system, obtains the optimal placement scheme of the backup copy of the gold copy through a multi-objective optimization algorithm based on decomposition, and reduces the risks of data loss and damage.
Description
Technical Field
The invention belongs to the field of cloud computing disaster recovery, and particularly relates to a hybrid cloud data backup management method based on a multi-objective optimal copy management strategy.
Background
With the development of the big data age, the importance of data has risen to the level of key resources. Cloud data backup and restoration is of great importance in government/enterprise cloud processes. It involves backing up data from a local machine room to the cloud or backing up data from the cloud to the cloud for recovery when data is lost, damaged or under attack. Has important significance for ensuring data security, improving data accessibility, reducing cost and simplifying data management. However, the backup and recovery of mass data make the traditional data protection method appear to have high maintenance cost and long recovery time, and the data security is reduced.
Data copy management policy (CDM) is an efficient backup data protection solution on the cloud, an important component of government, enterprise and personal protection critical information. An excellent backup copy management strategy can cope with data loss caused by faults or natural disasters, protect service continuity, cope with network security threats, improve data recovery speed and the like. Data backup management is a key to ensuring data security and business continuity. By implementing an effective data backup strategy, the risk of data loss can be reduced, the benefits of government, enterprises and clients are protected, and the data stability, reliability, corresponding speed and the like of the whole management system are greatly influenced.
In copy management policies, the reliability of the backup data is critical. Reliability means that the backed up data should be reliable and can be restored when any problem occurs. In addition, during backup and restore, it is also necessary to ensure that the backup data can be restored by repairing the data file when an error occurs.
In the copy management strategy, latency of the backup system is one of its important characteristics. Cloud storage systems often store multiple copies of a file in duplicate and place the files on different nodes to ensure the security of the stored information and reduce the delay of user access. In a cloud storage system, when the access amount of one file is small and the number of copies of the file is excessive, unnecessary storage overhead and storage cost of the system are increased; when the access amount of a file is high and the number of copies of the file is small, the access time delay of a user for accessing the file is increased, and the user requirement cannot be met.
Load balancing of the backup system is also an important factor to consider in the copy management policy. The load balancing rate of the server not only affects the resource utilization rate of the server, but also has larger influence on the whole system.
However, in actual copy management strategies, the backup system's pursuits in three aspects may be potentially conflicting and not optimal at the same time.
Disclosure of Invention
The invention aims to provide a hybrid cloud data backup management method based on a multi-objective optimal copy management strategy, which is used for overcoming the defects of the technical scheme. On the backup and recovery management system, the backup and recovery of different types of data can be realized through one platform. In the backup module, the reliability of backup data, the overall system time delay and the load balancing problem are considered, a multi-objective optimization model is established, an optimal copy management strategy of Pareto balance of the three problems is obtained, and backup copies are backed up to a bottom storage space of the system based on the optimal strategy. The method and the device have the advantages of simplifying user operation, improving service efficiency, improving reliability of backup data and stability of a backup system, reducing time delay of the backup system and ensuring stable operation of the service system.
In order to achieve the above object, the present invention provides a hybrid cloud data backup management method based on a multi-objective optimal copy management policy, including the following steps:
Defining a reliability objective function, a transmission delay objective function and a load balancing objective function of the data which are completely backed up in a plurality of storage nodes respectively;
Acquiring an optimized objective function of data backup management based on the reliability objective function, the transmission delay objective function and the load balancing objective function;
obtaining an optimal placement scheme matrix meeting the optimization objective function by adopting a multi-objective evolutionary algorithm based on decomposition;
and carrying out backup placement allocation on each storage node on the data based on the optimal placement scheme matrix.
Optionally, the reliability objective function is as follows:
wherein, Representing the reliability of the ith data file k i, χ j is the failure rate of the jth storage node T j in the storage cluster T, the usage time of the u storage nodes,/>For the failure rate function of storage node T j, K represents a data file, m is the number of storage nodes, T represents a cluster of m storage nodes, and tk ij represents a copy on node T j.
