CN110061881B - Energy consumption perception virtual network mapping algorithm based on Internet of things - Google Patents
Energy consumption perception virtual network mapping algorithm based on Internet of things Download PDFInfo
- Publication number
- CN110061881B CN110061881B CN201910339562.1A CN201910339562A CN110061881B CN 110061881 B CN110061881 B CN 110061881B CN 201910339562 A CN201910339562 A CN 201910339562A CN 110061881 B CN110061881 B CN 110061881B
- Authority
- CN
- China
- Prior art keywords
- node
- mapping
- virtual
- link
- physical
- 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.)
- Expired - Fee Related
Links
- 238000013507 mapping Methods 0.000 title claims abstract description 154
- 238000005265 energy consumption Methods 0.000 title claims abstract description 87
- 230000008447 perception Effects 0.000 title claims abstract description 28
- 238000000034 method Methods 0.000 claims abstract description 22
- 108090000623 proteins and genes Proteins 0.000 claims description 15
- 230000008569 process Effects 0.000 claims description 9
- 230000001174 ascending effect Effects 0.000 claims description 3
- 230000006870 function Effects 0.000 claims description 3
- 230000000694 effects Effects 0.000 abstract description 5
- 230000010354 integration Effects 0.000 abstract description 3
- 238000005516 engineering process Methods 0.000 description 7
- 230000005540 biological transmission Effects 0.000 description 4
- 238000011161 development Methods 0.000 description 4
- 230000008901 benefit Effects 0.000 description 3
- CURLTUGMZLYLDI-UHFFFAOYSA-N Carbon dioxide Chemical compound O=C=O CURLTUGMZLYLDI-UHFFFAOYSA-N 0.000 description 2
- 230000008859 change Effects 0.000 description 2
- 238000007726 management method Methods 0.000 description 2
- 238000012545 processing Methods 0.000 description 2
- 239000000758 substrate Substances 0.000 description 2
- 230000002776 aggregation Effects 0.000 description 1
- 238000004220 aggregation Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 229910002092 carbon dioxide Inorganic materials 0.000 description 1
- 239000001569 carbon dioxide Substances 0.000 description 1
- 239000003245 coal Substances 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 230000002596 correlated effect Effects 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 238000002955 isolation Methods 0.000 description 1
- 238000012423 maintenance Methods 0.000 description 1
- 230000007246 mechanism Effects 0.000 description 1
- 238000012544 monitoring process Methods 0.000 description 1
- 230000006855 networking Effects 0.000 description 1
- 238000005457 optimization Methods 0.000 description 1
- 238000010248 power generation Methods 0.000 description 1
- 230000009467 reduction Effects 0.000 description 1
- 238000013468 resource allocation Methods 0.000 description 1
Images
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L12/00—Data switching networks
- H04L12/02—Details
- H04L12/12—Arrangements for remote connection or disconnection of substations or of equipment thereof
-
- 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
- H04L41/142—Network analysis or design using statistical or mathematical methods
-
- 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
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L47/00—Traffic control in data switching networks
- H04L47/70—Admission control; Resource allocation
- H04L47/76—Admission control; Resource allocation using dynamic resource allocation, e.g. in-call renegotiation requested by the user or requested by the network in response to changing network conditions
Landscapes
- Engineering & Computer Science (AREA)
- Computer Networks & Wireless Communication (AREA)
- Signal Processing (AREA)
- Physics & Mathematics (AREA)
- Algebra (AREA)
- General Physics & Mathematics (AREA)
- Mathematical Analysis (AREA)
- Mathematical Optimization (AREA)
- Mathematical Physics (AREA)
- Probability & Statistics with Applications (AREA)
- Pure & Applied Mathematics (AREA)
- Data Exchanges In Wide-Area Networks (AREA)
Abstract
An energy consumption perception virtual network mapping algorithm based on the Internet of things is characterized in that a traditional energy consumption model is redefined in a virtual network mapping stage according to network characteristics of the Internet of things, so that the energy consumption perception virtual network mapping algorithm is suitable for new energy consumption characteristics in the environment of the Internet of things. Under the condition of ensuring the minimum energy consumption, the heterogeneous performance of the nodes and the timeliness of the links are considered emphatically, and the mapping of the virtual network resources is completed by adopting a two-stage method through distinguishing the timeliness characteristics of the links. The invention solves the problem of virtual network mapping by means of resource integration, realizes the maximization of resource utilization rate, shortens mapping time, achieves the effect of energy saving, and realizes a finer-grained mapping mode along with the increase of introduced parameters.
Description
Technical Field
The invention relates to a virtual network mapping algorithm, in particular to an energy consumption perception virtual network mapping algorithm based on the Internet of things.
