CN112579280B - Cloud resource scheduling method and device and computer storage medium - Google Patents
Cloud resource scheduling method and device and computer storage medium Download PDFInfo
- Publication number
- CN112579280B CN112579280B CN202011605368.2A CN202011605368A CN112579280B CN 112579280 B CN112579280 B CN 112579280B CN 202011605368 A CN202011605368 A CN 202011605368A CN 112579280 B CN112579280 B CN 112579280B
- Authority
- CN
- China
- Prior art keywords
- scheduling
- scheduling scheme
- candidate
- scheme
- tasks
- 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.)
- Active
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F9/00—Arrangements for program control, e.g. control units
- G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
- G06F9/46—Multiprogramming arrangements
- G06F9/48—Program initiating; Program switching, e.g. by interrupt
- G06F9/4806—Task transfer initiation or dispatching
- G06F9/4843—Task transfer initiation or dispatching by program, e.g. task dispatcher, supervisor, operating system
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F9/00—Arrangements for program control, e.g. control units
- G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
- G06F9/46—Multiprogramming arrangements
- G06F9/50—Allocation of resources, e.g. of the central processing unit [CPU]
- G06F9/5061—Partitioning or combining of resources
- G06F9/5072—Grid computing
Landscapes
- Engineering & Computer Science (AREA)
- Software Systems (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Mathematical Physics (AREA)
- Mobile Radio Communication Systems (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The embodiment of the application discloses a scheduling method and device of cloud resources and a computer storage medium, and belongs to the technical field of cloud computing. The method comprises the following steps: when a resource scheduling request is received, performing scheduling scheme initialization processing according to a plurality of tasks in the resource scheduling request and the number of the tasks corresponding to each task to obtain a reference scheduling scheme set, wherein the reference scheduling scheme set comprises a plurality of scheduling schemes corresponding to each task; performing optimization selection processing on the reference scheduling scheme set to obtain a target scheduling scheme set, wherein the target scheduling scheme set comprises a scheduling scheme corresponding to each task; and scheduling the plurality of tasks according to the scheduling scheme in the target scheduling scheme set. According to the embodiment of the application, the reference scheduling scheme can be optimized, selected and processed, so that the task completion time is shortened, and the cloud resource scheduling efficiency and the cloud service quality are improved.
Description
Technical Field
The embodiment of the application relates to the technical field of cloud computing, in particular to a cloud resource scheduling method and device and a computer storage medium.
Background
With the gradual development of cloud computing technology, the cloud computing technology is applied more and more widely in various aspects, and due to uncertainty of cloud resources and complexity of task scheduling, the service quality of a user and waste of partial cloud resources may be affected, so that effective scheduling of cloud resources is generally required to improve the waste of cloud resources and improve the service quality of the user.
Currently, cloud resources can generally be scheduled through heuristic algorithms. However, the heuristic algorithm has the problems of low convergence speed, easy falling into local optimization and the like, so that the acquired scheduling resources cannot quickly complete the task requested by the terminal, and the cloud service quality is reduced.
Disclosure of Invention
The embodiment of the application provides a scheduling method and device of cloud resources and a computer storage medium, which can be used for solving the problems of low resource scheduling efficiency, low task processing speed and low cloud service quality in the related technology. The technical scheme is as follows:
in one aspect, a method for scheduling cloud resources is provided, where the method includes:
when a resource scheduling request is received, performing scheduling scheme initialization processing according to a plurality of tasks in the resource scheduling request and the number of the tasks corresponding to each task to obtain a reference scheduling scheme set, wherein the reference scheduling scheme set comprises a plurality of scheduling schemes corresponding to each task;
performing optimization selection processing on the reference scheduling scheme set to obtain a target scheduling scheme set, wherein the target scheduling scheme set comprises a scheduling scheme corresponding to each task;
and scheduling the cloud resources for the plurality of tasks according to the scheduling scheme in the target scheduling scheme set.
In some embodiments, when receiving a resource scheduling request, performing scheduling scheme initialization processing according to a plurality of tasks in the resource scheduling request and a quantity corresponding to each task in the plurality of tasks to obtain a reference scheduling scheme set, includes:
when the resource scheduling request is received, acquiring the task number of the tasks, the number of the currently running virtual machines and a reference scheduling scheme;
according to the number of the tasks of the plurality of tasks and the number of the virtual machines, identifying the reference scheduling scheme through gene coding;
and constructing the reference scheduling scheme set according to the gene codes corresponding to the reference scheduling schemes.
In some embodiments, the performing optimization selection processing on the reference scheduling scheme set to obtain a target scheduling scheme set includes:
determining a first scheduling set and a second scheduling set in the reference scheduling scheme set, wherein the efficiency of processing tasks by the scheduling schemes in the first scheduling set is greater than the efficiency of processing tasks by the scheduling schemes in the second scheduling set;
determining a candidate scheduling scheme set from a scheduling scheme resource set through a differential evolution algorithm, wherein the scheduling scheme resource set is a set of all scheduling schemes capable of processing the plurality of tasks;
selecting an optimal scheduling scheme from the candidate scheduling scheme set;
updating the reference scheduling scheme by the optimal scheduling scheme;
determining an update number of times to update the reference scheduling scheme;
when the updating times are less than or equal to a time threshold value, returning to the operation of determining a first scheduling set and a second scheduling set in the reference scheduling scheme set until the updating times are greater than the time threshold value;
and when the updating times are larger than the time threshold value, acquiring a target scheduling scheme set from the reference scheduling scheme.
In some embodiments, the selecting an optimal scheduling scheme from the set of candidate scheduling schemes includes:
determining, by a classifier, a label of each candidate scheduling scheme in the set of candidate scheduling schemes, where the label is used to distinguish between the merits of each candidate scheduling scheme;
determining an average execution time of candidate scheduling schemes marked as a first identifier, wherein the first identifier is used for identifying a corresponding scheduling scheme as an excellent scheduling scheme;
and determining the candidate scheduling scheme with the average execution time meeting the time condition as the optimal scheduling scheme.