Optionally, the transmission delay objective function is as follows:
Wherein v j is the data transmission rate of the backup node t j; c i is the size of the data file k i, Representing the total access amount of k i on node t j; /(I)Representing the transmission delay of the a-th access on storage node t j for the backup copy of k i.
Optionally, the load balancing objective function is as follows:
Wherein σ j is the weight of each backup node t j, an
Optionally, the optimization objective function is as follows:
max S(X)＝[f1(X),f2(X),f3(X)]T
Where X is a decision variable, here a placement scheme matrix TK, f 1(X)＝s1 (TK), Ρ (k i) represents the size of the i-th data file k i, Γ (t j) represents the capacity size of the j-th storage node t j.
Optionally, the process of obtaining the optimal placement scheme matrix meeting the optimization objective function by adopting a decomposition-based multi-objective evolutionary algorithm comprises the following steps:
Initializing the population quantity and randomly generating weight vectors which are the same as the population quantity;
Decomposing the optimized objective function into a plurality of sub-problems with the same population number based on Tchebycheff decomposition algorithm;
Initializing the evolution times, and randomly generating a placement scheme matrix with the same population quantity;
initializing a front edge solution set and optimizing a reference point corresponding to an objective function;
initializing a neighborhood vector, obtaining a plurality of weight vectors nearest to the last weight vector, and constructing an index set based on the plurality of weight vectors;
setting the maximum iteration times, carrying out iteration update on the last weight vector and the corresponding placement scheme matrix based on the iteration times, stopping iteration when the maximum iteration times are reached, and obtaining the optimal placement scheme matrix.
Optionally, the updating the weight vector and the placement scheme matrix based on the iteration number includes:
Randomly acquiring an index number based on the index set, and acquiring a placement scheme matrix corresponding to the index number;
Acquiring a new placement scheme matrix based on a corresponding placement scheme matrix, and respectively carrying out iterative updating on reference points, populations and non-dominant solution sets based on the new placement scheme matrix;
When the maximum iteration times are reached, an updated non-dominant solution set is obtained and used as an optimal front solution set, a balance solution is selected in the optimal front solution set, and the balance solution is used as an optimal placement scheme matrix.
Optionally, during the updating of the reference point, ifThen Reference point representing sub-problem f k,/>Representing a new placement plan matrix; in the updating process of the population, for each neighborhood individual/>, in the index setIf/>Then/>Where g te represents the sub-problem based on Tchebycheff decomposition algorithm, λ represents the neighborhood individualIs used for the weight vector of (a).
Optionally, in the updating process of the non-dominant solution set, the dominant in INL (h-1) is reservedTo discard all quilt/>, in INL (h-1)Dominant solution, if all INL (h-1) solutions do not dominant/>Will/>Adding INL (h), wherein INL (h-1) represents the last updated non-dominant solution set,/>Representing the new placement plan matrix, INL (h) represents the current non-dominant solution set.
In order to achieve the above object, the present invention provides a hybrid cloud data backup management system based on a multi-objective optimal copy management policy, which is characterized in that the system includes a memory, a processor, and a computer program stored on the memory and capable of running on the processor, and the processor implements the hybrid cloud data backup management method based on the multi-objective optimal copy management policy when executing the computer program.
Compared with the prior art, the hybrid cloud data backup recovery management method based on the multi-objective optimal copy management strategy has the advantages that reliability of backup data, time delay of a backup system and stability of the backup system are comprehensively considered, and an optimal placement scheme of the backup copy of the gold copy is obtained through a multi-objective optimization algorithm based on decomposition, so that risks of data loss and damage are reduced. The user can set the backup recovery strategy through the invention, can flexibly and rapidly carry out data backup and recovery, help the administrator to rapidly locate faults, and furthest improve the efficiency and performance of the system.
