CN117032904B - Cloud computing resource scheduling method, computer device and storage medium - Google Patents
Cloud computing resource scheduling method, computer device and storage medium Download PDFInfo
- Publication number
- CN117032904B CN117032904B CN202311294056.8A CN202311294056A CN117032904B CN 117032904 B CN117032904 B CN 117032904B CN 202311294056 A CN202311294056 A CN 202311294056A CN 117032904 B CN117032904 B CN 117032904B
- Authority
- CN
- China
- Prior art keywords
- resource
- scheduling
- resources
- condition
- conditions
- 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
- 238000000034 method Methods 0.000 title claims abstract description 97
- 238000005457 optimization Methods 0.000 claims abstract description 36
- 238000013468 resource allocation Methods 0.000 claims abstract description 36
- 238000012544 monitoring process Methods 0.000 claims abstract description 35
- 239000002245 particle Substances 0.000 claims description 28
- 238000013528 artificial neural network Methods 0.000 claims description 12
- 238000012549 training Methods 0.000 claims description 9
- 238000012360 testing method Methods 0.000 claims description 6
- 238000013461 design Methods 0.000 description 3
- 238000005516 engineering process Methods 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 238000005265 energy consumption Methods 0.000 description 1
- 230000004927 fusion Effects 0.000 description 1
- 238000013507 mapping Methods 0.000 description 1
- 238000013508 migration Methods 0.000 description 1
- 230000005012 migration Effects 0.000 description 1
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
- G06F9/44—Arrangements for executing specific programs
- G06F9/455—Emulation; Interpretation; Software simulation, e.g. virtualisation or emulation of application or operating system execution engines
- G06F9/45533—Hypervisors; Virtual machine monitors
- G06F9/45558—Hypervisor-specific management and integration aspects
-
- 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]
- G06F9/5005—Allocation of resources, e.g. of the central processing unit [CPU] to service a request
- G06F9/5027—Allocation of resources, e.g. of the central processing unit [CPU] to service a request the resource being a machine, e.g. CPUs, Servers, Terminals
-
- 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
- G06F9/44—Arrangements for executing specific programs
- G06F9/455—Emulation; Interpretation; Software simulation, e.g. virtualisation or emulation of application or operating system execution engines
- G06F9/45533—Hypervisors; Virtual machine monitors
- G06F9/45558—Hypervisor-specific management and integration aspects
- G06F2009/4557—Distribution of virtual machine instances; Migration and load balancing
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02D—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
- Y02D10/00—Energy efficient computing, e.g. low power processors, power management or thermal management
Abstract
The invention discloses a cloud computing resource scheduling method, which comprises the following steps: monitoring the use conditions of virtual machine and physical machine resources in real time; receiving a task submitted by a user side; analyzing the task type and the protocol required by the task, and carrying out resource scheduling prediction; performing resource allocation by adopting a multi-objective optimization method to allocate tasks to virtual machine resources; starting a secondary resource allocation period, analyzing a resource placement strategy by adopting a multi-objective optimization method, pre-occupying and updating PM resource conditions and VM resource conditions according to the resource placement strategy, and starting a next secondary resource allocation period; the resources are arranged onto physical machine resources. In the invention, the primary resource scheduling and the secondary resource scheduling are sequentially carried out, in the secondary scheduling, the resources are occupied before the previous task resources are scheduled, so that the resource occupation conflict is avoided, the scheduling efficiency of cloud computing is ensured, the optimal scheme is found through a multi-objective optimization method, and the resource utilization rate is improved.
Description
Technical Field
The invention relates to the field of data communication, in particular to a cloud computing resource scheduling method, a computer device for realizing the method and a storage medium capable of realizing the method.
Background
The cloud computing system can be divided into two types, namely public cloud and private cloud, wherein the public cloud is operated and maintained by a third party, resources are provided for users through the Internet, the private cloud is autonomously built by enterprises, the scale is small, IT resources which are more suitable for the operation of the enterprises are provided, cloud computing is a product of fusion of traditional computer and network technology development, such as grid computing, distributed computing, parallel computing, utility, network storage, virtualization, load balancing and the like, the virtualization technology refers to that computing elements are operated on a virtual basis rather than a real basis, such as virtual machines are created on the basis of physical hosts, the virtual machines are common technical means for improving the performance and the use efficiency of the computers in the technical field of virtualization, and unnecessary servers are closed through migration of the virtual machines, so that the energy consumption of the system is reduced.