In some embodiments, before determining the average execution time of the candidate scheduling schemes labeled as the first identifier, the method further comprises:
determining cosine similarity between the candidate scheduling scheme labeled as the first identifier and a corresponding adjacent candidate scheduling scheme through a cosine similarity measurement method, wherein the corresponding adjacent candidate scheduling scheme is a scheduling scheme which is located in the reference scheduling scheme set and has a gene code adjacent to the gene code labeled as the candidate scheduling scheme of the first identifier;
setting a reference execution time as the time condition, where the reference execution time is an average execution time of the candidate scheduling schemes adjacent to the candidate scheduling schemes labeled as the first identifier and having a largest cosine similarity with the candidate scheduling schemes labeled as the first identifier.
In another aspect, an apparatus for scheduling cloud resources is provided, the apparatus including:
the device comprises an initialization module, a scheduling module and a scheduling module, wherein the initialization module is used for initializing a scheduling scheme according to a plurality of tasks in a resource scheduling request and the number of each task in the plurality of tasks when the resource scheduling request is received, so as to obtain a reference scheduling scheme set, and the reference scheduling scheme set comprises a plurality of scheduling schemes corresponding to each task;
the optimization module is used for carrying out optimization selection processing on the reference scheduling scheme set to obtain a target scheduling scheme set, wherein the target scheduling scheme set comprises scheduling schemes corresponding to the tasks;
and the scheduling module is used for scheduling the plurality of tasks according to the scheduling scheme in the target scheduling scheme set.
In some embodiments, the initialization module comprises:
the first obtaining submodule is used for obtaining the task quantity of the tasks, the quantity of the currently running virtual machines and a reference scheduling scheme when the resource scheduling request is received;
the identification submodule is used for identifying the reference scheduling scheme through gene coding according to the number of the tasks and the number of the virtual machines;
and the construction submodule is used for constructing the reference scheduling scheme set according to the gene codes corresponding to the reference scheduling schemes.
In some embodiments, the optimization module comprises:
the first determining submodule is used for determining a first scheduling set and a second scheduling set in the reference scheduling scheme set, and the efficiency of processing tasks by the scheduling schemes in the first scheduling set is greater than the efficiency of processing tasks by the scheduling schemes in the second scheduling set;
a second determining submodule, configured to determine a candidate scheduling scheme set from a scheduling scheme resource set through a differential evolution algorithm, where the scheduling scheme resource set is a set of all scheduling schemes capable of processing the plurality of tasks;
a selection submodule, configured to select an optimal scheduling scheme from the candidate scheduling scheme set;
an updating submodule, configured to update the reference scheduling scheme by the optimal scheduling scheme;
a third determining submodule, configured to determine an update number of times to update the reference scheduling scheme;
the triggering submodule is used for triggering the first determining submodule to determine a first scheduling set and a second scheduling set in the reference scheduling scheme set when the updating times are smaller than or equal to a time threshold value until the updating times are larger than the time threshold value;
and the second obtaining submodule is used for obtaining a target scheduling scheme set from the reference scheduling scheme when the updating times are greater than the time threshold.
In some embodiments, the selection submodule is to:
determining, by a classifier, a label of each candidate scheduling scheme in the set of candidate scheduling schemes, where the label is used to distinguish between the merits of each candidate scheduling scheme;
determining an average execution time of candidate scheduling schemes marked as a first identifier, wherein the first identifier is used for identifying a corresponding scheduling scheme as an excellent scheduling scheme;
and determining the candidate scheduling scheme with the average execution time meeting the time condition as the optimal scheduling scheme.
In some embodiments, the selection submodule is to:
determining cosine similarity between a candidate scheduling scheme labeled as the first identifier and a corresponding adjacent candidate scheduling scheme through a cosine similarity measurement method, wherein the corresponding adjacent candidate scheduling scheme is a scheduling scheme which is located in the reference scheduling scheme set and has a gene code adjacent to a gene code labeled as the candidate scheduling scheme of the first identifier;
setting a reference execution time as the time condition, where the reference execution time is an average execution time of the candidate scheduling schemes adjacent to the candidate scheduling schemes labeled as the first identifier and having a largest cosine similarity with the candidate scheduling schemes labeled as the first identifier.
In another aspect, a computer program product containing instructions is provided, which when run on a computer, causes the computer to execute the method for scheduling cloud resources according to the above aspect.
The beneficial effects brought by the technical scheme provided by the embodiment of the application at least comprise:
in the embodiment of the application, after the reference scheduling scheme is obtained, the reference scheduling scheme can be optimized and selected to obtain a target scheduling scheme set. The target scheduling scheme set is obtained by pre-selecting the reference scheduling scheme, so that the scheduling strategy is optimized, the task completion time is shortened, and the cloud resource scheduling efficiency and the cloud service quality are improved.
Drawings
In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings needed to be used in the description of the embodiments are briefly introduced below, and it is obvious that the drawings in the following description are only some embodiments of the present application, and it is obvious for those skilled in the art to obtain other drawings based on these drawings without creative efforts.
Fig. 1 is a flowchart of a scheduling method for cloud resources according to an embodiment of the present application;
fig. 2 is a flowchart of another scheduling method for cloud resources according to an embodiment of the present application;
fig. 3 is a schematic structural diagram of a scheduling apparatus of cloud resources according to an embodiment of the present application;
fig. 4 is a schematic structural diagram of an initialization module according to an embodiment of the present disclosure;
FIG. 5 is a schematic structural diagram of an optimization module provided in an embodiment of the present application;
fig. 6 is a schematic structural diagram of a scheduling apparatus according to an embodiment of the present application.
Detailed Description
To make the objects, technical solutions and advantages of the embodiments of the present application more clear, the embodiments of the present application will be further described in detail with reference to the accompanying drawings.
Before explaining the scheduling method of cloud resources provided in the embodiment of the present application in detail, an application scenario provided in the embodiment of the present application is introduced first.
With the rapid development of information technology and the increasing scale of the internet, the amount of traffic and data to be handled by the internet is rapidly increasing. In order to effectively process these massive data and services and optimize the user experience of using internet services, cloud computing technology has been developed. The key of the cloud computing technology is the scheduling of cloud resources. At present, cloud resources can be scheduled through a heuristic algorithm, but the heuristic algorithm has the problems of low convergence rate, easy falling into local optimum and the like, so that the acquired scheduling resources cannot quickly complete a task requested by a terminal, that is, the task is completed for a long time, and the cloud service quality is reduced.
