CN113645076B - Virtual network resource allocation method based on hypergraph matching algorithm - Google Patents

Virtual network resource allocation method based on hypergraph matching algorithm Download PDF

Info

Publication number
CN113645076B
CN113645076B CN202110925564.6A CN202110925564A CN113645076B CN 113645076 B CN113645076 B CN 113645076B CN 202110925564 A CN202110925564 A CN 202110925564A CN 113645076 B CN113645076 B CN 113645076B
Authority
CN
China
Prior art keywords
virtual
super
vertex
graph
network
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
Application number
CN202110925564.6A
Other languages
Chinese (zh)
Other versions
CN113645076A (en
Inventor
刘伟
张宇飞
李建东
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Xidian University
Original Assignee
Xidian University
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Xidian University filed Critical Xidian University
Priority to CN202110925564.6A priority Critical patent/CN113645076B/en
Publication of CN113645076A publication Critical patent/CN113645076A/en
Application granted granted Critical
Publication of CN113645076B publication Critical patent/CN113645076B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L12/00—Data switching networks
    • H04L12/28—Data switching networks characterised by path configuration, e.g. LAN [Local Area Networks] or WAN [Wide Area Networks]
    • H04L12/46—Interconnection of networks
    • H04L12/4641—Virtual LANs, VLANs, e.g. virtual private networks [VPN]
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L41/00—Arrangements for maintenance, administration or management of data switching networks, e.g. of packet switching networks
    • H04L41/08—Configuration management of networks or network elements
    • H04L41/0896—Bandwidth or capacity management, i.e. automatically increasing or decreasing capacities
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L41/00—Arrangements for maintenance, administration or management of data switching networks, e.g. of packet switching networks
    • H04L41/12—Discovery or management of network topologies
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L41/00—Arrangements for maintenance, administration or management of data switching networks, e.g. of packet switching networks
    • H04L41/14—Network analysis or design
    • 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 provides a virtual network resource allocation method based on a hypergraph matching algorithm, which comprises the following implementation steps of: and each super point represents a virtual node in the virtual network before mapping, each super edge represents a server in the physical network, the power consumption value of each super edge represents the power consumption of each server in the physical network, and the super points corresponding to the virtual nodes in the virtual network are mapped to the super edges corresponding to the servers in the physical network to form an initial super graph. Because the process of directly searching the maximum power consumption value of a vertex disjoint super-edge subset on the basis of the initial super-graph is NP-hard, the initial super-graph is converted into a conflict graph, and the independent set of the maximum edge disjoint vertex is approximately obtained, so that the total power consumption of the data center server is reduced while the virtual network receiving rate is ensured.

