CN115052002B - Virtual network resource allocation method based on optimal adaptation and shortest path algorithm - Google Patents
Virtual network resource allocation method based on optimal adaptation and shortest path algorithm Download PDFInfo
- Publication number
- CN115052002B CN115052002B CN202210614860.9A CN202210614860A CN115052002B CN 115052002 B CN115052002 B CN 115052002B CN 202210614860 A CN202210614860 A CN 202210614860A CN 115052002 B CN115052002 B CN 115052002B
- Authority
- CN
- China
- Prior art keywords
- virtual
- virtual network
- servers
- network
- mapping
- 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 24
- 238000013468 resource allocation Methods 0.000 title claims abstract description 23
- 230000006978 adaptation Effects 0.000 title claims abstract description 8
- 238000013507 mapping Methods 0.000 claims abstract description 27
- 230000001174 ascending effect Effects 0.000 claims description 3
- 238000012163 sequencing technique Methods 0.000 claims description 3
- 238000005265 energy consumption Methods 0.000 description 22
- 238000004088 simulation Methods 0.000 description 11
- 238000004891 communication Methods 0.000 description 3
- 230000007547 defect Effects 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 238000007726 management method Methods 0.000 description 2
- 230000003044 adaptive effect Effects 0.000 description 1
- 238000004458 analytical method Methods 0.000 description 1
- 238000013528 artificial neural network Methods 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 230000007246 mechanism Effects 0.000 description 1
- 230000006855 networking Effects 0.000 description 1
- 238000005457 optimization Methods 0.000 description 1
- 238000012545 processing Methods 0.000 description 1
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L67/00—Network arrangements or protocols for supporting network services or applications
- H04L67/01—Protocols
- H04L67/10—Protocols in which an application is distributed across nodes in the network
- H04L67/104—Peer-to-peer [P2P] networks
- H04L67/1074—Peer-to-peer [P2P] networks for supporting data block transmission mechanisms
-
- 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
- H04L41/145—Network analysis or design involving simulating, designing, planning or modelling of a network
Abstract
The invention discloses a virtual network resource allocation method based on an optimal adaptation and shortest path algorithm, which comprises the following implementation steps: the method comprises the steps of constructing a virtual network of each user, mapping each virtual node in each virtual network onto one server of a physical network by adopting an optimal applicable algorithm, mapping each virtual link in each virtual network onto one path between the servers of the physical network by adopting a shortest path algorithm, wherein the path is the shortest path between the servers respectively mapped by two virtual nodes of the virtual link, and taking all virtual nodes carried by the server after mapping as a virtual network resource allocation result.
Description
Technical Field
The invention belongs to the technical field of computers, and further relates to a virtual network resource allocation method based on an optimal adaptation and shortest path algorithm in the technical field of data communication. The invention can be applied to reasonably distributing virtual network resources in a data center environment.
Background
Currently, cloud computing has been widely used for real-time processing of large-scale data, physical resource management of a cloud data center can shield isomerism and complexity of underlying resources through a virtualization technology, and virtualized resources can be allocated to users as required according to requirements of the users. The huge demands of smart cities, cloud computing and the like continuously stimulate the increase of data volume, so that the increase of power consumption of the data center is promoted, and the power consumption of the data center is expected to be increased at a high speed in the future. However, most of the existing virtual network resource allocation schemes only focus on the receiving rate, the resource utilization rate and the cost ratio of the virtual network, and little focus on the energy consumption index, so that the energy consumption is increased in spite of the improvement of other performance indexes. Meanwhile, the existing resource allocation scheme reduces the energy consumption of the system at the cost of losing the fairness of virtual resources.