Drawings
Various other advantages and benefits will become apparent to those of ordinary skill in the art upon reading the following detailed description of the preferred embodiments. The drawings are only for purposes of illustrating the preferred embodiments and are not to be construed as limiting the invention. Also, like reference numerals are used to designate like parts throughout the figures. In the drawings:
FIG. 1 is a flowchart of a multi-objective optimal copy management method according to an embodiment of the present invention;
Fig. 2 is a schematic diagram of a design principle of a hybrid cloud data backup and recovery management system based on a multi-objective optimal copy management strategy in an embodiment of the present invention;
FIG. 3 is a schematic diagram of a hybrid cloud data backup and restoration management system based on a multi-objective optimal copy management strategy in an embodiment of the present invention;
FIG. 4 is a schematic diagram of a multi-objective backup copy management strategy design based on the idea of CDM golden copy in an embodiment of the invention;
FIG. 5 is a roadmap of a multi-objective optimal copy management method in an embodiment of the invention.
Detailed Description
Exemplary embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. While exemplary embodiments of the present disclosure are shown in the drawings, it should be understood that the present disclosure may be embodied in various forms and should not be limited to the embodiments set forth herein. Rather, these embodiments are provided so that this disclosure will be thorough and complete, and will fully convey the scope of the disclosure to those skilled in the art. It should be noted that, without conflict, the embodiments of the present invention and features of the embodiments may be combined with each other. The invention will be described in detail below with reference to the drawings in connection with embodiments.
1-5, In this embodiment, a hybrid cloud data backup management method and system based on a multi-objective optimal copy management policy are provided, where the method is shown in FIG. 1, and includes:
S100: defining a reliability objective function, a transmission delay objective function and a load balancing objective function of the data which are completely backed up in a plurality of storage nodes respectively;
S200: acquiring an optimized objective function of data backup management based on the reliability objective function, the transmission delay objective function and the load balancing objective function;
s300: obtaining an optimal placement scheme matrix meeting the optimization objective function by adopting a multi-objective evolutionary algorithm based on decomposition;
S400: and carrying out backup placement allocation on each storage node on the data based on the optimal placement scheme matrix.
The working principle of the system is shown in fig. 2, and the system comprises:
step 1: user and rights management settings. User authentication and authority management functions are realized, and full life cycle management is performed on the backup recovery system.
Step 1.1: the user authentication function is implemented. When a user logs in, the account name and password at the user are required to be compared with the account name and password stored in the system, if the account name and password are matched with the account name and password, the user is indicated to pass authentication, otherwise, the user is refused to access and the authentication is required to be carried out again.
Step 1.2: the rights management function is implemented. And setting role authorities for different users, such as common users, operation and maintenance administrators and the like, and setting authorities for different user roles.
Step 2: and (5) managing backup strategies. And setting a backup mode and a backup strategy for scheduling the operation of various backup tasks.
Step 2.1: and setting a backup mode. The invention supports two backup modes of real-time backup and timing backup. The user can select the corresponding backup mode according to different requirements.
Step 2.2: and setting a backup strategy. The invention supports three backup strategies of full backup, incremental backup and differential backup. When the user performs the first backup, the full backup is performed by default. And (5) carrying out subsequent backup operation, wherein a user can select a backup strategy according to the requirement.
Step 3: and (5) backing up data. And starting data backup, establishing encryption connection between the data node and the backup node, transmitting data, and performing data backup according to the backup mode applied by the data node. If the user selects backup to the third party backup software, the step is transferred to step 3.1; if the user selects backup to the bottom storage space of the system, the process goes to step 3.2.
Step 3.1: the backup recovery management system is in butt joint with the third party backup software, and directly backups the third party backup software storage space according to the backup strategy set by the user.
Step 3.2: the bottom storage space of the system adopts a multi-objective optimal backup copy management scheme. Based on the idea of CDM golden copy, the steps and principles of designing a multi-target backup copy management strategy are respectively shown in FIG. 3 and FIG. 4:
for a cluster T, t= (T 1,t2,…,tm) consisting of m heterogeneous and independent storage nodes, the user backs up its data files K, k= (K 1,k2,…,kn) into the storage system.