The resource scheduling in the cloud computing environment can be divided into primary scheduling of a virtual resource layer and secondary scheduling of a physical resource layer, wherein the primary resource scheduling process is to divide a job submitted by a user into a plurality of tasks, and the core of scheduling is to allocate proper virtual resources to the tasks of the user; the process of the secondary scheduling is a mapping relationship between virtual resources and physical resources (the virtual resources are scheduled to the physical resources).
As shown in fig. 6, the existing cloud computing scheduling model directly performs primary scheduling and secondary scheduling through tasks submitted by users, but under the condition that at least two tasks in the scheduling mode are similar in time, the tasks submitted first are allocated to resources first, and then resource scheduling is performed after the tasks are allocated to the resources, so that the allocated resources cannot be occupied, the tasks submitted later are allocated to the same area in a resource pool before the tasks submitted first are scheduled in place, the scheduling process is easy to occur the situation that resource areas occupied by different resources overlap, the resource pool occupies resource conflicts, and the cloud resource utilization rate is poor and the situation that resources are disordered easily occurs.
Disclosure of Invention
Therefore, the invention provides a cloud computing resource scheduling method, a computer device and a storage medium, which effectively solve the problems that in the prior art, the resource areas occupied by different resources are easy to overlap, the resource pool occupies resource conflict, the cloud resource utilization rate is poor, and the resource confusion is easy to occur in the scheduling process.
In order to solve the technical problems, the invention specifically provides the following technical scheme: the cloud computing resource scheduling method comprises the following steps:
monitoring the use condition of virtual machine and physical machine resources in real time, and updating PM resource condition and VM resource condition;
receiving a task submitted by a user terminal, and sending a resource request according to the task;
analyzing the task type and the protocol required by the task based on the resource request, and carrying out resource scheduling prediction to obtain a scheduling prediction result;
according to the VM resource condition and the scheduling prediction result, performing resource allocation by adopting a multi-objective optimization method so as to allocate tasks to virtual machine resources, and returning a resource request and an allocation result to a user;
starting a secondary resource allocation period, analyzing a resource placement strategy by adopting a multi-objective optimization method based on PM resource conditions and VM resource conditions, pre-occupying and updating the PM resource conditions and the VM resource conditions according to the resource placement strategy, and starting a next secondary resource allocation period;
the resource placement policy is transmitted to place virtual machine resources onto physical machine resources.
Further, the method comprises the steps of,
the real-time monitoring of the use condition of the virtual machine and the physical machine resources and updating of the PM resource condition and the VM resource condition comprise:
the virtual machine resource pool and the physical machine resource pool send the service condition of the resources to a resource monitoring module;
the resource monitoring module receives the use conditions of the virtual machine and the physical machine resources so as to update the PM resource conditions and the VM resource conditions;
the PM resource condition and the VM resource condition are resource distribution conditions in a resource pool.
Further, the method comprises the steps of,
the starting of the secondary resource allocation period, based on the PM resource condition and the VM resource condition, adopts a multi-objective optimization method to analyze the resource placement strategy, and pre-occupies and updates the PM resource condition and the VM resource condition according to the resource placement strategy, and the starting of the next secondary resource allocation period comprises the following steps:
setting a large period consisting of a plurality of resource secondary allocation periods;
setting a physical resource scheduling module, and starting a channel for receiving PM resource conditions and VM resource conditions in the starting point of the secondary resource allocation period by the physical resource scheduling module;
performing resource placement strategy analysis by adopting a multi-objective optimization method;
and sending the pre-occupation instruction to a resource monitoring module according to the resource placement strategy, and pre-updating the PM resource condition and the VM resource condition in the resource monitoring module.