The China Union network communication group limited company discloses a virtual network resource allocation method in the proprietary technology of 'a mapping method and device of a virtual network' (application number: 2019112849534 grant bulletin number: CN 111182037B). The method comprises the following implementation steps: acquiring a set of physical nodes which can be mapped by virtual nodes in a virtual network and bandwidth resource requirements of virtual links; and a second step of: determining the weight of a first physical node which can be mapped by a target virtual node; and a third step of: the target virtual node is any virtual node in the virtual network, and the first physical node is any physical node which can be mapped by the target virtual node; fourth step: and determining the first physical node with the maximum weight as a mapping node of the target virtual node, and sequentially carrying out routing and bandwidth resource allocation meeting preset conditions among mapping nodes corresponding to the virtual nodes at two ends of the virtual link according to the sequence from large to small of the bandwidth resource demand of the virtual link. The method has the defects that the priority is given to the virtual node by adopting a mode of determining the weight of the first physical node which can be mapped by the target virtual node, so that the fairness of the virtual network can be damaged, the virtual resource is improperly distributed, and the energy consumption of the system is increased.
Zhiyuan Wang, chao Guo et al in its published paper "Frequency-Adaptive VDC Embedding to Minimize Energy Consumption of Data Centers" (IEEE 2021-Transactions on Green Communications and Networking) propose a Frequency-adaptive virtual resource allocation method based on a dynamic Frequency scaling mechanism. The method comprises the following implementation steps: the frequency of each hardware can be adaptively adjusted according to the given service requirement, and the total energy consumption of the direct current hardware is reduced to the minimum when a specific virtual network is mapped; and a second step of: an integer linear programming optimization model and an effective heuristic algorithm are provided, so that energy consumption is effectively reduced. Although the method reduces the energy consumption of the data center system, the method still has the defect that the adoption of a mode that the frequency can be adaptively adjusted according to the given service requirement can negatively influence the CPU performance, and the receiving rate of the virtual network is reduced.
Disclosure of Invention
The invention aims to solve the problems that the resource allocation of the virtual network is improper and the energy consumption of the system is increased due to the fact that the fairness of the virtual network is damaged in the resource allocation of the virtual network, and the adoption frequency can be adaptively adjusted according to the given service requirement, which can negatively affect the CPU performance and reduce the receiving rate of the virtual network.
The invention adopts the best applicable algorithm to map each virtual node in the virtual network, maps each virtual node to one server of the data center, and solves the problem of high energy consumption caused by improper virtual resource allocation for the server with the minimum difference between the residual CPU core number resources of each server and the CPU core number resources of the virtual node. The invention adopts the shortest path algorithm to map each virtual link in the virtual network, maps each virtual link to one path between the data center servers, and solves the problem of low receiving rate of the virtual network because the path is the shortest path between the servers respectively mapped by two virtual nodes of the virtual link.
The technical scheme for realizing the purpose of the invention comprises the following steps:
step 1, constructing a virtual network of each user:
constructing a virtual network for each user task, and representing the topology of each virtual network node and link by using a weighted undirected graph, wherein each virtual node represents the CPU core number resource requirement of a user, and each virtual link represents the bandwidth resource requirement of the user;
step 2, selecting an unmapped virtual network:
step 3, selecting an unmapped virtual node from the selected virtual network;
step 4, mapping the selected virtual nodes by adopting an optimal applicable algorithm;
mapping the selected virtual nodes in sequence according to the serial numbers of the servers, mapping the selected virtual nodes into the servers when the CPU core numbers of the selected virtual nodes are smaller than or equal to the residual CPU core numbers of the servers, updating the residual CPU core number resources of the servers by using the difference value between the residual CPU core numbers of the servers and the CPU core numbers of the selected virtual nodes, sequencing the updated residual CPU core numbers of the servers according to the ascending order of the residual CPU core numbers of all the servers, and renumbering all the servers;
step 5, judging whether all virtual nodes in the selected virtual network are mapped, if yes, executing step 6, otherwise, executing step 3;
step 6, selecting an unmapped virtual link from the selected virtual network;
step 7, mapping the selected virtual link by adopting a shortest path algorithm;
mapping the selected virtual link to a path between servers in a physical network, wherein the path is the shortest path between servers to which two virtual nodes of the virtual link are respectively mapped;
step 8, judging whether all virtual links in the selected virtual network are mapped, if yes, executing step 9, otherwise, executing step 6;
step 9, judging whether all virtual networks are mapped, if yes, executing step 10, otherwise, executing step 2;
step 10, completing virtual network resource allocation:
and taking all virtual nodes borne by the server in the physical network after the mapping is completed as a virtual network resource allocation result.