Description

Virtual network resource allocation method based on hypergraph matching algorithm
Technical Field
The invention belongs to the technical field of computers, and further relates to a virtual network resource allocation method based on a hypergraph matching algorithm in the technical field of data communication. The method can be applied to reasonable distribution of virtual network resources in a data center environment.
Background
Cloud computing has been widely used for real-time processing of large-scale data, and physical resource management of a cloud data center can shield heterogeneity and complexity of underlying resources through a virtualization technology, and can allocate virtualized resources to users according to needs of the users. Due to the continuous expansion of the current internet Data center idc (internet Data center) scale, it is expected that the total energy consumption of our Data center will break through 2500 billion kilowatt-hours in 2023. However, most of the existing virtual network resource allocation schemes only focus on the receiving rate and the cost ratio of the virtual network request, and the index of energy consumption is rarely focused on; meanwhile, although the existing dynamic resource allocation scheme reduces the energy consumption of the system, the existing dynamic resource allocation scheme is at the cost of losing the fairness of virtual resources. Aiming at the defects, the multi-dimensional mapping relation between the objects can be represented by using the hypergraph, so that the relation of the virtual network request to the physical network mapping can be represented by using the hypergraph, and the method of energy consumption perception two-stage mapping is adopted for solving. In order to better distribute the user request task, the computing resources and the communication resources in the physical network must be considered comprehensively, and the virtual network resource distribution method based on the hypergraph matching algorithm is an optimization design technology for solving the problem.
Shenzhen advanced technology research institute disclosed a virtual network resource allocation method in the patent document 'a virtual network resource allocation method, system and electronic device' filed by Shenzhen advanced technology research institute (application number: 2018114303629 application publication number: CN 109412865A). The method comprises the following specific steps: generating a task graph according to network information of all virtual machines on a physical machine, and dividing the task graph by adopting a multi-level graph division algorithm to obtain k mutually disjoint division subsets; the second step: calculating the key path with the longest completion time in each segmentation subset to obtain a virtual machine positioned at a key path node; the third step: and distributing single-root input and output virtualized virtual equipment for the virtual machine at the key path node. The method has the disadvantages that the task graph is divided by adopting a multi-level graph division algorithm, so that the resource distribution is concentrated in a local part, the virtual network occupies too large physical network link resources, and the receiving rate of the virtual network resources is reduced.
Mengyang He, Lei Zhuang et al, in its published paper "DROI: Energy-efficiency virtual network embedded on dynamic regions of interest" (Elsevier 2020-Computer Networks) propose a virtual resource allocation method for dynamic regions of interest DROI (dynamic Region of interest). The method comprises the following specific steps: constructing a physical network weight matrix based on field theory; the second step is that: constructing a diagonal matrix and calculating a normalized Laplace matrix; the third step: and repeatedly enabling interested region ROI (region of interest) nodes and links to bear virtual requests by using k-means clustering, and reducing the number of the nodes and the links which are converted into an active state from a dormant state. Although the method for dynamically allocating resources reduces the energy consumption of the server system, the method still has the disadvantages that the fairness of the virtual resources cannot be ensured by performing region division on the virtual resources, so that the receiving rate of the virtual resources in different resource allocation modes is unstable.
Disclosure of Invention
The invention aims to provide a virtual network resource allocation method based on a hypergraph matching algorithm aiming at the defects of the prior art, and the method is used for solving the problems of overhigh energy consumption and unstable virtual network acceptance rate caused by improper virtual network resource allocation in a data center.
The technical idea of the invention is that each hyper point represents a virtual node in the virtual network before mapping, each hyper edge represents a server in the physical network, each hyper point is randomly mapped to d hyper edges, the mapping range of the virtual node is expanded, the problem of unstable receiving rate of the virtual network is solved, the hyper graph matching algorithm is utilized to ensure that the virtual network is distributed to a small amount of physical resources, and the problem of overhigh energy consumption caused by improper distribution of virtual network resources in a data center is solved.
The technical scheme for realizing the aim of the invention comprises the following steps:
step 1, constructing a virtual network and a physical network:
(1a) based on graph theory, K virtual networks are modeled, K is more than or equal to 1 and less than or equal to 20, and a weighted undirected graph is used
Figure BDA0003209167670000021
Expressing the topology of the kth virtual network node and link, wherein K is more than or equal to 1 and less than or equal to K;
(1b) based on graph theory, a physical network is modeled and weighted undirected graph G is usedsRepresenting the topology of the physical network nodes and links;
step 2, mapping the virtual network to the physical network:
(2a) randomly generating a 1 x d set, wherein each element in the set is a pre-mapped super edge of each pre-mapped super point, each super point represents a virtual node in a pre-mapped virtual network, each super edge represents a server in a physical network, and d represents the total number of the super edges mapped to the physical network by the super points in the virtual network;
(2b) if the residual cpu core number of each element in the set is greater than the cpu core number of the super point, the success probability of the super point mapping to the super edge is 1, otherwise, the probability is 0;
(2c) selecting an unmapped virtual link, searching a shortest path between two super edges of two super points at two ends of the selected virtual link by using a Floyd algorithm, and mapping the selected virtual link to the shortest path;
(2d) judging whether all the virtual links are mapped, if so, executing the step 3, otherwise, executing the step (2 c);
and 3, generating an independent set:
(3a) mapping the corresponding super point of the virtual node in the virtual network to the corresponding super edge of the server in the physical network to form an initial super graph;
(3b) correspondingly generating a vertex in the conflict graph by each hyper-edge in the initial hyper-graph, and if the hyper-edges have crossed hyper-points, forming an edge between the corresponding vertices in the conflict graph;
(3c) calculating the power consumption of a vertex corresponding to each server in the physical network according to the following formula:
ω(pi)＝-(Pmin+(Pmax-Pmin)×Ui)
wherein, ω (p)i) Representing the vertex p corresponding to the ith server in the physical networkiPower consumption of PminRepresenting the average power consumption, P, of all servers in the physical network in the idle statemaxRepresenting full power consumption, U, of all servers in a physical networkiRepresenting the CPU core number utilization rate of the ith server in the physical network;
(3d) initializing an empty set, selecting a vertex with the maximum power consumption value in the vertex set of the conflict graph, putting the vertex into the set, deleting the vertex and the adjacent vertex from the vertex set of the conflict graph, and so on to obtain an initial independent set of the vertices stored into the set;
(3e) arranging vertexes in the initial independent set in an ascending order according to the power consumption values, selecting a vertex which is not selected in the initial independent set and has the minimum power consumption value, initializing an empty set, putting vertexes which are adjacent to the vertexes in the vertex set of the conflict graph into the set, and arranging the vertexes in a descending order according to the power consumption values of the vertexes to obtain an adjacent set of the vertexes stored in the set;
(3f) using the vertex selected in the initial independent set as the central point of the claw graph, and finding out the adjacent set of the central point
Figure BDA0003209167670000031
A drawing; wherein, the first and the second end of the pipe are connected with each other,
Figure BDA0003209167670000032
to represent
Figure BDA0003209167670000033
The amount of the (c) middle claw,
Figure BDA0003209167670000034
(3g) judgment of E (S)*) If < E (S) is true, if so, then use S*Step (3E) is executed after S is replaced, otherwise, step 4 is executed, wherein E (S)*) To represent
Figure BDA0003209167670000035
In the figure, claw replaces the original independent set S
Figure BDA0003209167670000036
Initial independent set S after the center point of the graph*The total power consumption of all the vertexes in the initial independent set S is represented by E (S);
step 4, completing the virtual network resource allocation:
will S*Each vertex in (1) is converted into a corresponding server to complete the allocation of virtual network resources.
Compared with the prior art, the invention has the following advantages:
first, the invention maps the super point corresponding to the virtual node in the virtual network to the super edge corresponding to the server in the physical network to form the initial super graph, so that each virtual node in the virtual network has multiple mapping selections, and the problems that the virtual network occupies too large physical network link resources and the receiving rate of the virtual network resources is reduced due to one-time mapping of the virtual network in the prior art are solved, so that the receiving rate of the virtual network is improved.
Secondly, because the invention adopts the generation of the independent set to obtain the vertex disjoint super edge subsets, the super edges in the subsets receive more virtual networks, and the virtual network resources are distributed to the super edges as few as possible, thereby overcoming the defect of low virtual network receiving rate caused by reducing power consumption in the prior art, and reducing the power consumption of the data center server while ensuring the virtual network receiving rate.
Drawings
FIG. 1 is a flow chart of the present invention;
FIG. 2 is a simulation of the present invention.