Further, the method comprises the steps of,
when the resource monitoring module receives the resource occupation instruction, the current secondary resource allocation period is not ended, and the next secondary resource allocation period is started.
Further, the method comprises the steps of,
the follow-up work of the current secondary resource allocation period is as follows:
and sending the resource placement strategy to a resource monitoring module through a virtual machine planning module, and arranging corresponding virtual machine resources on a physical machine resource corresponding area by the resource monitoring module based on the resource placement strategy.
Further, the method comprises the steps of,
the method comprises the steps of analyzing the task type and the protocol required by the task based on the resource request, and carrying out resource scheduling prediction to obtain a scheduling prediction result, wherein the method comprises the following steps:
analyzing the task type and the protocol required by the task based on the resource request;
constructing an RBF neural network, and obtaining optimal parameters of the RBF neural network through a particle swarm optimization algorithm;
dividing real-time application resource demand data contained in PM resource conditions or VM resource conditions into two parts, wherein one part is used as a training sample, and the other part is used as a test sample;
training the BRF neural network through a training sample;
and predicting the actual application resource demand according to the test sample through the trained neural network.
Further, the method comprises the steps of,
the multi-objective optimization method comprises the following steps:
population p= { S consisting of m sub-groups 1 ，S 2 ，……，S M Each subgroup S k ={X 1 ，X 2 ，……，X N N particle swarms;
initializing a particle swarm;
calculating a fitness value based on the actual application resource demand, and determining the searching speed, the searching position, the optimal searching position and the global optimal searching position of the particles;
determining variant particles;
and (5) after multiple iterations, obtaining an optimal solution.
Further, the method comprises the steps of,
according to VM resource conditions and scheduling prediction results, a virtual resource allocation scheme is obtained by adopting a multi-objective optimization method so as to allocate resources;
each particle swarm is a virtual resource allocation scheme or a resource placement strategy.
In order to solve the technical problems, the invention further provides the following technical scheme: a computer apparatus, comprising:
at least one processor; and a memory communicatively coupled to the processor;
wherein the memory stores instructions executable by the processors, the instructions being executable by at least one of the processors to cause the processor to be configured to perform a cloud computing resource scheduling method.
In order to solve the technical problems, the invention further provides the following technical scheme: a computer readable storage medium having stored therein computer executable instructions that, when executed by a processor, enable the processor to perform a cloud computing resource scheduling method.
Compared with the prior art, the invention has the following beneficial effects:
in the invention, first-level resource scheduling is firstly carried out, a multi-objective optimization method is adopted to carry out resource allocation according to VM resource conditions and scheduling prediction results so as to allocate tasks to virtual machine resources, then the resource scheduling is carried out, and then second-level resource scheduling is carried out;
in the secondary resource scheduling process, a resource secondary allocation period is set, a multi-objective optimization method is adopted to analyze a resource placement strategy according to PM resource conditions and VM resource conditions, PM resource conditions and VM resource conditions are pre-occupied and updated according to the resource placement strategy, the next resource secondary allocation period is started after the pre-occupation of the resources, in the secondary scheduling process, time sequences are set for task scheduling processes of different tasks, the resource secondary allocation period of the next task is started after the occupation of the resources in the resource secondary allocation period of the previous task, the occupation of the resources is carried out before the previous task resource is scheduled, the situation that the resource pool area overlapped with the occupied area of the previous task is obtained in the analysis process of the resource placement strategy of the next task is avoided, the resource scheduling is carried out after the occupation of the resources, the situation that the resource occupation conflicts are caused due to the time consumption of the resource scheduling is avoided, the cloud computing scheduling efficiency is guaranteed, and the optimal virtual resource allocation scheme or the resource placement strategy is found through the multi-objective optimization method, and the resource utilization rate is improved.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below. It will be apparent to those of ordinary skill in the art that the drawings in the following description are exemplary only and that other implementations can be obtained from the extensions of the drawings provided without inventive effort.