Compared with the prior art, the invention has the following advantages:
firstly, because the invention adopts the best-fit algorithm to map each virtual node in the virtual network, the opened servers are utilized as much as possible, the unused servers are adjusted to the energy-saving sleep mode and even are completely closed, the problem of excessively high energy consumption caused by improper resource allocation in the prior art is overcome, the number of the servers occupied by the virtual network resource allocation is reduced as much as possible, thereby reducing the system energy consumption of the data center and improving the resource utilization rate of the data center.
Second, because the invention adopts the shortest path algorithm, map each virtual link in the virtual network to a route between the data center servers, this route is two virtual nodes of the virtual link to the shortest route between the servers separately, overcome the low disadvantage of the virtual network receiving rate in the prior art, make the invention reduce the physical bandwidth that the resource allocation of the virtual network occupies, thus has raised the receiving rate of the virtual network.
Drawings
FIG. 1 is a flow chart of the present invention;
fig. 2 is a simulation diagram of the present invention.
Detailed Description
The invention is described in further detail below with reference to the drawings and examples.
The specific steps of the implementation of the present invention will be described in further detail with reference to fig. 1.
Step 1, constructing a virtual network:
a virtual network is constructed for each user task, and the topology of each virtual network node and link is represented by a weighted undirected graph.
The embodiment of the invention constructs 20 virtual networks with different structures, the number of virtual nodes contained in the weighted undirected graph corresponding to each virtual network is uniformly distributed according to [2,4], the number of CPU cores requested by each virtual node is uniformly distributed according to [5,20], the probability of existence of virtual links between each pair of virtual nodes is 0.5, and the bandwidth request of each virtual link is uniformly distributed according to [5,20 ].
Step 2, selecting an unmapped virtual network;
step 3, selecting an unmapped virtual node from the selected virtual network;
step 4, mapping the selected virtual nodes by adopting an optimal applicable algorithm;
mapping the selected virtual nodes in sequence according to the serial numbers of the servers, mapping the selected virtual nodes into the servers when the CPU core numbers of the selected virtual nodes are smaller than or equal to the residual CPU core numbers of the servers, updating the residual CPU core number resources of the servers by using the difference value between the residual CPU core numbers of the servers and the CPU core numbers of the selected virtual nodes, sequencing the updated residual CPU core numbers of the servers according to the ascending order of the residual CPU core numbers of all the servers, and renumbering all the servers;
step 5, judging whether all virtual nodes in the selected virtual network are mapped, if yes, executing step 6, otherwise, executing step 3;
step 6, selecting an unmapped virtual link from the selected virtual network;
step 7, mapping the selected virtual link by adopting a shortest path algorithm;
mapping the selected virtual link to a path between servers in a physical network, wherein the path is the shortest path between servers to which two virtual nodes of the virtual link are respectively mapped;
step 8, judging whether all virtual links in the selected virtual network are mapped, if yes, executing step 9, otherwise, executing step 6;
step 9, judging whether all virtual networks are mapped, if yes, executing step 10, otherwise, executing step 2;
step 10, completing virtual network resource allocation:
and taking all virtual nodes borne by the server in the physical network after the mapping is completed as a virtual network resource allocation result.
The effects of the present invention will be further described with reference to simulation experiments.
1. Simulation experiment conditions:
the hardware platform of the simulation experiment of the invention is: the processor is Intel (R) Core (TM) i5-7200U CPU, the main frequency is 2.70GHz, and the memory is 128GB.
The software platform of the simulation experiment of the invention is: windows 10 operating system and MATLAB R2017a. The virtual nodes are mapped by adopting the best-fit algorithm.
According to the simulation experiment, a physical network with 30 servers is selected, the number of CPU cores of each server is set to be 100, the probability of connection between 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 150, the power consumption in full load is set to be 1500, and the basic cost of the server occupies 7 CPU cores.
2. Simulation content and result analysis:
the simulation experiment of the invention adopts the invention and a prior art (optimal adaptation algorithm) to map the virtual nodes according to the serial numbers of the servers in sequence, so as to obtain a change graph of the total energy consumption of the servers along with the number of the virtual networks.