Background
The internet greatly improves the life of people from birth to the present, and the internet is developed rapidly in recent years. A large number of applications and a wide variety of networking technologies are currently running on the internet. An ubiquitous, worldwide, intercommunicating network has been formed, which is an important infrastructure and productivity of today's society, far beyond what was envisioned when it was first set up. With the development of network technology, everything interconnection is more keen. The Internet of Things (IoT) is a ubiquitous network established on the Internet, and is an Internet connected with objects, and information exchange and communication between the objects and the people are completed according to a certain protocol through various information sensing devices, so that a network for intelligent identification, positioning, tracking, monitoring and management is realized. With the development of the internet of things in the fields of security, traffic, logistics, medical treatment and the like, the application mode of the internet of things is becoming mature. In order to acquire accurate information, more and more nodes are needed by the interconnection of everything, the connected data are more and more huge, and the problem of energy consumption is highlighted. Therefore, reducing network energy consumption, reducing network operation and maintenance cost, reducing delay, and building a green energy-saving network becomes a hot spot problem of concern.
In order to deal with the problem of energy consumption in the environment of the internet of things, how to effectively organize and manage increasing physical resources, and how to meet diversified application requirements of the internet of things, academic circles and industrial circles continuously explore and finally draw conclusions. The most representative of which is network virtualization technology.
The Network virtualization technology is proposed in recent years to solve the problem of internet "stiffness", and creates several coexisting Virtual Networks (VN) on a shared physical Substrate Network (SN) through mechanisms such as resource abstraction, aggregation, isolation, and the like, so as to meet the diversified demands of different users on the networks. Network virtualization is generally viewed as a enabler of polymorphic internet and a cornerstone of future internet architectures due to its potential to offer in diversifying existing networks and ensuring coexistence of heterogeneous network architectures over a shared substrate. The basic entity of network virtualization is a virtual network, a virtual network being a virtual topology formed by a set of virtual nodes and virtual links. A plurality of logic networks on the same physical network belong to different service providers, infrastructure providers can provide different network topology resources for different virtual networks, and the use and management of different virtual networks are independent and do not influence each other. Through the network virtualization technology, the utilization rate of network resources can be effectively improved, and the method has important significance for solving the network rigidity problem.
Due to a large number of uncertain factors existing in the network in the operation process, such as the change of a cache queue of a node in the network, the change of the bandwidth and delay of a link, and the like, the resources in the physical network are dynamically changed, and uncertainty exists. How to efficiently manage and allocate the bottom layer resources, complete the processing of the virtual network request, provide the required network resources for the tenants, and ensure the service quality is a problem that needs to be solved in network virtualization. How to efficiently allocate and manage underlying physical Network resources and provide services for Virtual networks is widely referred to as Virtual Network Embedding (VNE). The virtual network mapping problem is a core problem of network virtualization and is also an NP-hard problem, and the virtual request is mapped to a bottom layer network through certain constraint, so that resource allocation of the bottom layer physical network is reasonably and efficiently completed. In the environment of the internet of things, for massive information acquired by a sensor, the correctness and timeliness of data must be ensured in the transmission process of a network, and the method must be suitable for various heterogeneous networks, so that the heterogeneity of nodes and the delay problem in link transmission must be considered in the virtual network mapping process. The quality of the virtual network mapping algorithm determines the number of virtual networks that the physical network can carry, and ultimately determines the benefit of the physical infrastructure provider. Therefore, in a network virtualization environment, the virtual network mapping algorithm is very important, and the efficient virtual network mapping algorithm can increase the profit of an infrastructure provider to the maximum extent and improve the utilization rate of physical resources.
The network virtualization technology achieves the purpose of fully sharing physical resources by effectively managing the mapping from the virtual user request to the physical resources, and embodies the advantages of the network virtualization technology when solving the problem of the internet of things. On the basis of the existing Internet virtual network mapping algorithm, aiming at the characteristics of heterogeneity presented by physical nodes in the Internet of things environment and the delay characteristic of a physical link, an energy consumption perception resource mapping algorithm based on the Internet of things is provided.
The traditional method based on energy consumption perception comprises the following steps: resource integration strategies, traffic expansion strategies, energy consumption minimization, etc. In terms of energy saving, most of operators achieve the effect of saving energy by turning off nodes and links under low power consumption to be in a dormant state on the premise of meeting the virtual network resource request. By means of the resource combination, distributed resources are distributed to the underlying network in a centralized mode, energy consumption can be reduced, rejection rate in virtual network mapping can be increased, and benefits are reduced. And then, establishing load Energy consumption models of the nodes and the links by proposing Energy consumption models of the nodes and the links and taking minimized mapping Energy consumption as a target, and reducing the Energy consumption by solving an Energy-Aware Virtual Network mapping Heuristic (EA-VNEH) algorithm. The above solutions to the problem of energy consumption have all focused on the economic aspect of energy consumption, i.e. maximizing the profitability by reducing the energy costs, but neglecting the ecological aspects related to energy utilization, such as carbon dioxide emissions from power generation. In recent years, with the concern of ecology, there has been a further breakthrough in solving the problem of energy consumption, which is reduced by classifying underlying resources into different types of energy resources and mapping virtual network requests onto the cleanest underlying network. The method considers the type of energy, adopts renewable resources to replace the prior coal resources, and achieves the effect of energy saving from the aspect of energy types.