Fig. 1 is a flowchart of a cloud computing resource scheduling method provided by an embodiment of the present invention;
FIG. 2 is a flow chart of a virtual resource scheduling process in an embodiment of the present invention;
FIG. 3 is a flow chart of a physical resource scheduling process in an embodiment of the present invention;
FIG. 4 is a schematic diagram of a large period distribution in an embodiment of the present invention;
FIG. 5 is a flow chart of a resource prediction and multi-objective optimization process in an embodiment of the invention;
fig. 6 is a flow chart of a prior art resource scheduling process.
Detailed Description
The following description of the embodiments of the present invention will be made clearly and completely with reference to the accompanying drawings, in which it is apparent that the embodiments described are only some embodiments of the present invention, but not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
As shown in fig. 1, 2 and 3, the present invention provides a cloud computing resource scheduling method, which includes:
monitoring the use condition of virtual machine and physical machine resources in real time, and updating PM resource condition and VM resource condition;
receiving a task submitted by a user terminal, and sending a resource request according to the task;
analyzing the task type and the protocol required by the task based on the resource request, and carrying out resource scheduling prediction to obtain a scheduling prediction result;
according to the VM resource condition and the scheduling prediction result, performing resource allocation by adopting a multi-objective optimization method so as to allocate tasks to virtual machine resources, and returning a resource request and an allocation result to a user;
starting a secondary resource allocation period, analyzing a resource placement strategy by adopting a multi-objective optimization method based on PM resource conditions and VM resource conditions, pre-occupying and updating the PM resource conditions and the VM resource conditions according to the resource placement strategy, and starting a next secondary resource allocation period;
the resource placement policy is transmitted to place virtual machine resources onto physical machine resources.
In the invention, first-level resource scheduling is carried out, a multi-objective optimization method is adopted to carry out resource allocation according to VM resource conditions and scheduling prediction results so as to allocate tasks to virtual machine resources, then the resource scheduling is carried out, and then second-level resource scheduling is carried out.
In the secondary resource scheduling process, a resource secondary allocation period is set, a multi-objective optimization method is adopted to analyze a resource placement strategy according to PM resource conditions and VM resource conditions, PM resource conditions and VM resource conditions are pre-occupied and updated according to the resource placement strategy, the next resource secondary allocation period is started after the pre-occupation of the resources, in the secondary scheduling process, time sequences are set for task scheduling processes of different tasks, the resource secondary allocation period of the next task is started after the occupation of the resources in the resource secondary allocation period of the previous task, the occupation of the resources is carried out before the previous task resource is scheduled, the situation that the resource pool area overlapped with the occupied area of the previous task is obtained in the analysis process of the resource placement strategy of the next task is avoided, the resource scheduling is carried out after the occupation of the resources, the situation that the resource occupation conflicts are caused due to the time consumption of the resource scheduling is avoided, the cloud computing scheduling efficiency is guaranteed, and the optimal virtual resource allocation scheme or the resource placement strategy is found through the multi-objective optimization method, and the resource utilization rate is improved.
The invention also comprises a primary resource scheduling process and a secondary resource scheduling process which are respectively virtual resource scheduling and physical resource scheduling, and the difference between the virtual resource scheduling and the physical resource scheduling is the resource scheduling prediction and resource scheduling process compared with the prior art.
The invention mainly comprises a cloud computing platform, a resource analysis module, a virtual resource scheduling module, a resource scheduling prediction module, a resource monitoring module, a physical resource scheduling module and a virtual machine planning module, so as to form a whole system capable of realizing cloud computing resource scheduling.
The precondition for realizing the resource scheduling in the invention is to monitor the service condition of the resource in real time so as to facilitate the subsequent scheduling of the resource to the idle resource area.
Therefore, monitoring the use condition of the virtual machine and the physical machine resources in real time, and updating the PM resource condition and the VM resource condition includes:
the virtual machine resource pool and the physical machine resource pool send the service condition of the resources to a resource monitoring module;
the resource monitoring module receives the use conditions of the virtual machine and the physical machine resources so as to update the PM resource conditions and the VM resource conditions;
the PM resource condition and the VM resource condition are resource distribution conditions in a resource pool.