The same copy cannot be placed on the same storage node, and the placement scheme of the data copy in the backup node is represented by a matrix TK:
wherein,
Step 3.2.1 as shown in fig. 5, from three aspects of backup system reliability, time delay and load balancing, setting corresponding objective optimization functions:
Step 3.2.1.1 reliability objective function design. The failure rate of the jth storage node T j in the storage cluster T is χ j, and the failure rate of the storage node is related to the use time u (u is in hours), the failure rate function of the storage node T j is Data files K, k= (K 1,k2,…,kn), for each of which data files K i, if there is at least one full backup copy in m storage nodes, the i-th data file K i is considered to be reliable. The reliability of k i is defined as:
the whole system, the reliability objective function of the complete backup of the data file K is defined as:
Step 3.2.1.2, designing a time delay objective function. The total network access delay of the j-th storage node T j in the storage cluster T is recorded as tim_del j (including processing delay, queuing delay, sending delay and propagation delay). Considering the data file response time and transmission time, the transmission delay of the a-th access on storage node t j is the backup copy of k i:
Wherein v j is the data transmission rate of the backup node t j; c i is the size of the data file k i. The accesses of the copies of k i on different backup nodes are independent of each other, and the total access latency function of k i is defined as:
wherein, Representing the total access of k i on node t j. The overall system defines an average delay objective function for the data file K as:
Step 3.2.1.3 load balancing objective function design. Each backup storage node t j adopts a weighted polling method, and corresponding weight sigma j,σj is attached to the backup storage node t j according to the hardware bearing capacity of the backup storage node t j to meet the requirement The invention takes the absolute value of the actual load of all the storage nodes t j and the theoretical load difference calculated based on the hardware bearing capacity as the standard for measuring the load balance of the system. The smaller the standard deviation, the more balanced the load. Defining a load balancing objective function:
Step 3.2.2 defines a multi-objective copy management policy optimization objective function. When the backup is carried out, reliable cloud backup service is provided, time delay is guaranteed, and faults caused by too high load of a certain storage node are avoided. Based on this, an optimization objective function is defined:
Where X is called a decision variable, here a placement scheme matrix TK. The optimal placement scheme matrix can be searched through multi-objective optimization. f 1(X)＝s1 (TK), Ρ (k i) represents the size of the i-th data file k i, Γ (t j) represents the capacity size of the j-th storage node t j. The computational unit requirements of ρ (k i) and Γ (t j) remain the same. σ j represents the weight assigned to the storage node hardware according to its load-bearing capacity.
Step 3.2.3 solves the above multi-objective problem with a decomposition-based multi-objective evolutionary algorithm. The method comprises the following specific steps:
Step 3.2.3.1 initializing population size to P, randomly generating P weight vectors uniformly distributed by Tchebycheff decomposition algorithm P=1, 2, …, P, where/>L is the number of objective functions.
Step 3.2.3.2 decomposing the multi-objective problem consisting of l objectives into P sub-problems using Tchebycheff decomposition algorithm, each sub-problem being:
wherein, Z * is the reference point for the multi-objective function. /(I)Is the minimum value that establishes the sub-problem f k (X), representing the reference point of the sub-problem f k (X).
Step 3.2.3.3 initializes the evolution number h=0, randomly generating P placement scheme matrices { X 1(0),...,Xp(0),...XP (0) } from the feasible domain. Wherein X p (0) represents the p-th placement scheme matrix during initialization, and the feasible domain is the value range of each numerical value in the placement scheme matrix meeting the constraint condition.
Step 3.2.3.4 initializes Pareto front solution set INL (0) =Φ, initializes the reference point.
Step 3.2.3.5 initializing a neighborhood vector, finding a distance p-th weight vectorThe nearest Neb weight vectors and will be equal to/>The index set of the nearest Neb weight vectors is denoted B (P) = { P 1,p2,...,pNeb},pNeb =1, 2.
Step 3.2.3.6 sets the maximum iteration number H, starting from h=1, stopping the iteration when h=h is set, and for each weight vectorAnd its corresponding individuals/>P=1, 2, …, P performs an update operation according to the number of iterations h:
step 3.2.3.6.1 the evolutionary algorithm updates the individual:
Randomly selecting index numbers c and d from B (p), and finding out corresponding individuals And/>By/>And/>Generating a new solution/>
Where ζ is the evolution parameter.