The prior energy-saving method can be well applied to the Internet environment, but in the scene of the Internet of things, the information of an object is accurately transmitted in real time through a sensor by fusing various wired and wireless networks with the Internet. Since massive sensors are distributed in different regions, the heterogeneity of nodes is more prominent, and the initial state, the energy consumption rate and the time delay characteristics of each link are different, the existing energy consumption sensing method needs to be improved.
Disclosure of Invention
The invention aims to provide an energy consumption perception virtual network mapping algorithm based on the Internet of things on the basis of a traditional energy consumption model.
In order to achieve the purpose, the invention adopts the following technical scheme:
an energy consumption perception virtual network mapping algorithm based on the Internet of things comprises the following steps:
1) Building network model and energy consumption perception model
a. Defining an underlying physical network topology graph as a weighted undirected graph
Wherein N is s And L s Respectively representing the set of underlying physical network nodes and links,
and
respectively representing the attributes contained in the bottom layer physical node and the link;
defining a virtual network topology graph as a weighted undirected graph
Wherein, N v And L v Respectively representing a set of virtual network nodes and links,
and
respectively representing the attributes contained in the virtual nodes and the links;
weighted undirected graph G s And weighted undirected graph G v Forming a network model;
b. the energy consumption perception model is established as follows:
an objective function: min P N +P L
And (4) CPU resource constraint:
and (4) CPU position constraint:
CPU category constraint:
bandwidth resource constraints of the link:
delay constraints of the link:
and (3) connectivity constraint:
wherein,
the method belongs to a binary variable and represents the mapping relation between a virtual node j and a physical node i; cpu (j) represents a cpu value of the virtual node; cpu (i) represents a cpu value of the physical node; dis (loc (j), loc (M) n (j) ) represents the euclidean distance between the mapping node j and the surrounding nodes; gene (j) represents the node type of the virtual node; gene (i) represents the node type of the physical node;
representing a virtual link l uv With physical links l ij The mapping relationship between b (l) uv ) Representing a virtual link l uv Bandwidth resources of (a); b (l) ij ) Represents a physical link l ij Bandwidth resources of (a); d (l) uv ) Representing a virtual link l uv Delay of (2); d (l) ij ) Represents a physical link l ij Delay of (P) N Mapping the total energy consumption for the node; p L Mapping the total energy consumption for the link;
2) Finishing the mapping of each node in the virtual topological graph, and storing the successfully mapped physical nodes into a node mapping linked list;
3) And finishing the mapping of each link in the virtual topological graph according to the node mapping linked list, and storing the successfully mapped physical link into the link mapping linked list.
A further development of the invention consists in that, in step 1), when
When the mapping is successful, the virtual node j is mapped to the physical node i,
if so, the operation is unsuccessful;
when in use
Then, the virtual link l is illustrated uv Successful mapping to physical link/ ij In the above-mentioned manner,
if so, it is unsuccessful.
The invention is further improved in that the total energy consumption P of the node mapping in step 1) is N ：
Wherein N is w Indicating the number of host nodes that need to be changed from inactive to active, P b Representing the base power consumption, P, of the node l Representing the energy consumption of the forwarding node, util (n) j ) Represents the cpu utilization, S, required for mapping a virtual node j to a physical network node i i Representing the energy state of physical node i when unmapped.
In a further development of the invention, S i =1 denotes the state of node i when it is active, S i And =0 represents the other state.
The invention is further improved in that the link in step 1) maps the total energy consumption P L ：
Wherein, P n For a link where power consumption is a constant, d (i, j) represents the path distance from node i to node j, α represents a distance factor, N n Indicating the number of forwarding nodes that need to be changed from an inactive state to an active state,
representing a virtual link l uv With physical links l ik The mapping relationship between them.
The invention is further improved when
Then, the virtual link l is illustrated uv Successful mapping to physical link/ ik In the above-mentioned manner,
if so, it is unsuccessful.
The invention is further improved in that the specific process of the step 2) is as follows:
a. virtual node n of a computational network model v Resource capability NR (n) v ) And NR (n) v ) Arranging in descending order;
wherein:
representing the topological relation between the mapping node and the surrounding nodes;
b. aiming at the virtual node n arranged in the step a v Selecting physical nodes meeting the CPU resource constraint and the position constraint of the virtual nodes to obtain a candidate mapping node set omega (n) v )；
Ω(n v )＝{n s |cpu(n v )≤cpu(n s )&dis(loc(n v ),loc(n s ))≤D(n s ),n s ∈N s }
c. For a set of candidate mapping nodes Ω (n) v ) Each candidate node n in s According to their resource capabilities NR (n) s ) And energy consumption P of the node N (n s ) Calculating the comprehensive resource capacity of the nodes and arranging in descending order:
CR(n s )＝log(NR(n s )·P N (n s )+γ)
wherein γ → 0, avoids the case where 0 occurs inside log in the above formula;
d. judging the type of the nodes arranged in the step c, if true (n) v )＝genre(n s ) Will virtualize node n v Mapping to physical node n with minimal CR and satisfying constraints s The above step (1); wherein, gene (n) v ) A node type representing a virtual node; gene (n) s ) Representing a type of physical node; if gene (n) v )≠genre(n s ) Returning to the step b, in the candidate mapping node set omega (n) v ) Other physical nodes are continuously searched circularly until the nodes with the same node type are found, and node mapping is completed;
e. calculating the residual CPU resource CPU (n) of the successfully mapped physical node s )＝cpu(n s )-cpu(n v )；
f. And (b) returning to the step (a), sequentially finishing the mapping of each node in the virtual topological graph, and storing the successfully mapped physical nodes into a node mapping linked list.