Detailed Description
The invention is described in further detail below with reference to the figures and examples.
With reference to fig. 1, the specific steps for implementing the present invention are described as follows.
Step 1, constructing a virtual network and a physical network.
Based on graph theory, K virtual networks are modeled, K is more than or equal to 1 and less than or equal to 20, and a weighted undirected graph is used
Figure BDA0003209167670000041
And K is more than or equal to 1 and less than or equal to K, and the topology of the kth virtual network node and the link is represented.
The number of virtual networks is set to 20.
Based on graph theory, a physical network is modeled and a weighted undirected graph G is usedsRepresenting the topology of the physical network nodes and links.
And 2, mapping the virtual network to the physical network.
The first step, a 1 x d set is randomly generated, each element in the set is a pre-mapping super edge of each pre-mapping super point, each super point represents a virtual node in a pre-mapping virtual network, each super edge represents a server in a physical network, and d represents the total number of super edges in the virtual network which are mapped to the physical network.
Hypergraph H ═ (v (H), e (H)) is a generalization of general graphs, where v (H) is a vertex set, and e (H) is an edge set. Each hyper-edge E (H) is a subset of vertices, and is called a consistent hyper-graph if the number of vertices contained in each hyper-edge is the same. The graph is a special case of a hypergraph, i.e., each hyperedge contains two vertices.
And secondly, if the residual cpu core number of each element in the set is greater than the cpu core number of the super point, the success probability of the super point mapping to the super edge is 1, otherwise, the probability is 0.
And thirdly, selecting an unmapped virtual link, searching a shortest path between two super edges where two super points are positioned at two ends of the selected virtual link by adopting a Floyd algorithm, and mapping the selected virtual link to the shortest path.
And step four, judging whether all the virtual links are mapped, if so, executing the step 3, otherwise, executing the step three.
And 3, generating an independent set.
The first step, mapping the super point corresponding to the virtual node in the virtual network to the super edge corresponding to the server in the physical network to form the initial super graph.
And step two, correspondingly generating a vertex in the conflict graph by each hyper-edge in the initial hyper-graph, and if the hyper-edges have crossed over points, forming a side between the corresponding vertices in the conflict graph.
Because it is NP difficult to directly find the maximum power consumption value of a vertex disjoint super-edge subset on the basis of the initial super-graph, the initial super-graph is converted into a conflict graph, and the subset of the vertex disjoint with the maximum edge is approximately obtained to obtain the virtual resource allocation result.
Thirdly, calculating the power consumption of a vertex corresponding to each server in the physical network according to the following formula:
ω(pi)＝-(Pmin+(Pmax-Pmin)×Ui)
wherein, ω (p)i) Representing the vertex p corresponding to the ith server in the physical networkiPower consumption of PminRepresenting the average power consumption, P, of all servers in the physical network in the idle statemaxRepresents the full power consumption, U, of all servers in the physical networkiAnd indicating the CPU core utilization rate of the ith server in the physical network.
The cpu core utilization rate of the ith server in the physical network is obtained by the following formula:
Figure BDA0003209167670000051
where Σ denotes a summation operation, MkTo represent
Figure BDA0003209167670000052
The total number of the middle virtual nodes, u represents the serial number of the virtual nodes in the virtual network after mapping, and u is more than or equal to 1 and less than or equal to Mk，
Figure BDA0003209167670000061
Representing the u-th virtual node of the k-th virtual network after mapping, wherein each virtual node in the virtual network represents a user,
Figure BDA0003209167670000062
representing a user
Figure BDA0003209167670000063
Cpu core number resource requirement of xk,u,iRepresents the successful probability, x, of the mapping of the u-th virtual node of the k-th virtual network to the i-th serverk,u,i1 indicates that the mapping was successful, otherwise, xk,u,i＝0，
Figure BDA0003209167670000064
Representing physicsAn ith server in the network,
Figure BDA0003209167670000065
to represent
Figure BDA0003209167670000066
The remaining cpu core number resources.
And fourthly, initializing an empty set, selecting the vertex with the maximum power consumption value in the vertex set of the conflict graph, putting the vertex into the set, deleting the vertex and the adjacent vertex from the vertex set of the conflict graph, and repeating the steps to obtain an initial independent set of the vertices stored in the set.
Fifthly, arranging vertexes in the initial independent set in an ascending order according to the power consumption values, selecting a vertex which is not selected in the initial independent set and has the minimum power consumption value, initializing an empty set, putting vertexes which are adjacent to the vertexes in the vertex set of the conflict graph into the set, and arranging the vertexes in a descending order according to the power consumption values of the vertexes to obtain an adjacent set of the vertexes stored in the set;