In the simulation experiment, one prior art best-fit algorithm used refers to:
the best adaptation algorithm proposed by Farzad Haubi et al in "Accelerating Virtual Network Embedding with Graph Neural Networks" (2020 16th International Conference on Network and Service Management (CNSM), 2020:1-9).
The number of nodes of each virtual network selected by the simulation experiment is uniformly distributed according to [2,5], the number of CPU cores requested by each virtual node is uniformly distributed according to [5,20], the probability of virtual links between each pair of virtual nodes is 0.5, the bandwidth request of each virtual link is uniformly distributed according to [5,20], and the arrival of the virtual network is uniformly distributed according to poisson.
The effects of the present invention are further described below in conjunction with the simulation diagram of fig. 2.
The method of the invention is adopted to simulate the mapping of the same virtual network respectively with other methods, and each time one virtual network is mapped, an energy consumption value consumed by 30 servers is correspondingly generated, the total energy consumption value of the servers corresponding to 20 virtual networks is drawn into figure 2, and the total energy consumption value is the sum of the energy consumption values correspondingly generated by each virtual network after the mapping is completed. The abscissa in fig. 2 represents the number of virtual networks, and the ordinate represents the total server energy consumption. In fig. 2, two different broken lines represent two different methods, and the broken line marked by a square in fig. 2 represents a trend of variation of total server energy consumption with the number of mapped virtual networks, which is obtained by adopting the virtual network resource allocation method of the present invention. The broken line marked with an asterisk in fig. 2 represents the variation trend of the total energy consumption of the server along with the number of mapped virtual networks by adopting the virtual network resource adaptive NF (Next Fit) method in the prior art.
As can be seen from fig. 2, the broken line marked with square is significantly lower than the broken line marked with asterisk, which indicates that the total energy consumption of the server in the method of the present invention is lower than that in other prior art methods, so that the method of the present invention can effectively reduce the total energy consumption of the server compared with other prior art methods.
Claims (1)
1. The virtual network resource allocation method based on the optimal adaptation and shortest path algorithm is characterized in that each virtual node in the virtual network is mapped by utilizing the optimal adaptation algorithm; mapping each virtual link in the virtual network by utilizing a shortest path algorithm; the resource allocation method comprises the following specific steps:
step 1, constructing a virtual network of each user:
constructing a virtual network for each user task, and representing the topology of each virtual network node and link by using a weighted undirected graph, wherein each virtual node represents the CPU core number resource requirement of a user, and each virtual link represents the bandwidth resource requirement of the user;
step 2, selecting an unmapped virtual network:
step 3, selecting an unmapped virtual node from the selected virtual network;
step 4, mapping the selected virtual nodes by adopting an optimal applicable algorithm;
mapping the selected virtual nodes in sequence according to the serial numbers of the servers, mapping the selected virtual nodes into the servers when the CPU core numbers of the selected virtual nodes are smaller than or equal to the residual CPU core numbers of the servers, updating the residual CPU core number resources of the servers by using the difference value between the residual CPU core numbers of the servers and the CPU core numbers of the selected virtual nodes, sequencing the updated residual CPU core numbers of the servers according to the ascending order of the residual CPU core numbers of all the servers, and renumbering all the servers;
step 5, judging whether all virtual nodes in the selected virtual network are mapped, if yes, executing step 6, otherwise, executing step 3;
step 6, selecting an unmapped virtual link from the selected virtual network;
step 7, mapping the selected virtual link by adopting a shortest path algorithm;
mapping the selected virtual link to a path between servers in a physical network, wherein the path is the shortest path between servers to which two virtual nodes of the virtual link are respectively mapped;
step 8, judging whether all virtual links in the selected virtual network are mapped, if yes, executing step 9, otherwise, executing step 6;
