CN110166304A - A kind of method of combination, device, electronic equipment and the storage medium of cross-domain SFC - Google Patents
A kind of method of combination, device, electronic equipment and the storage medium of cross-domain SFC Download PDFInfo
- Publication number
- CN110166304A CN110166304A CN201910472411.3A CN201910472411A CN110166304A CN 110166304 A CN110166304 A CN 110166304A CN 201910472411 A CN201910472411 A CN 201910472411A CN 110166304 A CN110166304 A CN 110166304A
- Authority
- CN
- China
- Prior art keywords
- sfc
- node
- probability
- network function
- layout
- 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.)
- Pending
Links
Classifications
-
- 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
- H04L41/0893—Assignment of logical groups to network elements
-
- 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
- H04L41/14—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
Abstract
The embodiment of the invention provides the method for combination of cross-domain SFC a kind of, device, electronic equipment and storage medium, the described method includes: for a plurality of to layout SFC and multiple data center's nodes, generated VNF expense and bandwidth cost when being deployed in different data central node based on heterogeneous networks functional node, construct probability set, and construct transition probability matrix and output probability matrix, then, based on constructed probability set, transition probability matrix, and output probability matrix, construct hidden Markov model, utilize the probability in hidden Markov model, transition probability and output probability, calculate the hidden state probability that network function node deployment in SFC is corresponding in data center's node, and layout is carried out to each SFC in a plurality of SFC based on hidden state probability, obtain each S The corresponding hidden state subsequence of FC.The embodiment of the present invention can reduce the cross-domain bandwidth cost of SFC after layout.
Description
Technical field
The present invention relates to fields of communication technology, more particularly to method of combination, device, the electronic equipment of a kind of cross-domain SFC
And storage medium.
Background technique
With the fast development of the communication technology, network size go from strength to strength and the continuous extension of business, in order to complete
It is serviced accordingly in network, flow needs are handled in order by the network function for meeting particular order in network, SFC (Service
Function Chains, business function chain) define the sequence that network function is handled.In practical application, network operator needs
Flow is directed to proprietary hardware net function to complete the corresponding service of SFC, but the network function of traditional task equipment is in net
Network flexibility, scalability can all show apparent limitation in terms of management type and efficiency of operation, so that NFV
(Network Functions Virtualization, network function virtualization) comes into being.NFV is a kind of new network rack
Structure decouples network function from proprietary hardware device, with VNF (Virtualized Network Function, virtual network
Function) form operate on common apparatus.
With the extensive use of cloud computing and big data, it is distributed in more IDCN (Inter-DC of diverse geographic location
Network, data center network) it is widely deployed.By disposing the SFC based on NFV in more IDCN, so that user can be with
Computing resource, Internet resources and the storage resource etc. of data center are neatly used, and realizes economy efficiently customer service portion
Administration.Although disposing SFC tool in more IDCN to have great advantage, across data center layout SFC is that one of current SFC layout chooses
War.
It is directed to the method for SFC layout at present are as follows: volume is treated based on the correlation of VNF type in SFC for all SFC
The a plurality of SFC of row carries out the traversal integration of network function type one by one, until needed layout SFC is integrated into one or more
In service function figure, which includes the network function sequence of needed layout SFC request, then, selects scale most
Big service function figure carries out topological sorting, the network function for further selecting bandwidth demand amount minimum to the service function figure
Sequentially, it will be successively deployed in bottom-layer network to the network function of layout SFC using shortest path first.
However, SFC is usually required to be deployed in the data center of diverse geographic location distribution to meet in practical application
The performance requirement or position constraint of SFC, for example, " agent functionality " and " caching function " should be deployed in the number near enterprise network
According on center, " packet filtering " should be deployed in the data center in flow source etc..It is existing to be directed to SFC layout
Method, be typically based on the correlation of VNF type in SFC, will be integrated into layout SFC in one or more service function figure,
To realize the layout to SFC.But when the SFC quantity of layout is more, in the service function figure after integration, it may go out
Now in layout SFC front and continued functional node and the non-conterminous situation of follow-up function node, that is, front and continued functional node with it is subsequent
Functional node is likely to occur cross-domain situation, and the cross-domain bandwidth cost between front and continued functional node and follow-up function node is caused to increase
Greatly, the cross-domain bandwidth cost of SFC after layout is caused to increase.