Step 3.2.3.6.2 updates the reference point z *: for each component k=1, 2, l, ifThen/> The reference point for sub-problem f k is shown.
Step 3.2.3.6.3 updates the population: for each neighborhood individual in B (p)If it isThen/>Where g te represents a sub-problem based on Tchebycheff decomposition algorithm, λ represents the neighborhood individual/>Is used for the weight vector of (a).
Step 3.2.3.6.4 updates the non-dominant solution set INL (h): in INL (h), the retention of INL (h-1) dominatesIn rejection of all of the INL (h-1) quilt/>Dominant solution, if all INL (h-1) solutions are not dominantWill/>INL (h) was added.
If h=h, step 3.2.3.6.5, the obtained non-dominant solution set INL (H) is the Pareto optimal leading edge solution set of the placement scheme matrix TK, and the iteration is terminated, and the step 3.2.3.6 is performed; otherwise h=h+1, go to step 3.2.3.1
Step 3.2.3.6.6 selects a tradeoff solution TK optimal from the Pareto optimal solution set according to decision preference of a decision maker or randomly, wherein TK optimal is a placement scheme matrix meeting three targets of reliability, delay and load balancing. And completing the placement and distribution of the backup copies on each node according to the TK optimal.
Step 3.2.4: after the backup is completed, the backup system informs the management node that the data backup is completed, the management node generates a backup log through the backup information and files the backup log, and informs a user that the backup is completed.
Step 4: and (5) recovering the data. And the backup system receives the data recovery request, calculates proper backup data through a backup selection algorithm according to the recovery time point in the request, and performs data recovery. The method comprises the following specific steps:
Step 4.1: the backup node transmits the data to the recovery node. After the connection between the transmitting end and the receiving end is established, the transmitting end encrypts the data. And the receiving end receives the data, then carries out decryption operation, and carries out check value detection on the decrypted data. If the verification detection is passed, the data transmission is proved to be correct, and the data can be written into the disk. If error data occurs in the transmission process, the receiving end can request the transmitting end to retransmit the error data.
Step 4.2: after the data recovery is completed, the recovery node notifies the management node that the recovery is completed, and the management node generates a recovery log through the data recovery information and files the recovery log and notifies a user that the data recovery is completed.
Step 5: and (5) log management. A log file is respectively established for each backup task flow and recovery task flow, and information of the task flow is tracked: recording the period backup and recovery time, task error information and the like, and facilitating the searching of task records and error positioning.
Step 6: and (5) monitoring and managing. The backup recovery management system provides a unified monitoring large screen, monitors the whole backup recovery process, and can check the data backup recovery condition in real time to monitor the available capacity of the storage space at the bottom layer of the system and the states of all storage nodes.
The invention abstracts the reliability, time delay problem and load balancing problem of the backup system into a multi-objective optimization problem based on the thought of multi-objective optimization. The performance of the 3 aspects of the system is optimized simultaneously in the form of vectors. On the premise of keeping load balance, the reliability of the system is improved and the system time delay is minimized. The obtained Pareto optimal solution enables the system to achieve optimal performance under multiple targets, thereby improving performance of the system in multiple aspects.
Compared with the prior art, the invention has the advantages and effects that:
(1) The patent discloses a hybrid cloud data backup and restoration management system, which integrates different types of data and backup spaces in a system platform in a butt joint way. The user can operate on the platform through the management interface, and backups different types of data on the local/cloud end to the third party storage software or the bottom storage space of the system according to the requirements of the user, so that the user operation is simplified, and the service efficiency is improved.
(2) According to the hybrid cloud data backup and recovery management system based on the multi-objective optimal copy management strategy, reliability of backup data, time delay of the backup system and stability of the backup system are comprehensively considered, and an optimal placement scheme of the backup copy of the gold copy is obtained through a multi-objective optimization algorithm based on decomposition, so that risks of data loss and damage are reduced.