step 9, judging whether all virtual networks are mapped, if yes, executing step 10, otherwise, executing step 2;
step 10, completing virtual network resource allocation:
and taking all virtual nodes borne by the server in the physical network after the mapping is completed as a virtual network resource allocation result.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202210614860.9A CN115052002B (en) | 2022-05-31 | 2022-05-31 | Virtual network resource allocation method based on optimal adaptation and shortest path algorithm |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202210614860.9A CN115052002B (en) | 2022-05-31 | 2022-05-31 | Virtual network resource allocation method based on optimal adaptation and shortest path algorithm |
Publications (2)
Publication Number | Publication Date |
---|---|
CN115052002A CN115052002A (en) | 2022-09-13 |
CN115052002B true CN115052002B (en) | 2024-02-06 |
Family
ID=83160228
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202210614860.9A Active CN115052002B (en) | 2022-05-31 | 2022-05-31 | Virtual network resource allocation method based on optimal adaptation and shortest path algorithm |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN115052002B (en) |
Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN113630328A (en) * | 2021-07-09 | 2021-11-09 | 西安电子科技大学 | Data center virtual network mapping method and system |
CN113645076A (en) * | 2021-08-12 | 2021-11-12 | 西安电子科技大学 | Virtual network resource allocation method based on hypergraph matching algorithm |
-
2022
- 2022-05-31 CN CN202210614860.9A patent/CN115052002B/en active Active
Patent Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN113630328A (en) * | 2021-07-09 | 2021-11-09 | 西安电子科技大学 | Data center virtual network mapping method and system |
CN113645076A (en) * | 2021-08-12 | 2021-11-12 | 西安电子科技大学 | Virtual network resource allocation method based on hypergraph matching algorithm |
Non-Patent Citations (1)
Title |
---|
Accelerating Virtual Network Embedding with Graph Neural Networks;Farzad Habibi等;2020 16th International Conference on Network and Service Management(CNSM);第1-9页 * |
Also Published As
Publication number | Publication date |
---|---|
CN115052002A (en) | 2022-09-13 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN103294521B (en) | A kind of method reducing data center's traffic load and energy consumption | |
CN111538587B (en) | Service function chain reconfiguration method based on load balancing | |
CN111694636B (en) | Electric power Internet of things container migration method oriented to edge network load balancing | |
CN108923979B (en) | Software defined network virtual network mapping method | |
CN104734915B (en) | A kind of concurrent dynamic emulation method of Multi net voting of compound multi-process multithreading | |
CN106028463B (en) | A kind of spaceborne dynamic spectrum resource regulating method based on service speed control | |
CN107196806B (en) | Topological proximity matching virtual network mapping method based on sub-graph radiation | |
WO2022001086A1 (en) | Efficient gpu resource allocation optimization method and system | |
CN114374609B (en) | Deep learning job operation method and system based on RDMA equipment | |
Wu et al. | A mobile edge computing-based applications execution framework for Internet of Vehicles | |
CN113645076B (en) | Virtual network resource allocation method based on hypergraph matching algorithm | |
CN113315806B (en) | Multi-access edge computing architecture for cloud network fusion | |
CN115052002B (en) | Virtual network resource allocation method based on optimal adaptation and shortest path algorithm | |
CN111698703B (en) | Network reliability optimization method based on service priority and load balancing | |
CN111756655B (en) | Virtual network resource migration method based on resource reservation | |
CN111245701B (en) | Link priority virtual network mapping method based on maximum weighted matching | |
CN110351199A (en) | Flow smoothing method, server and forwarding device | |
CN116909738A (en) | Neural network call simplification method based on distributed resource pool | |
CN116684291A (en) | Service function chain mapping resource intelligent allocation method suitable for generalized platform | |
Yuan et al. | A DRL-Based Container Placement Scheme with Auxiliary Tasks. | |
CN104468379B (en) | Virtual Hadoop clustered nodes system of selection and device based on most short logical reach | |
CN111954230A (en) | Computing migration and resource allocation method based on fusion of mobile edge computing and dense cloud access network | |
CN115617526A (en) | Cloud data center energy-saving method based on cloud data center construction and virtual machine integration | |
CN109254844A (en) | A kind of triangle calculation method of Large Scale Graphs | |
He et al. | QoS-Aware and Resource-Efficient Dynamic Slicing Mechanism for Internet of Things. |
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 |