8. The energy consumption perception virtual network mapping algorithm based on the internet of things according to claim 7, wherein the specific process of the step 3) is as follows:
a. obtaining the bottom layer physical node n according to the node mapping chain table v (u)→n s (i)，n v (v)→n s (j) Wherein n is v (u)→n s (i) Representing the mapping of a virtual node u to a physical node i, n v (v)→n s (j) Representing the mapping of virtual node v to physical node j;
b. virtual link l uv Arranging according to the bandwidth requirement in a descending order;
c. aiming at each virtual link l arranged in descending order in step b uv Calculating a physical node n s (i) To n s (j) Set of shortest paths of
And for arbitrary paths
d. For shortest path set
Calculates the link energy consumption of each path
And arranging them in ascending order;
e. if the service type is low latency requirement, only the bandwidth requirement of the link is considered, if b (l) uv )≤b(l ij ) Virtual link l uv Mapping to link energy consumption
Minimum physical link l ij The above step (1); if the service type is high latency requirement, then not only the bandwidth requirement of the link is considered, but also the latency requirement of the link, if b (l) uv )≤b(l ij )&d(l uv )≥d(l ij ) Virtual link l uv Mapping to link energy consumption
Minimum physical link l ij Completing link mapping;
f. calculating the residual bandwidth resource b (l) of the physical link with successful mapping s )＝b(l s )-b(l v )；
g. And returning to the step b, sequentially finishing the mapping of each link in the virtual topological graph, and storing the successfully mapped physical links into a link mapping linked list.
A further improvement of the invention is that flag =1 indicates a high latency requirement; flag =0 indicates a low latency requirement.
The invention is further improved in that in step c, a k shortest path algorithm is adopted to calculate the physical node n s (i) To n s (j) Set of shortest paths of
And for arbitrary paths
Compared with the prior art, the invention has the following beneficial effects:
according to the invention, the virtual request is mapped to the minimum energy consumption and the mapping is completed based on the closest residual capacity (namely, the virtual request is mapped to the node/link with the minimum but enough capacity to meet the VNR) by constructing the energy consumption model, so that the number of the nodes for finally completing the mapping is relatively less, and the energy-saving effect is better achieved. The invention can reduce the energy consumption in the network mapping process and simultaneously meet the node isomerism and link timeliness in the environment of the Internet of things. The virtual network is mapped to the bottom layer network with the minimum energy consumption, and the optimization of the energy consumption is realized by reducing the number of working nodes, so that the resource utilization rate is realized to the maximum extent, the benefit-cost ratio is improved, and a resource mapping mode with finer granularity is realized along with the increase of introduced parameters.
Drawings
FIG. 1 is a split virtual request. Wherein, (a) is a virtual network topological graph, (b) is S-V1, and (c) is S-V2.
Fig. 2 is a flow chart of node mapping.
Fig. 3 is a link mapping flow chart.
Detailed Description
The present invention will be described in further detail with reference to the accompanying drawings.
In the environment of the internet of things, for massive information acquired by a sensor, the correctness and timeliness of data must be ensured in the transmission process of a network, and the method must be suitable for various heterogeneous networks, so that the heterogeneity of nodes and the delay problem in link transmission must be considered in the virtual network mapping process. Therefore, for the characteristics of the internet of things, the virtual network mapping algorithm based on the energy consumption perception of the internet of things is redefined as follows:
1) Building network model and energy consumption perception model
a. Defining an underlying physical network topology graph as a weighted undirected graph
Wherein N is s And L s Respectively representing the set of underlying physical network nodes and links,
and
representing the attributes contained by the underlying physical node and link, respectively. Each bottom node n s ∈N s And CPU processor capacity value CPU (n) s ) Associated, for the underlying link l s ∈L s And bandwidth capacity b (l) s ) Are correlated and are represented by d (l) s ) Indicating the delay attribute of each underlying link.
Defining a virtual network topology graph as a weighted undirected graph
Wherein N is v And L v Respectively representing a set of virtual network nodes and links,
and
respectively representing virtual nodesAnd the attributes contained by the link. Each virtual node n v ∈N v And CPU processor capacity value CPU (n) v ) Associated, each virtual link l v ∈L v And bandwidth capacity b (l) v ) And (4) associating. Virtual nodes and virtual links have the same attributes as nodes and links in the underlying physical network. In addition, for heterogeneous nodes, the category of the node is represented in the form of node category + sequence number, such as A 1 。
Weighted undirected graph G s And weighted undirected graph G v And forming a network model.
b. The energy consumption of the virtual network mapping consists of two parts of node energy consumption and link energy consumption. The energy consumption of the node is linearly related to the utilization rate of the CPU resource, and the energy consumption of the link is linearly related to the bandwidth utilization rate and the link length.