(3) According to the hybrid cloud data backup and recovery management system based on the multi-target optimal copy management strategy, which is disclosed by the invention, the management is performed through a graphical interface, log information is recorded in real time, a unified monitoring platform is provided, a user can set the backup and recovery strategy, the data backup and recovery can be flexibly and rapidly performed, an administrator is helped to rapidly locate faults, and the efficiency and performance of the system are improved to the greatest extent.
Finally, it should be noted that: the above embodiments are only for illustrating the technical aspects of the present invention and not for limiting the same, and although the present invention has been described in detail with reference to the above embodiments, it should be understood by those of ordinary skill in the art that: modifications and equivalents may be made to the specific embodiments of the invention without departing from the spirit and scope of the invention, which is intended to be covered by the claims.
Claims (10)
1. The data backup management method based on the multi-objective optimal copy management strategy is characterized by comprising the following steps of:
Defining a reliability objective function, a transmission delay objective function and a load balancing objective function of the data which are completely backed up in a plurality of storage nodes respectively;
Acquiring an optimized objective function of data backup management based on the reliability objective function, the transmission delay objective function and the load balancing objective function;
obtaining an optimal placement scheme matrix meeting the optimization objective function by adopting a multi-objective evolutionary algorithm based on decomposition;
and carrying out backup placement allocation on each storage node on the data based on the optimal placement scheme matrix.
2. The method for managing data backup based on a multi-objective optimal copy management strategy according to claim 1, wherein the reliability objective function is as follows:
wherein, Representing the reliability of the ith data file k i, χ j is the failure rate of the jth storage node T j in the storage cluster T, the usage time of the u storage nodes,/>For the failure rate function of storage node T j, K represents a data file, m is the number of storage nodes, T represents a cluster of m storage nodes, and tk ij represents a copy on node T j.
3. The data backup management method based on the multi-objective optimal copy management policy according to claim 1, wherein the transmission delay objective function is as follows:
Wherein v j is the data transmission rate of the backup node t j; c i is the size of the data file k i, Representing the total access amount of k i on node t j; /(I)Representing the transmission delay of the a-th access on storage node t j for the backup copy of k i.
4. The method for managing data backup based on a multi-objective optimal copy management strategy according to claim 1, wherein the load balancing objective function is as follows:
Wherein σ j is the weight of each backup node t j, an
5. The method for managing data backup based on a multi-objective optimal copy management strategy according to claim 1, wherein the optimization objective function is as follows:
max S(X)＝[f1(X),f2(X),f3(X)]T
Where X is a decision variable, here a placement scheme matrix TK, f 1(X)＝s1 (TK), Ρ (k i) represents the size of the i-th data file k i, Γ (t j) represents the capacity size of the j-th storage node t j.
6. The method for managing data backup based on a multi-objective optimal copy management strategy according to claim 1, wherein the process of obtaining an optimal placement scheme matrix satisfying the optimization objective function by using a decomposition-based multi-objective evolutionary algorithm comprises:
Initializing the population quantity and randomly generating weight vectors which are the same as the population quantity;
Decomposing the optimized objective function into a plurality of sub-problems with the same population number based on Tchebycheff decomposition algorithm;
Initializing the evolution times, and randomly generating a placement scheme matrix with the same population quantity;
initializing a front edge solution set and optimizing a reference point corresponding to an objective function;
initializing a neighborhood vector, obtaining a plurality of weight vectors nearest to the last weight vector, and constructing an index set based on the plurality of weight vectors;
setting the maximum iteration times, carrying out iteration update on the last weight vector and the corresponding placement scheme matrix based on the iteration times, stopping iteration when the maximum iteration times are reached, and obtaining the optimal placement scheme matrix.
7. The method for managing data backup based on the multi-objective optimal copy management strategy of claim 6, wherein updating each weight vector and the corresponding placement scheme matrix based on the number of iterations comprises:
Randomly acquiring an index number based on the index set, and acquiring a placement scheme matrix corresponding to the index number;
Acquiring a new placement scheme matrix based on a corresponding placement scheme matrix, and respectively carrying out iterative updating on reference points, populations and non-dominant solution sets based on the new placement scheme matrix;
When the maximum iteration times are reached, an updated non-dominant solution set is obtained and used as an optimal front solution set, a balance solution is selected in the optimal front solution set, and the balance solution is used as an optimal placement scheme matrix.