Based on the application scenario, the embodiment of the application provides a cloud resource scheduling method capable of improving resource scheduling efficiency and shortening task completion time.
Fig. 1 is a flowchart of a cloud resource scheduling method provided in an embodiment of the present application, where the cloud resource scheduling method may include the following steps:
step 101: when a resource scheduling request is received, performing scheduling scheme initialization processing according to a plurality of tasks in the resource scheduling request and the number of the tasks corresponding to each task to obtain a reference scheduling scheme set, wherein the reference scheduling scheme set comprises a plurality of scheduling schemes corresponding to each task.
Step 102: and performing optimization selection processing on the reference scheduling scheme set to obtain a target scheduling scheme set, wherein the target scheduling scheme set comprises a scheduling scheme corresponding to each task.
Step 103: and scheduling the cloud resources for the plurality of tasks according to the scheduling scheme in the target scheduling scheme set.
In the embodiment of the application, after the reference scheduling scheme is obtained, the reference scheduling scheme can be optimized and selected to obtain a target scheduling scheme set. The target scheduling scheme set is obtained by pre-selecting the reference scheduling scheme, so that the scheduling strategy is optimized, the task completion time is shortened, and the cloud resource scheduling efficiency and the cloud service quality are improved.
In some embodiments, when a resource scheduling request is received, performing scheduling scheme initialization processing according to a plurality of tasks in the resource scheduling request and a quantity corresponding to each task in the plurality of tasks, to obtain a reference scheduling scheme set, including:
when the resource scheduling request is received, acquiring the task number of the tasks, the number of the currently running virtual machines and a reference scheduling scheme;
according to the number of the tasks of the plurality of tasks and the number of the virtual machines, identifying the reference scheduling scheme through gene coding;
and constructing the reference scheduling scheme set according to the gene codes corresponding to the reference scheduling schemes.
In some embodiments, performing optimization selection processing on the reference scheduling scheme set to obtain a target scheduling scheme set includes:
determining a first scheduling set and a second scheduling set in the reference scheduling scheme set, wherein the efficiency of processing tasks by the scheduling schemes in the first scheduling set is greater than the efficiency of processing tasks by the scheduling schemes in the second scheduling set;
determining a candidate scheduling scheme set from a scheduling scheme resource set through a differential evolution algorithm, wherein the scheduling scheme resource set is a set of all scheduling schemes capable of processing the plurality of tasks;
selecting an optimal scheduling scheme from the candidate scheduling scheme set;
updating the reference scheduling scheme through the optimal scheduling scheme;
determining the number of updating times for updating the reference scheduling scheme;
when the updating times are less than or equal to the times threshold, returning to the operation of determining the first scheduling set and the second scheduling set in the reference scheduling scheme set until the updating times are greater than the times threshold;
and when the updating times are larger than the time threshold value, acquiring a target scheduling scheme set from the reference scheduling scheme.
In some embodiments, selecting the optimal scheduling scheme from the set of candidate scheduling schemes comprises:
determining a label of each candidate scheduling scheme in the candidate scheduling scheme set through a classifier, wherein the label is used for distinguishing the quality of each scheduling scheme;
determining an average execution time of candidate scheduling schemes marked as a first identifier, wherein the first identifier is used for identifying the corresponding scheduling scheme as an excellent scheduling scheme;
and determining the candidate scheduling scheme with the average execution time meeting the time condition as the optimal scheduling scheme.
In some embodiments, before determining the average execution time of the candidate scheduling schemes labeled as the first identifier, further comprises:
determining cosine similarity between the candidate scheduling scheme labeled as the first identifier and a corresponding adjacent candidate scheduling scheme through a cosine similarity measurement method, wherein the corresponding adjacent candidate scheduling scheme is a scheduling scheme which is positioned in the reference scheduling scheme set and has a gene code adjacent to the gene code labeled as the candidate scheduling scheme of the first identifier;
setting a reference execution time as the time condition, wherein the reference execution time is an average execution time of the candidate scheduling schemes which are adjacent to the candidate scheduling scheme marked as the first identifier and have the largest cosine similarity with the candidate scheduling scheme marked as the first identifier.
All the above optional technical solutions can be combined arbitrarily to form an optional embodiment of the present application, and the present application embodiment is not described in detail again.
Fig. 2 is a flowchart of a scheduling method for cloud resources provided in an embodiment of the present application, where the scheduling method for cloud resources is applied to a scheduling device for example, the scheduling method for cloud resources may include the following steps:
step 201: the scheduling device receives a resource scheduling request.
It should be noted that the resource scheduling request can carry a plurality of tasks and the number of the tasks corresponding to each task.
Step 202: and the scheduling equipment performs scheduling scheme initialization processing according to the plurality of tasks in the resource scheduling request and the number corresponding to each task in the plurality of tasks to obtain a reference scheduling scheme set, wherein the reference scheduling scheme set comprises a plurality of scheduling schemes corresponding to each task.
As an example, the scheduling device performs scheduling scheme initialization processing according to a plurality of tasks in the resource scheduling request and a number corresponding to each task in the plurality of tasks, and the operation of obtaining the reference scheduling scheme set at least includes the following operations: when a resource scheduling request is received, acquiring the task number of a plurality of tasks, the number of currently-running virtual machines and a reference scheduling scheme; according to the number of tasks and the number of virtual machines of a plurality of tasks, identifying a reference scheduling scheme through gene coding; and constructing a reference scheduling scheme set according to the gene codes corresponding to the reference scheduling schemes.
Because the cloud resources in the scheduling device are rich, different scheduling schemes may exist for any task, and therefore, when the scheduling device receives a resource scheduling request, the scheduling device may obtain any reference scheduling scheme in advance, but the obtained reference scheduling scheme is not necessarily a scheduling scheme with high efficiency and capable of shortening the task completion time.
The reference scheduling scheme set is constructed by the gene coding corresponding to the reference scheduling scheme, so that the gene coding mode of each scheduling scheme in the reference scheduling scheme set is the same as the gene coding mode corresponding to the reference scheduling scheme.