Summary of the invention
The method of combination for being designed to provide a kind of cross-domain SFC, device, electronic equipment and the storage of the embodiment of the present invention are situated between
Matter, to reduce the cross-domain bandwidth cost of SFC after layout.Specific technical solution is as follows:
In a first aspect, the embodiment of the invention provides the method for combination of cross-domain SFC a kind of, which comprises
The a plurality of business function chain SFC to layout and multiple data center's nodes are obtained, includes in every SFC
Multiple network function nodes, data center's node are used for the network function node disposed in the SFC, a network
Functional node is deployed in data center's node;
Generated virtual network function VNF is opened when being deployed in different data central node based on heterogeneous networks functional node
Pin and bandwidth cost determine heterogeneous networks functional node being deployed in the probability in different data central node, and be based on
Identified multiple probabilities construct probability set;
It is transferred to transition probability determined by the second state based on first state, constructs transition probability matrix;Described first
State is state corresponding to process that first network functional node in SFC is deployed in first data center's node, described
Second state be by state corresponding to process of the second network function node deployment in second data center's node in SFC,
The second network function node is the adjacent node of the first network functional node;
Based on the output probability for exporting the corresponding function type of third network function node in SFC under the third state, building
Output probability matrix;The third state is by third network function node deployment in the SFC in third data center node
In process corresponding to state;
Based on the constructed probability set, the transition probability matrix and the output probability matrix, structure
Build hidden Markov model;
Using the probability, the transition probability and the output probability in the hidden Markov model,
The hidden state probability that network function node deployment in the SFC is corresponding in data center's node is calculated, and based on described
Hidden state probability carries out layout to each SFC in a plurality of SFC, obtains the corresponding hidden state subsequence of each SFC, obtains
To the corresponding multiple hidden state subsequences of a plurality of SFC;The element for including in the hidden state subsequence are as follows: in the SFC
Data center's node that network function node is disposed.
Optionally, it using the first default expression formula, determines and the heterogeneous networks functional node is deployed in the different numbers
According to the probability in central node;
Using the second default expression formula, determine that the first state is transferred to the transition probability of second state;
Expression formula is preset using third, determines and exports third network function node pair described in SFC under the third state
The output probability for the function type answered;
The first default expression formula are as follows:
In formula, πmIndicate that by the network function node deployment, in the probability of than the m-th data central node, M is indicated
The number of data center's node,Indicate s-th of VNF example in than the m-th data central node,It indicates first
Expense caused by s-th of VNF example of the network function node deployment in than the m-th data central node,It indicates to rise
Beginning network function node deployment is in initial data central node, and first network function node deployment is in than the m-th data centromere
Bandwidth cost is shifted caused by point, 01 indicates the initial network functional node of SFC to the state of first network function node
Transfer, σ m are indicated from initial data central node to the transfer of than the m-th data central node;
The second default expression formula are as follows:
In formula, Expression stateIt is transferred to stateTransfer
Probability, stateIt indicates corresponding to the process by (i-1)-th network function node deployment in nth data central node
State, stateIt indicates corresponding to the process by i-th of network function node deployment in than the m-th data central node
State,It indicates (i-1) a network function node deployment in data center node n, i-th of network function node
It is deployed on the transfer bandwidth cost of data center m, (i-1) i indicates (i-1) a network function node to i-th of network function section
The transfer of point, nm indicate from nth data central node to the transfer of than the m-th data central node,Indicate data center
The unit bandwidth expense of node n to data center node m, N indicate the network topology quantity of data center's node,It indicates
Bandwidth on demand amount between (i-1) a network function node of pth article SFC and i-th of network function node;
The third presets expression formula are as follows:
In formula,Expression stateLower output network function typeProbability, stateIt indicates i-th
State corresponding to process of a network function node deployment in than the m-th data central node,Indicate the i-th of pth SFC
The function type of a network function node,Expression stateUnder do not export network function typeProbability,It indicates to open caused by s-th of VNF example by i-th of network function node deployment in than the m-th data central node
Pin,Indicate the reliability of s-th of VNF example in than the m-th data central node,Indicate than the m-th data centromere
The function type of s-th of VNF example in point.