sixthly, using the vertex selected in the initial independent set as the central point of the claw image, and searching the adjacent set of the central point
Figure BDA0003209167670000067
A drawing; wherein the content of the first and second substances,
Figure BDA0003209167670000068
to represent
Figure BDA0003209167670000069
The amount of claw in (a) and (b),
Figure BDA00032091676700000610
will be provided with
Figure BDA00032091676700000611
Is set to 2.
Seventh, judging E (S)*) If < E (S) is true, if so, then use S*After S is replaced, the fifth step of the step is executed, otherwise, the step is executedLine step 4, where E (S)*) To represent
Figure BDA00032091676700000612
In the figure, claw replaces the original independent set S
Figure BDA00032091676700000613
Initial independent set S after the center point of the graph*And e (S) represents the total power consumption of all vertices in the initial independent set S.
And 4, completing the virtual network resource allocation.
Will S*Each vertex in (2) is converted into a corresponding server to complete the allocation of virtual network resources.
The effect of the present invention will be further described with reference to simulation experiments.
1. Simulation experiment conditions are as follows:
the hardware platform of the simulation experiment of the invention is as follows: the processor is Intel (R) core (TM) i5-7200U CPU with a main frequency of 2.70GHz and a memory of 128 GB.
The software platform of the simulation experiment of the invention is as follows: windows 10 operating system and MATLAB R2017 a.
2. Simulation content and result analysis thereof:
the simulation experiment of the invention is to adopt the invention and a prior art (hypergraph matching algorithm) to search an independent set for the formed initial hypergraph, and obtain a graph of the total power consumption of the server along with the change of the number of virtual networks.
In the simulation experiment, the adopted prior art hypergraph matching algorithm is as follows:
the hypergraph matching algorithm proposed by Long Zhang et al in "Virtual resource allocation for mobile edge computing: A hypergraph matching (IEEE) Global Communications Conference (GLOBECOM). IEEE 2019:1-6.
The simulation experiment selects a physical network with 20 servers, the number of CPU cores of each server is set to be 100, the probability of connection among the servers is 0.5, the bandwidth resource size of each physical link is uniformly distributed according to 200,300, the idle power consumption of each server is set to be 100, the power consumption when the server is fully loaded is set to be 1000, and the basic cost of the server accounts for 7 CPU cores.
The number of nodes of each virtual network selected by a simulation experiment of the invention obeys the uniform distribution of [2,5], the number of CPU cores requested by each virtual node obeys the uniform distribution of [5,20], the probability of virtual links existing between each pair of virtual nodes is 0.5, the bandwidth request of each virtual link obeys the uniform distribution of [5,20], and the arrival of the virtual network obeys the poisson distribution. And mapping the corresponding super point of the virtual node in the virtual network to the corresponding super edge of the server in the physical network to form an initial super graph.
The effect of the present invention is further described below with reference to the simulation diagram of fig. 2.
By adopting the method of the invention and other methods, mapping simulation of the same virtual network is respectively carried out, a total power consumption value consumed by 20 servers is correspondingly generated every time one virtual network is mapped, and the total power consumption values of the servers corresponding to 15 virtual networks are drawn into a graph 2. The abscissa in fig. 2 represents the number of virtual networks, and the ordinate represents the total power consumption of the server. Fig. 2 shows that 3 different methods are adopted by using 3 different broken lines, and the broken line marked by an asterisk in fig. 2 represents the trend of the total power consumption of the server along with the change of the number of the mapped virtual networks by using the virtual network resource allocation method based on the hypergraph matching algorithm in the prior art. The broken line marked by a plus sign in fig. 2 represents that the virtual network resource in the boxing algorithm of the prior art is firstly adapted to the ff (first) method, and the change trend of the total power consumption of the server along with the number of the mapped virtual networks is obtained. The broken line marked by a circle in fig. 2 represents that the virtual network resource in the boxing algorithm of the prior art is adopted to circularly adapt to the nf (next fit) method for the first time, and the change trend of the total power consumption of the server along with the number of the mapped virtual networks is obtained.
As can be seen from fig. 2, when the number of the virtual networks is less than 9, the broken line marked with an asterisk is slightly higher than other broken lines, and when the number of the virtual networks is greater than 9, the broken line marked with an asterisk is significantly lower than other broken lines, which indicates that the total power consumption of the server in the method of the present invention is lower than that of other existing methods, so that the method of the present invention can effectively reduce the total power consumption of the server compared with other methods in the prior art.