As an example, the scheduling device can encode according to the encoding scheme shown in fig. 3, where t in fig. 3 i Identify task number, i ∈ [1,m]，r j Identify resource number, j belongs to [1,n]And according to the coding scheme shown in FIG. 3, obtain the corresponding decoding r 3 :{t 1 }，r 1 :{t 2 }，r 2 :{t 3 }，…，r n :{t m }。
Step 203: and the scheduling equipment performs optimization selection processing on the reference scheduling scheme set to obtain a target scheduling scheme set, wherein the target scheduling scheme set comprises a scheduling scheme corresponding to each task.
Because there are many scheduling schemes in the reference scheduling scheme set, the scheduling device needs to select a target scheduling scheme set from the reference scheduling scheme set, where the scheduling scheme in the target scheduling scheme set is a scheduling scheme with high resource processing efficiency.
As an example, the operation of the scheduling scheme performing optimization selection processing on the reference scheduling scheme set to obtain the target scheduling scheme set at least includes the following steps: determining a first scheduling set and a second scheduling set in a reference scheduling scheme set, wherein the efficiency of processing tasks by the scheduling schemes in the first scheduling set is greater than the efficiency of processing tasks by the scheduling schemes in the second scheduling set; determining a candidate scheduling scheme set from a scheduling scheme resource set through a differential evolution algorithm, wherein the scheduling scheme resource set is a set of all scheduling schemes capable of processing a plurality of tasks; selecting an optimal scheduling scheme from the candidate scheduling scheme set; updating the reference scheduling scheme through the optimal scheduling scheme; determining the number of updating times of the updated reference scheduling scheme; when the updating times are less than or equal to the time threshold, returning to the operation of determining the first scheduling set and the second scheduling set in the reference scheduling scheme set until the updating times are greater than the time threshold; and when the updating times are larger than the time threshold value, acquiring a target scheduling scheme set from the reference scheduling scheme.
In some embodiments, it is efficient that an existing scheduling scheme in the reference scheduling scheme set processes a plurality of tasks, and it is inefficient that an existing scheduling scheme processes a plurality of tasks, and therefore, in order to facilitate updating the reference scheduling scheme set, the scheduling apparatus can determine a first scheduling scheme set and a second scheduling scheme set in the reference scheduling scheme set, and the number of scheduling schemes in the first scheduling scheme set is the same as the number of scheduling schemes in the second scheduling scheme set.
In some embodiments, in order to distinguish between the scheduling schemes in the first set of scheduling schemes and the scheduling schemes in the second set of scheduling schemes, the scheduling schemes in the first set of scheduling schemes are labeled with a first identifier, and the scheduling schemes in the second set of scheduling schemes are labeled with a second identifier, wherein the labels are used for distinguishing the advantages and the disadvantages of each scheduling scheme.
It should be noted that the number threshold can be set in advance according to the requirement, for example, the number threshold can be 10 times, 30 times, 50 times, and so on. The first identifier can be 1, the second identifier can be-1, and so on.
As an example, the operation of the scheduling device selecting the optimal scheduling scheme from the candidate scheduling scheme set at least comprises the following operations: determining a label of each candidate scheduling scheme in the candidate scheduling scheme set through a classifier, wherein the label is used for distinguishing the advantages and the disadvantages of each scheduling scheme; determining an average execution time of candidate scheduling schemes marked as a first identifier, wherein the first identifier is used for identifying the corresponding scheduling scheme as an excellent scheduling scheme; and determining the candidate scheduling scheme with the average execution time meeting the time condition as the optimal scheduling scheme.
It should be noted that the classifier can be a k-neighbor classifier.
In some embodiments, the scheduler can set < x, y > as a set of training data, x is a candidate scheduling scheme in the set of candidate scheduling schemes, y is a set of labels to which the candidate scheduling scheme in the set of candidate scheduling schemes corresponds, and y ∈ L, L = { +1, -1}. The correspondence between the candidate scheduling schemes and the labels can be expressed as y = Class (x).
In some embodiments, the training process for label y can be represented by the following first formula.
It should be noted that, in the first formula (1) above, the training data of the k-neighbor classifier includes the current entire reference scheduling scheme set. N (x) represents K nearest neighbor individuals in the training data, which are nearest to x, and K can be set in advance according to requirements, for example, K is set to 3.
In some embodiments, sign (x) can be performed as: a scheduling scheme is selected from N (x), and the tag values of the selected scheduling schemes are added. The result is 1, which indicates an excellent scheduling scheme, otherwise, a bad scheduling scheme.
In some embodiments, since the candidate scheduling set may include a plurality of excellent scheduling schemes after being classified by the classifier, it is necessary to further determine an optimal scheduling scheme set from the plurality of excellent scheduling schemes.
Since each specific task completion time is from task submission until task calculation completion is fed back to the user. Generally, the task completion time is defined as the time taken by the task with the longest completion time among all tasks, however, the time taken by the task with the longest completion time cannot reflect the advantages of the cloud resource scheduling policy as a whole, and therefore, the scheduling device needs to determine the average execution time of the tasks.
As an example, the terminal can determine the average execution time by the following second formula.
In the second equation (2), RT is i,j For the theoretical running time, TC, of the current task i on the virtual resource j i，j =1 denotes that the current task i runs on virtual resource j, TC i，j =0 indicates that the current task i is not running on the virtual resource j.
In some embodiments, after the scheduling apparatus determines the label of each candidate scheduling scheme in the candidate scheduling scheme set through the classifier, in order to further determine the quality of each candidate scheduling scheme, before the scheduling apparatus determines the average execution time of the candidate scheduling scheme labeled as the first identifier, the scheduling apparatus may further determine, through a cosine similarity measurement method, a cosine similarity between the candidate scheduling scheme labeled as the first identifier and a corresponding adjacent candidate scheduling scheme, where the corresponding adjacent candidate scheduling scheme is a scheduling scheme located in the reference scheduling scheme set and having a gene code adjacent to the gene code labeled as the candidate scheduling scheme of the first identifier; and setting a reference execution time as a time condition, wherein the reference execution time is the average execution time of the candidate scheduling scheme which is adjacent to the candidate scheduling scheme marked as the first identifier and has the largest cosine similarity with the candidate scheduling scheme marked as the first identifier.