Optionally, a plurality of SFC to layout is SFC set;The institute using in the hidden Markov model
Probability, the transition probability and the output probability are stated, is calculated network function node deployment in the SFC in data
Corresponding hidden state probability in central node, and each SFC in a plurality of SFC is carried out based on the hidden state probability
Layout obtains the corresponding hidden state subsequence of each SFC, obtains the corresponding multiple hidden state subsequences of a plurality of SFC are as follows:
Using the probability, the transition probability and the output probability in the hidden Markov model,
The hidden state probability that network function node deployment in the SFC is corresponding in data center's node is calculated, and based on described
Hidden state probability carries out layout to each SFC in a plurality of SFC, obtains the corresponding hidden state subsequence of each SFC, obtains
Gather corresponding hidden status switch set to the SFC.
Optionally, the probability using in the hidden Markov model, the transition probability and institute
Output probability is stated, the hidden state probability that network function node deployment in the SFC is corresponding in data center's node is calculated,
And layout is carried out to each SFC in a plurality of SFC based on the hidden state probability, obtain the corresponding hidden shape of each SFC
State subsequence obtains the step of SFC gathers corresponding hidden status switch set, comprising:
Judge in the SFC set with the presence or absence of to layout SFC;
If existed in the SFC set to layout SFC, longest SFC in the SFC set is selected to compile as current
Arrange SFC;
Using the probability, the transition probability and the output probability in the hidden Markov model,
It is general to calculate the hidden state that each network function node deployment in the current layout SFC is corresponding in data center's node
Rate, and layout is carried out to the current layout SFC based on the hidden state probability, it is corresponding hidden to obtain the current layout SFC
State subsequence;
If exporting the SFC there is no to layout SFC in the SFC set and gathering corresponding hidden set of state sequence
It closes.
Optionally, the probability using in the hidden Markov model, the transition probability and institute
Output probability is stated, is calculated each network function node deployment in the current layout SFC is corresponding in data center's node
Hidden state probability, and layout is carried out to the current layout SFC based on the hidden state probability, obtains the current layout
The step of SFC corresponding hidden state subsequence, comprising:
Judge the current layout SFC whether complete by layout;
If the non-layout of current layout SFC is completed, the current layout network function of the current layout SFC is judged
Node whether first network function node for being the current layout SFC;
If the current layout network function node of the current layout SFC is first net of the current layout SFC
Network functional node is calculated then based on the probability in the hidden Markov model by the current layout network function
Initial hidden state probability of the node deployment in each data center's node;
If the current layout network function node of the current layout SFC is not first of the current layout SFC
Network function node, then using the transition probability and the output probability in the hidden Markov model, calculating will
Maximum hidden state probability of the current layout network function node deployment in each data center's node, and record described in acquisition
The position of the corresponding advance data central node of the hidden state probability of maximum;
By next network function node of the current layout network function node of the current layout SFC, work as described
The current layout network function node of preceding layout SFC, execution judge the current layout SFC whether layout completion the step of;
If the current layout SFC layout is completed, using the end network function node of the current layout SFC as
Current network functional node, selection make maximum corresponding 4th data center of the hidden state probability of the current network functional node
Current network functional node described in node deployment, and the 4th data center's node is stored in described SFC pairs of current layout
In the hidden state subsequence answered;
The maximum corresponding advance data central node of hidden state probability that will make the current network functional node, is determined as
5th data center's node corresponding to the previous network function node of the current network functional node, in the 5th data
Dispose the previous network function node of the current network functional node in central node, and by the 5th data center's node
It is stored in the corresponding hidden state subsequence of the current layout SFC, by the previous network function of the current network functional node
Energy node is as current network functional node;
Whether the previous network function node for judging the current network functional node is the first of the current layout SFC
A network function node;
If the previous network function node of the current network functional node is not first of the current layout SFC
Network function node then executes the maximum corresponding advance data center of hidden state probability that will make the current network functional node
Node is determined as the step of the 5th data center's node corresponding to the previous network function node of the current network functional node
Suddenly;
If the previous network function node of the current network functional node is first net of the current layout SFC
Network functional node then deletes the current layout SFC from SFC set.
Optionally, the method also includes:
Based on each corresponding reliability value of SFC and the hidden state sequence after the hidden status switch set, layout
In column set in the corresponding SFC of each hidden state subsequence VNF corresponding to each network function node cost-effectiveness value, to institute
VNF is stated to be backed up.