In the above embodiment, the PM resource situation and the VM resource situation are sent by the virtual machine resource pool and the physical machine resource pool to the resource monitoring module independently, the real-time PM resource situation and the VM resource situation are stored in the resource monitoring module, and the physical or virtual resource scheduling module acquires the real-time PM resource situation and the VM resource situation data stored in the resource monitoring module before resource scheduling.
In the above case, updating the real-time PM resource condition and VM resource condition stored in the resource monitoring module after the occupation of the resource scheduling is completed may cause deviation of the judgment of the resource usage condition by the physical or virtual resource scheduling module, for example, when the first time data is scheduled to the 1-10 area in the resource pool, but the data is still occupied in the scheduling process, in the early stage of the second time data scheduling process, the 1-10 area in the resource pool is judged to be in an idle state in the judgment process of the area in the resource pool, that is, the second time resource placement strategy may finally obtain the result of the 1-10 area placed in the resource pool, resulting in the scheduling conflict of the data, in order to solve the above problem, the present invention sets the secondary allocation period of the resource in the physical resource scheduling module, and further makes the following design:
starting a secondary resource allocation period, analyzing a resource placement strategy by adopting a multi-objective optimization method based on PM resource conditions and VM resource conditions, and pre-occupying and updating the PM resource conditions and the VM resource conditions according to the resource placement strategy, and starting a next secondary resource allocation period, wherein the method comprises the following steps:
setting a large period consisting of a plurality of resource secondary allocation periods;
setting a physical resource scheduling module, and starting a channel for receiving PM resource conditions and VM resource conditions at a starting point of a secondary resource allocation period by the physical resource scheduling module;
performing resource placement strategy analysis by adopting a multi-objective optimization method;
and sending the pre-occupation instruction to a resource monitoring module according to the resource placement strategy, and pre-updating the PM resource condition and the VM resource condition in the resource monitoring module.
In the above embodiment, each time of resource scheduling is periodically divided, that is, as shown in fig. 4, each time of resource scheduling corresponds to a secondary allocation period of resources, and when the secondary allocation period of resources starts, the corresponding resource scheduling process can only start.
When the resource monitoring module receives the resource occupation instruction, the current secondary resource allocation period is not ended, and the next secondary resource allocation period is started.
In a specific process, as shown in fig. 3 and fig. 4, a large period is started, a first secondary resource allocation period is started, a first resource scheduling process is started, a physical resource scheduling module starts a channel for receiving the PM resource condition and the VM resource condition, a multi-objective optimization method is adopted to perform resource placement policy analysis, a pre-occupied instruction is sent to a resource monitoring module according to the resource placement policy, the PM resource condition and the VM resource condition are updated in advance in the resource monitoring module, at this time, a pre-occupied time point is reached, although the secondary resource allocation period is not completely ended, and then the placement of resources is performed, at this time, a second secondary resource allocation period is started, a second resource scheduling process is started, a physical resource scheduling module starts a channel for receiving the real-time PM resource condition and the VM resource condition, at this time, the received PM resource condition and the VM resource condition are in a resource pool after the first resource scheduling is completed, and data in the resource pool may not be placed yet, but the corresponding PM resource condition and VM resource condition data are updated.
The advanced occupation design and the design that the overlapping area exists in the secondary allocation period of the adjacent resources can avoid the conflict of resource distribution and ensure the efficient performance of resource allocation.
The follow-up work of the current secondary resource allocation period is as follows: and sending the resource placement strategy to a resource monitoring module through a virtual machine planning module, and arranging corresponding virtual machine resources on a physical machine resource corresponding area by the resource monitoring module based on the resource placement strategy.
In the invention, the virtual resource scheduling process and the physical resource scheduling process both comprise a resource scheduling prediction process and a multi-objective optimization process.
As shown in fig. 5, the resource scheduling prediction process includes analyzing a task type and a protocol required by a task based on a resource request, and performing resource scheduling prediction to obtain a scheduling prediction result, where the method includes:
analyzing the task type and the protocol required by the task based on the resource request;
constructing an RBF neural network, and obtaining optimal parameters of the RBF neural network through a particle swarm optimization algorithm;
dividing real-time application resource demand data contained in PM resource conditions or VM resource conditions into two parts, wherein one part is used as a training sample, and the other part is used as a test sample;
training the BRF neural network through a training sample;
and predicting the actual application resource demand according to the test sample through the trained neural network.