Claims (2)

1. A virtual network resource allocation method based on a hypergraph matching algorithm is characterized in that an initial hypergraph is constructed on the basis that a hypergraph represents the mapping relation of a virtual network to a physical network; the resource allocation method comprises the following specific steps:
step 1, constructing a virtual network and a physical network:
(1a) based on graph theory, K virtual networks are modeled, K is more than or equal to 1 and less than or equal to 20, and a weighted undirected graph is used
Figure DEST_PATH_IMAGE002
Expressing the topology of the kth virtual network node and link, wherein K is more than or equal to 1 and less than or equal to K;
(1b) based on graph theory, a physical network is modeled and weighted undirected graph G is usedsRepresenting the topology of the physical network nodes and links;
step 2, mapping the virtual network to the physical network:
(2a) randomly generating a 1 x d set, wherein each element in the set is a pre-mapping super edge of each pre-mapping super point, each super point represents a virtual node in a pre-mapping virtual network, each super edge represents a server in a physical network, and d represents the total number of the super edges mapped to the physical network by the super points in the virtual network;
(2b) if the residual cpu core number of each element in the set is greater than the cpu core number of the super point, the success probability of the super point mapping to the super edge is 1, otherwise, the probability is 0;
(2c) selecting an unmapped virtual link, searching a shortest path between two super edges of two super points at two ends of the selected virtual link by using a Floyd algorithm, and mapping the selected virtual link to the shortest path;
(2d) judging whether all the virtual links are mapped, if so, executing the step 3, otherwise, executing the step (2 c);
and 3, generating an independent set:
(3a) mapping the corresponding super point of the virtual node in the virtual network to the corresponding super edge of the server in the physical network to form an initial super graph;
(3b) correspondingly generating a vertex in the conflict graph by each hyper-edge in the initial hyper-graph, and if the hyper-edges have crossed hyper-points, forming an edge between the corresponding vertices in the conflict graph;
(3c) calculating the power consumption of a vertex corresponding to each server in the physical network according to the following formula:
ω(pi)＝-(Pmin+(Pmax-Pmin)×Ui)
wherein, ω (p)i) Representing the vertex p corresponding to the ith server in the physical networkiPower consumption of (P)minRepresents the average power consumption, P, of all servers in the physical network in the idle statemaxRepresenting full power consumption, U, of all servers in a physical networkiRepresenting the CPU core number utilization rate of the ith server in the physical network;
(3d) initializing an empty set, selecting a vertex with the maximum power consumption value in the vertex set of the conflict graph, putting the vertex into the set, deleting the vertex and the adjacent vertex from the vertex set of the conflict graph, and so on to obtain an initial independent set of the vertices stored into the set;
(3e) arranging vertexes in the initial independent set in an ascending order according to the power consumption values, selecting a vertex which is not selected in the initial independent set and has the minimum power consumption value, initializing an empty set, putting vertexes which are adjacent to the vertexes in the conflict graph vertex set into the set, and arranging the vertexes in a descending order according to the power consumption values of the vertexes to obtain an adjacent set of the vertexes stored in the set;
(3f) using the vertex selected in the initial independent set as the central point of the claw graph, and finding out the adjacent set of the central point
Figure FDA0003209167660000021
A drawing; wherein, the first and the second end of the pipe are connected with each other,
Figure FDA0003209167660000022
represent
Figure FDA0003209167660000023
The amount of claw in (a) and (b),
Figure FDA0003209167660000024
(3g) judgment of E (S)*) If < E (S) is true, if so, then use S*Step (3E) is executed after S is replaced, otherwise, step 4 is executed, wherein E (S)*) Represent
Figure FDA0003209167660000025
In the figure, claw replaces the original independent set S
Figure FDA0003209167660000026
Initial independent set S after the center point of the graph*The total power consumption of all the vertexes in the initial independent set S is represented by E (S);
step 4, completing the virtual network resource allocation:
will S*Each vertex in (1) is converted into a corresponding server to complete the allocation of virtual network resources.
2. The method for allocating virtual network resources based on hypergraph matching algorithm of claim 1, wherein the cpu core utilization of the ith server in the physical network in step (3c) is obtained by the following formula:
Figure FDA0003209167660000027
where Σ denotes a summation operation, MkRepresent
Figure FDA0003209167660000028
The total number of the middle virtual nodes, u represents the serial number of the virtual nodes in the virtual network after mapping, and u is more than or equal to 1 and less than or equal to Mk，
Figure FDA0003209167660000029
A u-th virtual node representing the mapped k-th virtual network, each virtual node in the virtual network representing a user,
Figure FDA00032091676600000210
representing a user
Figure FDA00032091676600000211
Cpu core number resource requirement of (x)k,u,iRepresenting the probability, x, of successful mapping of the u-th virtual node of the k-th virtual network to the i-th serverk,u,i1 indicates that the mapping was successful, otherwise, xk,u,i＝0，
Figure FDA0003209167660000031
Representing the ith server in the physical network,
Figure FDA0003209167660000032
to represent
Figure FDA0003209167660000033
The remaining cpu core number resources.
CN202110925564.6A 2021-08-12 2021-08-12 Virtual network resource allocation method based on hypergraph matching algorithm Active CN113645076B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202110925564.6A CN113645076B (en) 2021-08-12 2021-08-12 Virtual network resource allocation method based on hypergraph matching algorithm