Defining node mapping total energy consumption as
Wherein N is w Representing the number of host nodes that need to be changed from an inactive (off) state to an active (on) state, P b Representing the base power consumption, P, of the node l Representing the energy consumption of the forwarding node, util (n) j ) Represents the cpu utilization, S, required for mapping a virtual node j to a physical network node i i Representing the energy state, S, of a physical node i when unmapped i =1 denotes the state when node i is active (on), S i And =0 represents the other state.
The link mapping total energy consumption is:
wherein, P n For the link power consumption to be a constant, d (i, j) represents the path distance from node i to node j, α represents the distance factor, N n Indicating the number of forwarding nodes that need to be changed from an inactive (off) state to an active (on) state,
representing a virtual link l uv With physical links l ik A mapping relationship between them when
Then, the virtual link l is illustrated uv Successful mapping to physical link/ ik In the above-mentioned manner,
if so, it is unsuccessful.
c. The energy consumption perception model is established as follows:
an objective function: min P N +P L
And (4) CPU resource constraint:
and (4) CPU position constraint:
CPU category constraint:
bandwidth resource constraints of the link:
delay constraints of the link:
connectivity constraint:
wherein,
belonging to binary variables, representingMapping relation between the virtual node j and the physical node i; when is coming into contact with
When the mapping is successful, the virtual node j is mapped to the physical node i,
if so, the operation is unsuccessful; cpu (j) represents a cpu value of the virtual node; cpu (i) represents a cpu value of the physical node; dis (loc (j), loc (M) n (j) ) represents the euclidean distance between the mapping node j and the surrounding nodes; gene (j) represents the node type of the virtual node; gene (i) represents the node type of the physical node;
representing a virtual link l uv With physical links l ij A mapping relationship between them when
Then, the virtual link l is illustrated uv Successful mapping to physical link/ ij In the above-mentioned manner,
if so, the operation is unsuccessful; b (l) uv ) Representing a virtual link l uv Bandwidth resources of (a); b (l) ij ) Represents a physical link l ij Bandwidth resources of (a); d (l) uv ) Representing a virtual link l uv Delay of (2); d (l) ij ) Represents a physical link l ij The delay of (2).
2) Splitting virtual requests
Because the internet of things has a high requirement on real-time performance, the delay characteristic of a link needs to be considered in virtual network mapping. By splitting an incoming virtual network topology map, splitting each virtual request large topology into a group of virtual request small clusters, and marking each link to distinguish the delay characteristics (flag =1 represents a high delay requirement, and flag =0 represents a low delay requirement), as shown in (a), (b) and (c) in fig. 1, splitting a virtual request of a network model into a request with delay constraint and a request with only bandwidth constraint according to service types of different services in an internet of things environment (i.e. splitting the virtual network topology map into two small topology maps of S-V1 and S-V2). The S-V1 comprises four nodes and three links, and the topological graph has higher requirement on delay; the S-V2 comprises three nodes and two links, belongs to a topological graph with lower delay requirement, wherein the nodes with black marks are the nodes mapped by the S-V1, and the node mapping problem is not considered.
Virtual network mapping is generally divided into two parts, node mapping and link mapping.
3) Node mapping
Referring to fig. 2, the specific steps are as follows:
a. virtual node n of a computational network model v Resource capability NR (n) v ) And mixing NR (n) v ) And (5) arranging in descending order.
Wherein:
representing the topological relationship of the mapping node to surrounding nodes.
b. Aiming at the virtual node n arranged in the step a v Selecting physical nodes meeting the CPU resource constraint and the position constraint of the virtual nodes to obtain a candidate mapping node set omega (n) v )；
Ω(n v )＝{n s |cpu(n v )≤cpu(n s )&dis(loc(n v ),loc(n s ))≤D(n s ),n s ∈N s }
c. For a set of candidate mapping nodes Ω (n) v ) Each candidate node n in s According to their resource capabilities NR (n) s ) And energy consumption P of the node N (n s ) Calculating the comprehensive resource capacity of the nodes and arranging in descending order:
CR(n s )＝log(NR(n s )·P N (n s )+γ)
where γ → 0, avoids the case where 0 occurs inside log in the above formula.
d. Judging the type of the nodes arranged in the step c, if gene (n) v )＝genre(n s ) Will virtualize node n v Mapping to physical node n where CR is minimal and satisfies constraints s The above. Wherein, gene (n) v ) A node type representing a virtual node; gene (n) s ) Representing the node type of the physical node. If gene (n) v )≠genre(n s ) Returning to the step b, in the candidate mapping node set omega (n) v ) And continuously and circularly searching other physical nodes until the nodes with the same node type are found, and finishing node mapping.
e. Calculating the residual CPU resource CPU (n) of the successfully mapped physical node s )＝cpu(n s )-cpu(n v )。
f. And returning to the step a, sequentially finishing the mapping of each node in the virtual topological graph, and storing the physical nodes which are successfully mapped into the node mapping linked list NodeMappingList.