As shown in fig. 5, the multi-objective optimization method includes:
population p= { S consisting of m sub-groups 1 ，S 2 ，……，S M Each subgroup S k ={X 1 ，X 2 ，……，X N N particle swarms;
initializing a particle swarm;
calculating a fitness value based on the actual application resource demand, and determining the searching speed, the searching position, the optimal searching position and the global optimal searching position of the particles;
determining variant particles;
and (5) after multiple iterations, obtaining an optimal solution.
In the above embodiment, the particle swarm optimization algorithm introduces the maximum speed to generate variant particles, and overcomes the problem of local optimization by the variant particles, and the population P= { S composed of m sub-groups 1 ，S 2 ，……，S M Each subgroup S k ={X 1 ，X 2 ，……，X N And (3) N particle groups, and the speed and position updating formula of the particles is as follows: ；；/>；
wherein,for the position of the ith particle in subgroup k, < +.>Is the speed of the ith particle, +.>Is individual optimal location information,/->Is a speed threshold (maximum speed of particles, specific value of the data can be adjusted according to the situation during actual operation),>for the optimal position in subgroup k +.>Is the optimal position of the adjacent subgroup of subgroup k, < ->、/>、/>For learning factors->、/>、/>Is a random number set in the (0, 1) interval;
and obtaining the speed, the position, the optimal position and the global optimal position of the particles according to the formula, determining the variant particles through a third formula, and outputting an optimal solution to obtain an optimal virtual resource allocation scheme or resource placement strategy.
In the virtual resource scheduling process, a virtual resource allocation scheme is obtained by adopting a multi-objective optimization method according to VM resource conditions and scheduling prediction results so as to allocate resources, and each particle swarm is a virtual resource allocation scheme or a resource placement strategy.
In the physical resource scheduling process, a multi-objective optimization method is adopted to analyze the resource placement strategy according to the PM resource condition, the VM resource condition and the scheduling prediction result, and the resource placement strategy is obtained so as to perform resource placement.
The present invention also provides a computer apparatus comprising: at least one processor; and a memory communicatively coupled to the processor;
the memory stores instructions executable by the processor, the instructions being executable by the at least one processor to cause the processor to be configured to perform the cloud computing resource scheduling method described above.
In addition, the invention also provides a computer readable storage medium, wherein computer executable instructions are stored in the computer readable storage medium, and when the instructions in the storage medium are executed by a processor, the processor is enabled to execute the cloud computing resource scheduling method.
The above embodiments are only exemplary embodiments of the present application and are not intended to limit the present application, the scope of which is defined by the claims. Various modifications and equivalent arrangements may be made to the present application by those skilled in the art, which modifications and equivalents are also considered to be within the scope of the present application.