8. The hybrid cloud data backup management method based on the multi-objective optimal copy management policy of claim 7, wherein, in the updating process of the reference point, ifThen/> Reference point representing sub-problem f k,/>Representing a new placement plan matrix;
In the updating process of the population, for each neighborhood individual in the index set If it isThen/>Where g te represents a sub-problem based on Tchebycheff decomposition algorithm, λ represents the neighborhood individual/>Is used for the weight vector of (a).
9. The hybrid cloud data backup management method based on a multi-objective optimal copy management policy of claim 7, wherein in the updating process of the non-dominant solution set, dominant in the reserved INL (h-1) is maintainedTo discard all quilt/>, in INL (h-1)Dominant solution, if all INL (h-1) solutions do not dominant/>Will/>Adding INL (h), wherein INL (h-1) represents the last updated non-dominant solution set,/>Representing the new placement plan matrix, INL (h) represents the current non-dominant solution set.
10. A hybrid cloud data backup management system based on a multi-objective optimal copy management strategy, characterized in that the system comprises a memory, a processor and a computer program stored on the memory and capable of running on the processor, which processor implements the method of any of the preceding claims 1 to 9 when executing the computer program.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311701095.5A CN117931513A (en) | 2023-12-12 | 2023-12-12 | Mixed cloud data backup management method based on multi-objective optimal copy management strategy |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311701095.5A CN117931513A (en) | 2023-12-12 | 2023-12-12 | Mixed cloud data backup management method based on multi-objective optimal copy management strategy |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN117931513A true CN117931513A (en) | 2024-04-26 |
Family
ID=90769325
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202311701095.5A Pending CN117931513A (en) | 2023-12-12 | 2023-12-12 | Mixed cloud data backup management method based on multi-objective optimal copy management strategy |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117931513A (en) |
-
2023
- 2023-12-12 CN CN202311701095.5A patent/CN117931513A/en active Pending
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US20210081383A1 (en) | Lifecycle support for storage objects | |
| US10642654B2 (en) | Storage lifecycle pipeline architecture | |
| US7647327B2 (en) | Method and system for implementing storage strategies of a file autonomously of a user | |
| KR101547719B1 (en) | Maintaining data integrity in data servers across data centers | |
| US9355060B1 (en) | Storage service lifecycle policy transition management | |
| US8504517B2 (en) | Systems and methods for selective data replication | |
| US20090125751A1 (en) | System and Method for Correlated Analysis of Data Recovery Readiness for Data Assets | |
| US20100241616A1 (en) | Perpetual archival of data | |
| US20070094323A1 (en) | Managed resource sharing method and apparatus | |
| CN103713951A (en) | Managing execution of programs by multiple computing systems | |
| JP2006268846A (en) | System and method of efficient data backup in networking environment | |
| EP3513296B1 (en) | Hierarchical fault tolerance in system storage | |
| US20190138402A1 (en) | Optimizing database migration in high availability and disaster recovery computing environments | |
| US11210150B1 (en) | Cloud infrastructure backup system | |
| US20070091809A1 (en) | Managed network resource sharing and optimization method and apparatus | |
| CN106101213A (en) | Information-distribution type storage method | |
| US9135002B1 (en) | Systems and methods for recovering an application on a computing device | |
| Alshayeji et al. | A study on fault tolerance mechanisms in cloud computing | |
| CN105827744A (en) | Data processing method of cloud storage platform | |
| US10664190B1 (en) | Geographically dispersed data protection and replication | |
| Sun et al. | Modelling and evaluating a high serviceability fault tolerance strategy in cloud computing environments | |
| CN117931513A (en) | Mixed cloud data backup management method based on multi-objective optimal copy management strategy | |
| US11188393B1 (en) | Systems and methods for performing load balancing and distributed high-availability | |
| US20080235392A1 (en) | Network file system | |
| CN111767251A (en) | Alliance chain beneficial to large file storage |
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 |