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202110925564.6A CN113645076B (en) 2021-08-12 2021-08-12 Virtual network resource allocation method based on hypergraph matching algorithm

Publications (2)

Publication Number Publication Date
CN113645076A CN113645076A (en) 2021-11-12
CN113645076B true CN113645076B (en) 2022-06-03

Family

ID=78421311

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202110925564.6A Active CN113645076B (en) 2021-08-12 2021-08-12 Virtual network resource allocation method based on hypergraph matching algorithm

Country Status (1)

Country Link
CN (1) CN113645076B (en)

Families Citing this family (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN115052002B (en) * 2022-05-31 2024-02-06 成都亚光电子股份有限公司 Virtual network resource allocation method based on optimal adaptation and shortest path algorithm
CN116232921B (en) * 2023-05-08 2023-09-12 中国电信股份有限公司四川分公司 Deterministic network data set construction device and method based on hypergraph

Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN109041236A (en) * 2018-08-23 2018-12-18 北京邮电大学 A kind of wireless resource allocation methods and device of difference weight business
CN111586697A (en) * 2020-04-29 2020-08-25 电子科技大学 Channel resource allocation method based on directed hyper-graph greedy coloring
CN113163408A (en) * 2021-02-05 2021-07-23 北京中电飞华通信有限公司 Wireless resource allocation method and device

Family Cites Families (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US8812653B2 (en) * 2010-08-05 2014-08-19 Novell, Inc. Autonomous intelligent workload management

Patent Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN109041236A (en) * 2018-08-23 2018-12-18 北京邮电大学 A kind of wireless resource allocation methods and device of difference weight business
CN111586697A (en) * 2020-04-29 2020-08-25 电子科技大学 Channel resource allocation method based on directed hyper-graph greedy coloring
CN113163408A (en) * 2021-02-05 2021-07-23 北京中电飞华通信有限公司 Wireless resource allocation method and device

Non-Patent Citations (2)

* Cited by examiner, † Cited by third party
Title
"V2V通信中基于超图理论的资源分配算法";张家波等;《计算机工程与设计》;20200930;第2408-2413页 *
"Virtual Resource Allocation for Mobile Edge Computing: A Hypergraph Matching Approach";Long Zhang等;《2019 IEEE Global Communications Conference (GLOBECOM)》;20200227;第1-6页 *

Also Published As

Publication number Publication date
CN113645076A (en) 2021-11-12

Similar Documents

Publication Publication Date Title
CN113645076B (en) Virtual network resource allocation method based on hypergraph matching algorithm
CN113193984B (en) Air-space-ground integrated network resource mapping method and system
CN113110933B (en) System with Nginx load balancing technology
WO2022001086A1 (en) Efficient gpu resource allocation optimization method and system
CN111556516B (en) Distributed wireless network task cooperative distribution method facing delay and energy efficiency sensitive service
CN110347515B (en) Resource optimization allocation method suitable for edge computing environment
CN112202599B (en) Topology-aware mapping method and system for heterogeneous multi-core platform communication optimization
CN105357322A (en) Virtual machine distribution method based on topology partition
CN111182037A (en) Mapping method and device of virtual network
CN112235344A (en) Distributed machine learning-oriented sparse communication model implementation method
CN103825946A (en) Virtual machine placement method based on network perception
CN104901999A (en) Method for reliability perceptive cross-domain mapping of virtual data center
CN109412865B (en) Virtual network resource allocation method, system and electronic equipment
Liu et al. Traffic-aware and reliability-guaranteed virtual machine placement optimization in cloud datacenters
CN113900810A (en) Distributed graph processing method, system and storage medium
CN114327811A (en) Task scheduling method, device and equipment and readable storage medium
CN108833162B (en) Network node sequencing method and virtual network mapping method
CN110689175A (en) Energy consumption optimization method for distributed green cloud data center with chaotic multiple universes
CN111538560B (en) Virtual machine deployment method and device, electronic equipment and storage medium thereof
CN115983032A (en) Digital twin-assisted edge computing resource allocation method
CN113377866A (en) Load balancing method and device for virtualized database proxy service
Tang et al. A classification-based virtual machine placement algorithm in mobile cloud computing
CN106227600A (en) A kind of multidimensional virtual resource allocation method based on Energy-aware
CN107360210B (en) Virtual machine allocation method for cloud computing data center considering energy consumption and access delay
Yao et al. An energy efficient virtual machine placement algorithm based on graph partitioning in 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