4) Link mapping
For the virtual link mapping, a K shortest path algorithm is adopted, which is common. However, in the environment of the internet of things, the information of the object needs to be accurately transmitted in real time through the sensor, so that the requirements on the energy consumption of network resources and the real-time performance of the link are high, and therefore, in the link mapping stage, the K shortest path algorithm which has priority on energy consumption and meets the link delay is comprehensively designed by integrating factors such as energy consumption and link delay in the link mapping process. Referring to fig. 3, the specific steps are as follows:
a. obtaining the bottom layer physical node n according to the node mapping list NodeMappisList v (u)→n s (i)，n v (v)→n s (j) Wherein n is v (u)→n s (i) Representing the mapping of a virtual node u to a physical node i, n v (v)→n s (j) Representing the mapping of virtual node v to physical node j.
b. Virtual link l uv And arranging according to the bandwidth requirement in a descending order.
c. For reduction in step bEach virtual link l after the sequence uv Calculating physical node n by adopting k shortest path algorithm s (i) To n s (j) Set of shortest paths of
And for arbitrary paths
d. For shortest path set
Calculates the link energy consumption of each path
And arranged in ascending order.
e. Judging the service type of the link by combining the virtual request splitting processing, only considering the bandwidth requirement of the link if the service type is the low delay requirement, and if b (l) uv )≤b(l ij ) Virtual link l uv Mapping to link energy consumption
Minimum physical link l ij The above step (1); if the service type is high latency requirement, then not only the bandwidth requirement of the link is considered, but also the latency requirement of the link, if b (l) uv )≤b(l ij )&d(l uv )≥d(l ij ) Virtual link l uv Mapping to link energy consumption
Minimum physical link l ij And finishing the link mapping.
f. Calculating the residual bandwidth resource b (l) of the physical link with successful mapping s )＝b(l s )-b(l v )。
g. And returning to the step b, sequentially finishing the mapping of each link in the virtual topological graph, and storing the successfully mapped physical link into a link mapping linked list LinkMappinList.
According to the method, aiming at the network characteristics of the Internet of things, the traditional energy consumption model is redefined in the virtual network mapping stage, so that the method is suitable for new energy consumption characteristics in the environment of the Internet of things. Under the condition of ensuring the minimum energy consumption, the heterogeneous performance of the nodes and the timeliness of the links are considered emphatically, firstly, the virtual requests are split, the timeliness characteristics of the links are distinguished, and then the two-stage method is adopted to complete the mapping of the virtual network resources. The algorithm is different from the existing energy consumption perception virtual network mapping algorithm, the virtual network mapping problem is completed in a resource integration mode, the maximization of the resource utilization rate is realized, the mapping time is shortened, the energy-saving effect is achieved, and the fine-grained mapping mode is realized along with the increase of introduced parameters.
Claims (8)
1. An energy consumption perception virtual network mapping algorithm based on the Internet of things is characterized by comprising the following steps:
1) Building network model and energy consumption perception model
a. Defining an underlying physical network topology graph as a weighted undirected graph
Wherein, N s And L s Respectively representing the set of underlying physical nodes and links,
and
respectively representing the attributes contained in the bottom layer physical node and the link;
defining a virtual network topology graph as a weighted undirected graph
Wherein N is v And L v Respectively representing a set of virtual nodes and links,
and
respectively representing the attributes contained in the virtual nodes and the links;
weighted undirected graph G s And weighted undirected graph G v Forming a network model;
b. the energy consumption perception model is established as follows:
an objective function: min P N +P L
CPU resource constraint:
CPU position constraint:
CPU category constraint:
bandwidth resource constraints of the link:
delay constraints of the link:
and (3) connectivity constraint:
wherein,
the method belongs to a binary variable and represents the mapping relation between a virtual node u and a physical node i; cpu (u) represents a cpu value of the virtual node; cpu (i) represents a cpu value of a physical node; dis (loc (u), loc (M) n (u))) represents the euclidean distance between the mapping node u and the surrounding nodes; gene (u) represents the node type of the virtual node; gene (i) represents the node type of the physical node;
representing a virtual link l uv With physical links l ij A mapping relationship between, b (l) uv ) Representing a virtual link l uv Bandwidth resources of (a); b (l) ij ) Represents a physical link l ij Bandwidth resources of (a); d (l) uv ) Representing a virtual link l uv Delay of (2); d (l) ij ) Represents a physical link l ij Delay of (P) N Mapping the total energy consumption for the node; p is L Mapping the total energy consumption for the link; total energy consumption P for link mapping L The following were used:
wherein, P n For the link power consumption to be a constant, d (i, j) represents the path distance from node i to node j, α represents the distance factor, N n Indicating the number of forwarding nodes that need to be changed from an inactive state to an active state,