As an example, the scheduling apparatus can determine a cosine similarity between the candidate scheduling scheme labeled as the first identification and the corresponding neighbor candidate scheduling scheme by the following third formula.
In the third formula (3), x is 1 And x 2 Represents two scheduling schemes with the same dimension in vector space, and sim (x) 1 ，x 2 ) The closer to 1,x 1 And x 2 The more similar.
In some embodiments, the time condition can also be set in advance according to requirements, for example, the time condition is set to be a time threshold, and the time threshold can be 5 seconds, 3 seconds, and the like. Accordingly, the average execution time satisfying the time condition at this time means that the average execution time is less than or equal to the time threshold.
In some embodiments, the operation of the scheduling device updating the reference scheduling scheme by the optimal scheduling scheme at least comprises the following operations: and replacing any reference scheduling scheme in the second scheduling set in the reference scheduling scheme by the optimal scheduling scheme, and re-determining the first scheduling set and the second scheduling set in the updated reference scheduling scheme.
In some embodiments, since the efficiency of processing tasks by the scheduling schemes in the first scheduling set is greater than the efficiency of processing tasks by the scheduling schemes in the second scheduling set, the scheduling device can obtain the target scheduling scheme set from the first scheduling set of the reference scheduling scheme.
Step 204: and scheduling the cloud resources for the plurality of tasks by the scheduling equipment according to the scheduling scheme in the target scheduling scheme set.
As an example, the scheduling device can select any scheduling scheme from the target scheduling scheme set to schedule the cloud resources for the plurality of tasks.
The target scheduling scheme set can be obtained from the first scheduling set, so that the obtained scheduling scheme is guaranteed to be a scheduling scheme with high task processing efficiency, and the task processing time is shortened.
In the embodiment of the application, the scheduling device can determine the candidate scheduling scheme set from the scheduling scheme resource set through a differential evolution algorithm, and obtain the optimal scheduling scheme through a classifier and a cosine similarity algorithm, so that the times of determining the execution time is reduced, and the efficiency of determining the optimal scheduling scheme is improved. In addition, the target scheduling scheme set is obtained by performing pre-selection processing on the reference scheduling scheme, so that the scheduling strategy is optimized, the task completion time is shortened, and the cloud resource scheduling efficiency and the cloud service quality are improved.
Fig. 3 is a schematic structural diagram of a scheduling apparatus of cloud resources according to an embodiment of the present disclosure, where the scheduling apparatus of cloud resources may be implemented by software, hardware, or a combination of the software and the hardware. The scheduling apparatus of cloud resources may include: an initialization module 301, an optimization module 302, and a scheduling module 303.
An initialization module 301, configured to, when a resource scheduling request is received, perform scheduling scheme initialization processing according to a plurality of tasks in the resource scheduling request and a number corresponding to each task in the plurality of tasks, to obtain a reference scheduling scheme set, where the reference scheduling scheme set includes a plurality of scheduling schemes corresponding to each task;
an optimizing module 302, configured to perform optimization selection processing on the reference scheduling scheme set to obtain a target scheduling scheme set, where the target scheduling scheme set includes a scheduling scheme corresponding to each task;
and the scheduling module 303 is configured to schedule the cloud resources for the plurality of tasks according to a scheduling scheme in the target scheduling scheme set.
In some embodiments, referring to fig. 4, the initialization module 301 comprises:
a first obtaining sub-module 3011, configured to obtain, when the resource scheduling request is received, the number of tasks of the multiple tasks, the number of currently running virtual machines, and a reference scheduling scheme;
the identification submodule 3012 is configured to identify the reference scheduling scheme through a gene code according to the number of the tasks of the multiple tasks and the number of the virtual machines;
and the constructing submodule 3013 is configured to construct the reference scheduling scheme set according to the gene code corresponding to the reference scheduling scheme.
In some embodiments, referring to fig. 5, the optimization module 302 includes:
a first determining submodule 3021, configured to determine a first scheduling set and a second scheduling set in the reference scheduling scheme set, where efficiency of processing tasks by a scheduling scheme in the first scheduling set is greater than efficiency of processing tasks by a scheduling scheme in the second scheduling set;
a second determining submodule 3022, configured to determine a candidate scheduling scheme set from a scheduling scheme resource set through a differential evolution algorithm, where the scheduling scheme resource set is a set of all scheduling schemes capable of processing the plurality of tasks;
a selecting submodule 3023 configured to select an optimal scheduling scheme from the candidate scheduling scheme set;
an updating submodule 3024, configured to update the reference scheduling scheme by the optimal scheduling scheme;
a third determining sub-module 3025 configured to determine an update number of times to update the reference scheduling scheme;
a triggering sub-module 3026, configured to, when the update time is less than or equal to a time threshold, trigger the first determining sub-module 3021 to determine a first scheduling set and a second scheduling set in the reference scheduling scheme set until the update time is greater than the time threshold;
a second obtaining sub-module 3027, configured to obtain a target scheduling scheme set from the reference scheduling scheme when the update time is greater than the time threshold.
In some embodiments, the selection submodule 3023 is configured to:
determining, by a classifier, a label of each candidate scheduling scheme in the set of candidate scheduling schemes, where the label is used to distinguish between the merits of each candidate scheduling scheme;
determining an average execution time of candidate scheduling schemes marked as a first identifier, wherein the first identifier is used for identifying a corresponding scheduling scheme as an excellent scheduling scheme;
and determining the candidate scheduling scheme with the average execution time meeting the time condition as the optimal scheduling scheme.
In some embodiments, the selection submodule is to:
determining cosine similarity between the candidate scheduling scheme labeled as the first identifier and a corresponding adjacent candidate scheduling scheme through a cosine similarity measurement method, wherein the corresponding adjacent candidate scheduling scheme is a scheduling scheme which is located in the reference scheduling scheme set and has a gene code adjacent to the gene code labeled as the candidate scheduling scheme of the first identifier;
setting a reference execution time as the time condition, where the reference execution time is an average execution time of the candidate scheduling schemes adjacent to the candidate scheduling schemes labeled as the first identifier and having a largest cosine similarity with the candidate scheduling schemes labeled as the first identifier.