Claims (8)
1. The cloud computing resource scheduling method is characterized by comprising the following steps of:
monitoring the use condition of virtual machine and physical machine resources in real time, and updating PM resource condition and VM resource condition;
receiving a task submitted by a user terminal, and sending a resource request according to the task;
analyzing the task type and the protocol required by the task based on the resource request, and carrying out resource scheduling prediction to obtain a scheduling prediction result;
according to the VM resource condition and the scheduling prediction result, performing resource allocation by adopting a multi-objective optimization method so as to allocate tasks to virtual machine resources, and returning a resource request and an allocation result to a user;
starting a secondary resource allocation period, analyzing a resource placement strategy by adopting a multi-objective optimization method based on PM resource conditions and VM resource conditions, pre-occupying and updating the PM resource conditions and the VM resource conditions according to the resource placement strategy, and starting a next secondary resource allocation period;
transmitting a resource placement policy to place virtual machine resources onto physical machine resources;
the real-time monitoring of the use condition of the virtual machine and the physical machine resources and updating of the PM resource condition and the VM resource condition comprise:
the virtual machine resource pool and the physical machine resource pool send the service condition of the resources to a resource monitoring module;
the resource monitoring module receives the use conditions of the virtual machine and the physical machine resources so as to update the PM resource conditions and the VM resource conditions;
wherein, the PM resource condition and the VM resource condition are resource distribution conditions in a resource pool;
the starting of the secondary resource allocation period, based on the PM resource condition and the VM resource condition, adopts a multi-objective optimization method to analyze the resource placement strategy, and pre-occupies and updates the PM resource condition and the VM resource condition according to the resource placement strategy, and the starting of the next secondary resource allocation period comprises the following steps:
setting a large period consisting of a plurality of resource secondary allocation periods;
setting a physical resource scheduling module, and starting a channel for receiving PM resource conditions and VM resource conditions in the starting point of the secondary resource allocation period by the physical resource scheduling module;
performing resource placement strategy analysis by adopting a multi-objective optimization method;
and sending the pre-occupation instruction to a resource monitoring module according to the resource placement strategy, and pre-updating the PM resource condition and the VM resource condition in the resource monitoring module.
2. The method for scheduling cloud computing resources of claim 1,
when the resource monitoring module receives the resource occupation instruction, the current secondary resource allocation period is not ended, and the next secondary resource allocation period is started.
3. The method for scheduling cloud computing resources of claim 2,
the follow-up work of the current secondary resource allocation period is as follows:
and sending the resource placement strategy to a resource monitoring module through a virtual machine planning module, and arranging corresponding virtual machine resources on a physical machine resource corresponding area by the resource monitoring module based on the resource placement strategy.
4. The method for scheduling cloud computing resources as recited in claim 3, wherein,
the method comprises the steps of analyzing the task type and the protocol required by the task based on the resource request, and carrying out resource scheduling prediction to obtain a scheduling prediction result, wherein the method comprises the following steps:
analyzing the task type and the protocol required by the task based on the resource request;
constructing an RBF neural network, and obtaining optimal parameters of the RBF neural network through a particle swarm optimization algorithm;
dividing real-time application resource demand data contained in PM resource conditions or VM resource conditions into two parts, wherein one part is used as a training sample, and the other part is used as a test sample;
training the BRF neural network through a training sample;
and predicting the actual application resource demand according to the test sample through the trained neural network.
5. The method for scheduling cloud computing resources of claim 4,
the multi-objective optimization method comprises the following steps:
the population p= { S1, S2, … …, SM } consisting of m sub-groups, each sub-group S k ={X 1 ，X 2 ，……，X N N particle swarms;
initializing a particle swarm;
calculating a fitness value based on the actual application resource demand, and determining the searching speed, the searching position, the optimal searching position and the global optimal searching position of the particles;
determining variant particles;
and (5) after multiple iterations, obtaining an optimal solution.
6. The method for scheduling cloud computing resources of claim 5,
according to VM resource conditions and scheduling prediction results, a virtual resource allocation scheme is obtained by adopting a multi-objective optimization method so as to allocate resources;
each particle swarm is a virtual resource allocation scheme or a resource placement strategy.
7. A computer apparatus, comprising:
at least one processor; and a memory communicatively coupled to the processor;
wherein the memory stores instructions executable by the processor, the instructions being executable by at least one of the processors to cause the processor to be configured to perform the cloud computing resource scheduling method of any of claims 1-6.