representing a virtual link l uv With physical links l ik The mapping relation between S k Indicating the number of host nodes that need to be changed from inactive to active, N w Indicating the number of host nodes which need to be changed from the inactive state to the active state;
2) Finishing the mapping of each node in the virtual topological graph, and storing the successfully mapped physical nodes into a node mapping linked list; the specific process is as follows:
a. virtual node n of a computational network model v Resource capability NR (n) v ) And NR (n) v ) Arranging in descending order;
wherein:
representing the topological relation between a mapping node and surrounding nodes, wherein n represents a virtual node needing mapping, and nb represents all nodes connected with the mapping node; n is nb Denotes the nth connected node, L: ( n ) Representing all links between virtual nodes;
b. aiming at the virtual node n arranged in the step a v Selecting the virtual nodes meeting the CPU resource constraint and the position constraint of the virtual nodes to obtain a candidate mapping node set omega (n) v )；
Ω(n v )＝{n s |cpu(n v )≤cpu(n s )&dis(loc(n v ),loc(n s ))≤D(n s ),n s ∈N s }
c. For a set of candidate mapping nodes Ω (n) v ) Each candidate node n in s According to their resource capabilities NR (n) s ) And energy consumption P of the node N (n s ) Calculating the comprehensive resource capacity of the nodes and arranging in descending order:
CR(n s )＝log(NR(n s )·P N (n s )+γ)
wherein γ → 0, avoids the case where 0 occurs inside log in the above formula;
d. judging the type of the nodes arranged in the step c, if true (n) v )＝genre(n s ) Will virtual node n v Mapping to physical node n where CR is minimal and satisfies constraints s C, removing; wherein, gene (n) v ) A node type representing a virtual node; gente (n) s ) Representing node types of physical nodes(ii) a If gene (n) v )≠genre(n s ) Returning to the step b, in the candidate mapping node set omega (n) v ) Continuously and circularly searching other physical nodes until the nodes with the same node type are found, and finishing node mapping;
e. computing the remaining CPU resource CPU (n) of the successfully mapped physical node s )＝cpu(n s )-cpu(n v )；
f. Returning to the step a, sequentially finishing the mapping of each node in the virtual topological graph, and storing the successfully mapped physical nodes into a node mapping linked list;
3) And finishing the mapping of each link in the virtual topological graph according to the node mapping linked list, and storing the successfully mapped physical link into the link mapping linked list.
2. The energy consumption perception virtual network mapping algorithm based on the internet of things as claimed in claim 1, wherein in step 1), when
When the mapping is successful, the virtual node u is mapped to the physical node i,
if so, the operation is unsuccessful;
when the temperature is higher than the set temperature
Then, the virtual link l is illustrated uv Successful mapping to physical link/ ij In the above-mentioned manner,
if so, it is not successful.
3. The energy consumption perception virtual network mapping algorithm based on the Internet of things as claimed in claim 1, wherein the nodes in step 1) map total energy consumption P N ：
Wherein N is w Indicating the number of host nodes, P, that need to be changed from inactive to active b Representing the base power consumption, P, of a physical node l Representing the energy consumption of the forwarding node, util (n) u ) Represents the cpu utilization, S, required by the mapping of virtual node u to physical node i i Representing the energy state of physical node i when unmapped.
4. The Internet of things-based energy consumption perception virtual network mapping algorithm according to claim 3, wherein S is i =1 represents the state of the physical node i when activated, S i And =0 represents the other state.
5. The Internet of things-based energy consumption perception virtual network mapping algorithm according to claim 1, wherein when the energy consumption perception virtual network mapping algorithm is used, the energy consumption perception virtual network mapping algorithm is used
Then, the virtual link l is illustrated uv Successful mapping to physical link/ ik In the above-mentioned manner,
if so, it is unsuccessful.
6. The energy consumption perception virtual network mapping algorithm based on the internet of things as claimed in claim 1, wherein the specific process of step 3) is as follows:
a. obtaining a bottom layer physical node n according to a node mapping linked list v (u)→n s (i),n v (v)→n s (j) Wherein n is v (u)→n s (i) Representing the mapping of a virtual node u to a physical node i, n v (v)→n s (j) Representing the mapping of virtual node v to physical node j;
b. virtual link l uv In descending order of bandwidth demandArranging;
c. aiming at each virtual link l arranged in descending order in step b uv Calculating a physical node n s (i) To n s (j) Set of shortest paths of
And for arbitrary paths
d. For shortest path set
Calculates the link energy consumption of each path in the path
And arranging the same in an ascending order;
e. if the service type is low latency requirement, only the bandwidth requirement of the link is considered, if b (l) uv )≤b(l ij ) Virtual link l uv Mapping to link energy consumption
Minimum physical link l ij The above step (1); if the service type is high latency requirement, then not only the bandwidth requirement of the link is considered, but also the latency requirement of the link, if b (l) uv )≤b(l ij )&d(l uv )≥d(l ij ) Virtual link l uv Mapping to link energy consumption
Minimum physical link l ij Completing link mapping;
f. computingMapping the residual bandwidth resource b (l) of the successful physical link s )＝b(l s )-b(l v )；
g. And returning to the step b, sequentially finishing the mapping of each link in the virtual topological graph, and storing the successfully mapped physical links into a link mapping linked list.