In the embodiment of the application, the scheduling device can determine the candidate scheduling scheme set from the scheduling scheme resource set through a differential evolution algorithm, and obtain the optimal scheduling scheme through a classifier and a cosine similarity algorithm, so that the times of determining the execution time is reduced, and the efficiency of determining the optimal scheduling scheme is improved. In addition, the target scheduling scheme set is obtained by performing pre-selection processing on the reference scheduling scheme, so that the scheduling strategy is optimized, the task completion time is shortened, and the cloud resource scheduling efficiency and the cloud service quality are improved.
It should be noted that: in the scheduling apparatus for cloud resources provided in the foregoing embodiment, when scheduling cloud resources, only the division of each functional module is described as an example, and in practical applications, the function allocation may be completed by different functional modules according to needs, that is, an internal structure of the apparatus is divided into different functional modules, so as to complete all or part of the functions described above. In addition, the scheduling apparatus for cloud resources and the scheduling method embodiment for cloud resources provided in the foregoing embodiments belong to the same concept, and specific implementation processes thereof are described in detail in the method embodiments, and are not described again here.
Fig. 6 is a schematic structural diagram illustrating a scheduling apparatus according to an exemplary embodiment. The scheduling device can be a server, and the server can be a server in a background server cluster. Specifically, the method comprises the following steps:
the server 600 includes a Central Processing Unit (CPU) 601, a system memory 604 including a Random Access Memory (RAM) 602 and a Read Only Memory (ROM) 603, and a system bus 605 connecting the system memory 604 and the central processing unit 601. The server 600 also includes a basic input/output system (I/O system) 606, which facilitates transfer of information between devices within the computer, and a mass storage device 607, which stores an operating system 613, application programs 614, and other program modules 615.
The basic input/output system 606 includes a display 608 for displaying information and an input device 609 such as a mouse, keyboard, etc. for user input of information. Wherein a display 608 and an input device 609 are connected to the central processing unit 601 through an input output controller 610 connected to the system bus 605. The basic input/output system 606 may also include an input/output controller 610 for receiving and processing input from a number of other devices, such as a keyboard, mouse, or electronic stylus. Similarly, input/output controller 610 may also provide output to a display screen, a printer, or other type of output device.
The mass storage device 607 is connected to the central processing unit 601 through a mass storage controller (not shown) connected to the system bus 605. The mass storage device 607 and its associated computer-readable media provide non-volatile storage for the server 700. That is, mass storage device 607 may include a computer-readable medium (not shown), such as a hard disk or CD-ROM drive.
Without loss of generality, computer readable media may comprise computer storage media and communication media. Computer storage media includes volatile and nonvolatile, removable and non-removable media implemented in any method or technology for storage of information such as computer readable instructions, data structures, program modules or other data. Computer storage media includes RAM, ROM, EPROM, EEPROM, flash memory or other solid state storage technology, CD-ROM, DVD, or other optical storage, magnetic cassettes, magnetic tape, magnetic disk storage or other magnetic storage devices. Of course, those skilled in the art will appreciate that computer storage media is not limited to the foregoing. The system memory 604 and mass storage device 607 described above may be collectively referred to as memory.
According to various embodiments of the present application, server 700 may also operate as a remote computer connected to a network via a network, such as the Internet. That is, the server 700 may be connected to the network 612 through the network interface unit 611 connected to the system bus 605, or may be connected to another type of network or a remote computer system (not shown) using the network interface unit 611.
The memory further includes one or more programs, and the one or more programs are stored in the memory and configured to be executed by the CPU. The one or more programs include a scheduling method for performing the cloud resource provided by the above embodiments.
The embodiment of the present application further provides a non-transitory computer-readable storage medium, and when instructions in the storage medium are executed by a processor of a server, the server is enabled to execute the scheduling method of cloud resources provided by the foregoing embodiment.
The embodiment of the present application further provides a computer program product containing instructions, which when run on a server, causes the server to execute the cloud resource scheduling method provided by the foregoing embodiment.
It will be understood by those skilled in the art that all or part of the steps for implementing the above embodiments may be implemented by hardware, or may be implemented by a program instructing relevant hardware, where the program may be stored in a computer-readable storage medium, and the above-mentioned storage medium may be a read-only memory, a magnetic disk or an optical disk, etc.
The above description is only a preferred embodiment of the present application and should not be taken as limiting the present application, and any modifications, equivalents, improvements, etc. made within the spirit and principle of the present application should be included in the protection scope of the present application.
Claims (7)
1. A scheduling method of cloud resources is characterized by comprising the following steps:
when a resource scheduling request is received, performing scheduling scheme initialization processing according to a plurality of tasks in the resource scheduling request and the number of the tasks corresponding to each task to obtain a reference scheduling scheme set, wherein the reference scheduling scheme set comprises a plurality of scheduling schemes corresponding to each task;
performing optimization selection processing on the reference scheduling scheme set to obtain a target scheduling scheme set, wherein the target scheduling scheme set comprises a scheduling scheme corresponding to each task;
scheduling the plurality of tasks according to a scheduling scheme in the target scheduling scheme set;
the performing optimization selection processing on the reference scheduling scheme set to obtain a target scheduling scheme set includes:
determining a first scheduling set and a second scheduling set in the reference scheduling scheme set, wherein the efficiency of processing tasks by the scheduling schemes in the first scheduling set is greater than the efficiency of processing tasks by the scheduling schemes in the second scheduling set;
determining a candidate scheduling scheme set from a scheduling scheme resource set through a differential evolution algorithm, wherein the scheduling scheme resource set is a set of all scheduling schemes capable of processing the plurality of tasks; selecting an optimal scheduling scheme from the candidate scheduling scheme set; updating the reference scheduling scheme by the optimal scheduling scheme; determining an update number of times to update the reference scheduling scheme;
when the updating times are less than or equal to a time threshold value, returning to the operation of determining a first scheduling set and a second scheduling set in the reference scheduling scheme set until the updating times are greater than the time threshold value; when the updating times are larger than the time threshold value, acquiring a target scheduling scheme set from the reference scheduling scheme;
the selecting an optimal scheduling scheme from the candidate scheduling scheme set includes:
determining, by a classifier, a label of each candidate scheduling scheme in the set of candidate scheduling schemes, where the label is used to distinguish whether each candidate scheduling scheme is good or bad; determining an average execution time of candidate scheduling schemes marked as a first identifier, wherein the first identifier is used for identifying a corresponding scheduling scheme as an excellent scheduling scheme; and determining the candidate scheduling scheme with the average execution time meeting the time condition as the optimal scheduling scheme.