8. A computer-readable storage medium comprising,
the computer-readable storage medium has stored therein computer-executable instructions that, when executed by a processor, enable the processor to perform the cloud computing resource scheduling method of any one of claims 1-6.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311294056.8A CN117032904B (en) | 2023-10-09 | 2023-10-09 | Cloud computing resource scheduling method, computer device and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311294056.8A CN117032904B (en) | 2023-10-09 | 2023-10-09 | Cloud computing resource scheduling method, computer device and storage medium |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN117032904A CN117032904A (en) | 2023-11-10 |
| CN117032904B true CN117032904B (en) | 2024-01-26 |
Family
ID=88645255
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202311294056.8A Active CN117032904B (en) | 2023-10-09 | 2023-10-09 | Cloud computing resource scheduling method, computer device and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117032904B (en) |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106027318A (en) * | 2016-07-24 | 2016-10-12 | 成都育芽科技有限公司 | Cloud computing-based two-level optimal scheduling management platform for virtual machine |
| CN114003389A (en) * | 2021-11-08 | 2022-02-01 | 苏州全时空信息技术有限公司 | Cloud platform resource allocation method and system based on joint selection and virtual machine placement |
Family Cites Families (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US9183016B2 (en) * | 2013-02-27 | 2015-11-10 | Vmware, Inc. | Adaptive task scheduling of Hadoop in a virtualized environment |
| US20180260251A1 (en) * | 2016-08-28 | 2018-09-13 | Vmware, Inc. | Use of nested hypervisors by a resource-exchange system to enhance data and operational security and to facilitate component installation |
-
2023
- 2023-10-09 CN CN202311294056.8A patent/CN117032904B/en active Active
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106027318A (en) * | 2016-07-24 | 2016-10-12 | 成都育芽科技有限公司 | Cloud computing-based two-level optimal scheduling management platform for virtual machine |
| CN114003389A (en) * | 2021-11-08 | 2022-02-01 | 苏州全时空信息技术有限公司 | Cloud platform resource allocation method and system based on joint selection and virtual machine placement |
Non-Patent Citations (1)
| Title |
|---|
| 多媒体云计算中的资源管理研究;王家庆;中国优秀硕士学位论文全文数据库信息科技辑（月刊）(第02期);I139-171 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN117032904A (en) | 2023-11-10 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Chen et al. | Deep reinforcement learning for computation offloading in mobile edge computing environment | |
| CN109669768B (en) | Resource allocation and task scheduling method for edge cloud combined architecture | |
| Li et al. | SmartPC: Hierarchical pace control in real-time federated learning system | |
| CN111274036B (en) | Scheduling method of deep learning task based on speed prediction | |
| CN111064633B (en) | Cloud-edge cooperative power information communication equipment automated testing resource allocation method | |
| Han et al. | Tailored learning-based scheduling for kubernetes-oriented edge-cloud system | |
| Wang et al. | A deep learning based energy-efficient computational offloading method in Internet of vehicles | |
| Kaur et al. | Deep‐Q learning‐based heterogeneous earliest finish time scheduling algorithm for scientific workflows in cloud | |
| Duggan et al. | A reinforcement learning approach for the scheduling of live migration from under utilised hosts | |
| CN113157422A (en) | Cloud data center cluster resource scheduling method and device based on deep reinforcement learning | |
| CN109684078A (en) | Resource dynamic distributing method and system for spark streaming | |
| CN109710372B (en) | Calculation intensive cloud workflow scheduling method based on owl search algorithm | |
| CN110086855A (en) | Spark task Intellisense dispatching method based on ant group algorithm | |
| CN116450312A (en) | Scheduling strategy determination method and system for pipeline parallel training | |
| Zhang et al. | An approach of multi-objective computing task offloading scheduling based NSGS for IOV in 5G | |
| CN105740059A (en) | Particle swarm scheduling method for divisible task | |
| Chai | Task scheduling based on swarm intelligence algorithms in high performance computing environment | |
| Hu et al. | Lars: A latency-aware and real-time scheduling framework for edge-enabled internet of vehicles | |
| CN117032904B (en) | Cloud computing resource scheduling method, computer device and storage medium | |
| CN116755867B (en) | Satellite cloud-oriented computing resource scheduling system, method and storage medium | |
| CN113190342A (en) | Method and system architecture for multi-application fine-grained unloading of cloud-edge cooperative network | |
| Cao et al. | Performance and stability of application placement in mobile edge computing system | |
| Cao et al. | Online cost-rejection rate scheduling for resource requests in hybrid clouds | |
| Barika et al. | Adaptive scheduling for efficient execution of dynamic stream workflows | |
| Luo et al. | IPSO: Improved particle swarm optimization based task scheduling at the Cloud Data Center |
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 |