7. The Internet of things-based energy consumption perception virtual network mapping algorithm according to claim 6, wherein flag =1 represents a high latency requirement; flag =0 indicates a low latency requirement.
8. The energy consumption perception virtual network mapping algorithm based on the Internet of things of claim 6, wherein in the step c, a k shortest path algorithm is adopted to calculate the physical node n s (i) To n s (j) Set of shortest paths of
And for arbitrary paths
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910339562.1A CN110061881B (en) | 2019-04-25 | 2019-04-25 | Energy consumption perception virtual network mapping algorithm based on Internet of things |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910339562.1A CN110061881B (en) | 2019-04-25 | 2019-04-25 | Energy consumption perception virtual network mapping algorithm based on Internet of things |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN110061881A CN110061881A (en) | 2019-07-26 |
| CN110061881B true CN110061881B (en) | 2022-10-28 |
Family
ID=67320779
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910339562.1A Expired - Fee Related CN110061881B (en) | 2019-04-25 | 2019-04-25 | Energy consumption perception virtual network mapping algorithm based on Internet of things |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110061881B (en) |
Families Citing this family (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111130858B (en) * | 2019-12-09 | 2023-05-19 | 网络通信与安全紫金山实验室 | Dynamic multi-target virtual network mapping method in SD-WAN scene |
| CN111371681B (en) * | 2020-03-12 | 2021-11-19 | 郑州轻工业大学 | Resource and energy consumption perception network service function chain mapping method |
| CN112436991A (en) * | 2020-11-09 | 2021-03-02 | 辽宁工程技术大学 | Virtual network mapping method based on energy consumption perception of enterprise network |
| CN112653580B (en) * | 2020-12-16 | 2022-11-08 | 国网河南省电力公司信息通信公司 | Virtual network resource allocation method based on active detection under network slice |
| CN114697222B (en) * | 2022-03-22 | 2023-06-20 | 西安交通大学 | Survivability virtual network mapping algorithm based on space-time-ground fusion network |
-
2019
- 2019-04-25 CN CN201910339562.1A patent/CN110061881B/en not_active Expired - Fee Related
Non-Patent Citations (1)
| Title |
|---|
| 云数据中心高效的虚拟机整合方法;喻新荣等;《计算机应用》;20180210(第02期);全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN110061881A (en) | 2019-07-26 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110061881B (en) | Energy consumption perception virtual network mapping algorithm based on Internet of things | |
| Zhao et al. | A modified cluster-head selection algorithm in wireless sensor networks based on LEACH | |
| Naranjo et al. | FOCAN: A Fog-supported smart city network architecture for management of applications in the Internet of Everything environments | |
| WO2021237996A1 (en) | Fuzzy c-means-based adaptive energy consumption optimization vehicle clustering method | |
| Huang et al. | A services routing based caching scheme for cloud assisted CRNs | |
| William et al. | Analysis of data aggregation and clustering protocol in wireless sensor networks using machine learning | |
| Yang et al. | Social-energy-aware user clustering for content sharing based on D2D multicast communications | |
| Li et al. | Resource scheduling based on improved spectral clustering algorithm in edge computing | |
| Khodaparas et al. | A software-defined caching scheme for the Internet of Things | |
| Zhao et al. | An adaptive vehicle clustering algorithm based on power minimization in vehicular ad-hoc networks | |
| Liu et al. | QEGWO: Energy-efficient clustering approach for industrial wireless sensor networks using quantum-related bioinspired optimization | |
| Wala et al. | Energy efficient data collection in smart cities using iot | |
| Ye et al. | Processing capability and QoE driven optimized computation offloading scheme in vehicular fog based F-RAN | |
| Zhang et al. | An Efficient Caching and Offloading Resource Allocation Strategy in Vehicular Social Networks | |
| Sharma et al. | PRESEP: Cluster based metaheuristic algorithm for energy-efficient wireless sensor network application in internet of things | |
| Su et al. | Boost decentralized federated learning in vehicular networks by diversifying data sources | |
| Ren et al. | Increasing network throughput based on dynamic caching policy at wireless access points | |
| Wu et al. | Collaborative caching relay algorithm based on recursive deep reinforcement learning in mobile vehicle edge network | |
| Yin et al. | Node importance evaluation method based on multi-attribute decision-making model in wireless sensor networks | |
| Gupta et al. | Energy Efficient Dynamic Sink Multi Level Heterogeneous Extended Distributed Clustering Routing for Scalable WSN: ML-HEDEEC | |
| Jain et al. | Performance analysis of hierarchical agglomerative clustering in a wireless sensor network using quantitative data | |
| CN112436991A (en) | Virtual network mapping method based on energy consumption perception of enterprise network | |
| ZHAO et al. | Collaborative offloading strategy in internet of vehicles based on asynchronous deep reinforcement learning | |
| Yu | Application of mobile edge computing technology in civil aviation express marketing | |
| CN109831337B (en) | Grouping mapping method for adjacent nodes in wireless network virtualization |
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 | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20221028 |
|
| CF01 | Termination of patent right due to non-payment of annual fee |