2. The method of claim 1, wherein when receiving a resource scheduling request, performing scheduling scheme initialization processing according to a plurality of tasks in the resource scheduling request and a number corresponding to each task in the plurality of tasks to obtain a reference scheduling scheme set, comprises:
when the resource scheduling request is received, acquiring the task number of the tasks, the number of the currently running virtual machines and a reference scheduling scheme;
according to the number of the tasks of the plurality of tasks and the number of the virtual machines, identifying the reference scheduling scheme through gene coding;
and constructing the reference scheduling scheme set according to the gene codes corresponding to the reference scheduling schemes.
3. The method of claim 1, wherein prior to determining the average execution time of the candidate scheduling schemes labeled as the first identifier, further comprising:
determining cosine similarity between the candidate scheduling scheme labeled as the first identifier and a corresponding adjacent candidate scheduling scheme through a cosine similarity measurement method, wherein the corresponding adjacent candidate scheduling scheme is a scheduling scheme which is located in the reference scheduling scheme set and has a gene code adjacent to the gene code labeled as the candidate scheduling scheme of the first identifier;
setting a reference execution time as the time condition, where the reference execution time is an average execution time of the candidate scheduling schemes adjacent to the candidate scheduling schemes labeled as the first identifier and having a largest cosine similarity with the candidate scheduling schemes labeled as the first identifier.
4. An apparatus for scheduling cloud resources, the apparatus comprising:
the device comprises an initialization module, a scheduling module and a scheduling module, wherein the initialization module is used for initializing a scheduling scheme according to a plurality of tasks in a resource scheduling request and the number of each task in the plurality of tasks when the resource scheduling request is received, so as to obtain a reference scheduling scheme set, and the reference scheduling scheme set comprises a plurality of scheduling schemes corresponding to each task;
the optimization module is used for carrying out optimization selection processing on the reference scheduling scheme set to obtain a target scheduling scheme set, wherein the target scheduling scheme set comprises scheduling schemes corresponding to all the tasks;
the scheduling module is used for scheduling the plurality of tasks according to the scheduling scheme in the target scheduling scheme set;
the optimization module comprises:
the first determining submodule is used for determining a first scheduling set and a second scheduling set in the reference scheduling scheme set, and the efficiency of processing tasks by the scheduling schemes in the first scheduling set is greater than the efficiency of processing tasks by the scheduling schemes in the second scheduling set;
a second determining submodule, configured to determine a candidate scheduling scheme set from a scheduling scheme resource set through a differential evolution algorithm, where the scheduling scheme resource set is a set of all scheduling schemes capable of processing the plurality of tasks;
a selection submodule, configured to select an optimal scheduling scheme from the candidate scheduling scheme set;
an updating submodule, configured to update the reference scheduling scheme by the optimal scheduling scheme;
a third determining sub-module, configured to determine an update time for updating the reference scheduling scheme;
the triggering submodule is used for triggering the first determining submodule to determine a first scheduling set and a second scheduling set in the reference scheduling scheme set when the updating times are smaller than or equal to a times threshold value until the updating times are larger than the times threshold value;
a second obtaining sub-module, configured to obtain a target scheduling scheme set from the reference scheduling scheme when the update time is greater than the time threshold;
the selection submodule is configured to:
determining, by a classifier, a label of each candidate scheduling scheme in the set of candidate scheduling schemes, the label being used for distinguishing the superiority and inferiority of each candidate scheduling scheme; determining an average execution time of candidate scheduling schemes marked as a first identifier, wherein the first identifier is used for identifying a corresponding scheduling scheme as an excellent scheduling scheme; and determining the candidate scheduling scheme with the average execution time meeting the time condition as the optimal scheduling scheme.
5. The apparatus of claim 4, wherein the initialization module comprises:
the first obtaining submodule is used for obtaining the task quantity of the tasks, the quantity of the currently running virtual machines and a reference scheduling scheme when the resource scheduling request is received;
the identification submodule is used for identifying the reference scheduling scheme through gene coding according to the number of the tasks and the number of the virtual machines;
and the construction submodule is used for constructing the reference scheduling scheme set according to the gene codes corresponding to the reference scheduling scheme.
6. The apparatus of claim 4, wherein the selection submodule is further to:
determining cosine similarity between the candidate scheduling scheme labeled as the first identifier and a corresponding adjacent candidate scheduling scheme through a cosine similarity measurement method, wherein the corresponding adjacent candidate scheduling scheme is a scheduling scheme which is located in the reference scheduling scheme set and has a gene code adjacent to the gene code labeled as the candidate scheduling scheme of the first identifier;
setting a reference execution time as the time condition, where the reference execution time is an average execution time of the candidate scheduling schemes adjacent to the candidate scheduling schemes labeled as the first identifier and having a largest cosine similarity with the candidate scheduling schemes labeled as the first identifier.
7. A computer-readable storage medium having stored thereon instructions which, when executed by a processor, carry out the steps of the method of any of the preceding claims 1 to 3.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011605368.2A CN112579280B (en) | 2020-12-30 | 2020-12-30 | Cloud resource scheduling method and device and computer storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011605368.2A CN112579280B (en) | 2020-12-30 | 2020-12-30 | Cloud resource scheduling method and device and computer storage medium |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN112579280A CN112579280A (en) | 2021-03-30 |
| CN112579280B true CN112579280B (en) | 2022-11-29 |
Family
ID=75144237
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202011605368.2A Active CN112579280B (en) | 2020-12-30 | 2020-12-30 | Cloud resource scheduling method and device and computer storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112579280B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114416351B (en) * | 2021-12-29 | 2022-10-21 | 北京百度网讯科技有限公司 | Resource allocation method, device, equipment, medium and computer program product |
Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103345384A (en) * | 2013-07-18 | 2013-10-09 | 北京航空航天大学 | Multi-target test task scheduling method and platform based on decomposed variable neighborhoods |
| CN105740949A (en) * | 2016-03-03 | 2016-07-06 | 浙江工业大学 | Group global optimization method based on randomness best strategy |
| CN109522104A (en) * | 2018-10-15 | 2019-03-26 | 华南理工大学 | Utilize the method for differential evolution algorithm optimization two goal task of Iaas scheduling |
| CN109936141A (en) * | 2019-03-28 | 2019-06-25 | 广州番禺职业技术学院 | A kind of Economic Dispatch method and system |
| CN110147274A (en) * | 2019-05-21 | 2019-08-20 | 武汉轻工大学 | Multiple target cloud task balance dispatching method, server and storage medium |
| CN110162393A (en) * | 2019-05-30 | 2019-08-23 | 奇瑞汽车股份有限公司 | Method for scheduling task, device and storage medium |
| CN110321217A (en) * | 2019-03-25 | 2019-10-11 | 深圳大学 | A kind of cloud resource dispatching method, device, equipment and the storage medium of multiple target |
| CN110489229A (en) * | 2019-07-17 | 2019-11-22 | 长沙学院 | A kind of multiple target method for scheduling task and system |
| CN111343259A (en) * | 2020-02-17 | 2020-06-26 | 武汉轻工大学 | Binary code-based cloud task scheduling method, server and storage medium |
Family Cites Families (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US8621414B2 (en) * | 2010-04-21 | 2013-12-31 | National Cheng Kung University | Method and algorithm analyzer for determining a design framework |
| CN102232282B (en) * | 2010-10-29 | 2014-03-26 | 华为技术有限公司 | Method and apparatus for realizing load balance of resources in data center |
| AU2016203619A1 (en) * | 2016-05-31 | 2017-12-14 | Canon Kabushiki Kaisha | Layer-based operations scheduling to optimise memory for CNN applications |
-
2020
- 2020-12-30 CN CN202011605368.2A patent/CN112579280B/en active Active
Patent Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103345384A (en) * | 2013-07-18 | 2013-10-09 | 北京航空航天大学 | Multi-target test task scheduling method and platform based on decomposed variable neighborhoods |
| CN105740949A (en) * | 2016-03-03 | 2016-07-06 | 浙江工业大学 | Group global optimization method based on randomness best strategy |
| CN109522104A (en) * | 2018-10-15 | 2019-03-26 | 华南理工大学 | Utilize the method for differential evolution algorithm optimization two goal task of Iaas scheduling |
| CN110321217A (en) * | 2019-03-25 | 2019-10-11 | 深圳大学 | A kind of cloud resource dispatching method, device, equipment and the storage medium of multiple target |
| CN109936141A (en) * | 2019-03-28 | 2019-06-25 | 广州番禺职业技术学院 | A kind of Economic Dispatch method and system |
| CN110147274A (en) * | 2019-05-21 | 2019-08-20 | 武汉轻工大学 | Multiple target cloud task balance dispatching method, server and storage medium |
| CN110162393A (en) * | 2019-05-30 | 2019-08-23 | 奇瑞汽车股份有限公司 | Method for scheduling task, device and storage medium |
| CN110489229A (en) * | 2019-07-17 | 2019-11-22 | 长沙学院 | A kind of multiple target method for scheduling task and system |
| CN111343259A (en) * | 2020-02-17 | 2020-06-26 | 武汉轻工大学 | Binary code-based cloud task scheduling method, server and storage medium |
Non-Patent Citations (3)
| Title |
|---|
| A New Two-objective Single Machine Scheduling Problem Considers a Past-sequence-dependent Setup Time and Learning Effect;Shih-Hsin Chen ET AL;《 2018 1st IEEE International Conference on Knowledge Innovation and Invention (ICKII)》;20181209;全文 * |
| 一种基于优先级的独立任务调度算法;吴进等;《微计算机信息》;20090824(第24期);全文 * |
| 云计算下的资源负载均衡性调度仿真;孟令玺等;《计算机仿真》;20180415(第04期);全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN112579280A (en) | 2021-03-30 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US11762697B2 (en) | Method and apparatus for scheduling resource for deep learning framework | |
| CN109857555B (en) | Memory recovery method and device, storage medium and electronic equipment | |
| CN111813554A (en) | Task scheduling processing method and device, electronic equipment and storage medium | |
| CN102480512B (en) | For the method and apparatus of expansion servers end disposal ability | |
| US8307366B2 (en) | Post-processing phase in a distributed processing system using assignment information | |
| CN110389816B (en) | Method, apparatus and computer readable medium for resource scheduling | |
| CN112416585A (en) | GPU resource management and intelligent scheduling method for deep learning | |
| CN112667405B (en) | Information processing method, device, equipment and storage medium | |
| US20230037783A1 (en) | Resource scheduling method and related apparatus | |
| CN112148468A (en) | Resource scheduling method and device, electronic equipment and storage medium | |
| CN107977275B (en) | Task processing method based on message queue and related equipment | |
| CN112579280B (en) | Cloud resource scheduling method and device and computer storage medium | |
| CN113657634A (en) | Bank branch queuing waiting time prediction method and device | |
| CN113902120A (en) | Heterogeneous cloud resolving platform hybrid computing task dynamic self-adaptive partitioning scheduling method and system | |
| CN113010288B (en) | Scheduling method and device of cloud resources and computer storage medium | |
| CN111813517A (en) | Task queue allocation method and device, computer equipment and medium | |
| CN110389817B (en) | Scheduling method, device and computer readable medium of multi-cloud system | |
| CN116820714A (en) | Scheduling method, device, equipment and storage medium of computing equipment | |
| CN116089477A (en) | Distributed training method and system | |
| CN116109102A (en) | Resource allocation method and system based on genetic algorithm | |
| CN113537392B (en) | Method and device for identifying similar images, computing equipment and computer storage medium | |
| CN111475424A (en) | Method, apparatus and computer program product for managing a storage system | |
| CN114282968A (en) | Serial number acquisition method, device, server and storage medium | |
| CN113254200A (en) | Resource arrangement method and intelligent agent | |
| CN112540844A (en) | Container scheduling method and device in cluster, storage medium and electronic equipment |
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 |