Optionally, it is described based on the hidden status switch set, the corresponding reliability value of each SFC and institute after layout
State the cost of VNF corresponding to each network function node in the corresponding SFC of each hidden state subsequence in hidden status switch set
Benefit value, the step of backup to the VNF, comprising:
Each hidden state subsequence in the hidden status switch set is traversed, by SFC corresponding to the hidden state subsequence
As current SFC;
Calculate the reliability value of the current SFC;
Judge whether the reliability value of the current SFC is less than default reliability value;
When the reliability value of the current SFC is less than default reliability value, the current SFC is placed in first set
In, and will be placed in second set by the VNF of the current SFC;
Judge whether the first set is empty;
When the first set is not sky, the cost-effectiveness value of every VNF in the second set is calculated;
VNF corresponding to maximum cost-effectiveness value is backed up;
Calculate the reliability value of every SFC in the first set after backing up;
If the reliability value of SFC is not less than default reliability value in the first set after backup, by the SFC from institute
It states in first set and deletes, and execute and described judge whether the first set is empty step.
Second aspect, the embodiment of the invention provides the layout device of cross-domain SFC a kind of, described device includes:
Module is obtained, for obtaining a plurality of business function chain SFC to layout and multiple data center's nodes, every institute
Stating includes multiple network function nodes in SFC, and data center's node is used for the network function section disposed in the SFC
Point, a network function node deployment is in data center's node;
First building module, it is generated when for being deployed in different data central node based on heterogeneous networks functional node
Heterogeneous networks functional node is deployed in different data central node by virtual network function VNF expense and bandwidth cost, determination
Probability, and construct probability set based on identified multiple probabilities;
Second building module, for being transferred to transition probability determined by the second state, building transfer based on first state
Probability matrix;The first state is the process being deployed in first network functional node in SFC in first data center's node
Corresponding state, second state be by the second network function node deployment in SFC in second data center's node
State corresponding to process, the second network function node are the adjacent node of the first network functional node;
Third constructs module, and the corresponding function class of third network function node in SFC is exported under the third state for being based on
The output probability of type constructs output probability matrix;The third state be by third network function node deployment in the SFC in
State corresponding to process in third data center node;
4th building module, for based on the constructed probability set, the transition probability matrix, Yi Jisuo
Output probability matrix is stated, hidden Markov model is constructed;
Orchestration module, for using in the hidden Markov model the probability, the transition probability and
It is general to calculate the hidden state that network function node deployment in the SFC is corresponding in data center's node for the output probability
Rate, and layout is carried out to each SFC in a plurality of SFC based on the hidden state probability, it is corresponding hidden to obtain each SFC
State subsequence obtains the corresponding multiple hidden state subsequences of a plurality of SFC;The element for including in the hidden state subsequence
Are as follows: data center's node that network function node is disposed in the SFC.
The third aspect, the embodiment of the invention also provides a kind of electronic equipment, including processor, communication interface, memory
And communication bus, wherein processor, communication interface, memory complete mutual communication by communication bus;
Memory, for storing computer program;
Processor when for executing the program stored on memory, realizes that one kind described in above-mentioned first aspect is cross-domain
The method of combination of SFC.
Fourth aspect, it is described computer-readable to deposit the embodiment of the invention also provides a kind of computer readable storage medium
Instruction is stored in storage media, when run on a computer, so that computer executes one kind described in above-mentioned first aspect
The method of combination of cross-domain SFC.
Method of combination, device, electronic equipment and the storage medium of a kind of cross-domain SFC provided in an embodiment of the present invention, because of HMM
Hidden state cannot observe directly, but can be arrived by observable sequence inspection, each observable sequence is by general
Rate Density Distribution shows as various states, each observable sequence is by a state with corresponding probability density distribution
Sequence generates, and in the embodiment of the present invention, the layout of cross-domain SFC is modeled as hidden Markov model, recycles hidden Markov
Model carries out layout to each SFC in a plurality of SFC, obtains the corresponding hidden state subsequence of each SFC, because being to be based on calculating
To the hidden state probability that network function node deployment in SFC is corresponding in data center's node, to the network function of SFC
Node is disposed, and realizes the layout of SFC, can reduce the cross-domain bandwidth cost of SFC after layout, and then after reduction SFC layout
Consumed bandwidth resources.
Certainly, implement any of the products of the present invention or method it is not absolutely required at the same reach all the above excellent
Point.
Detailed description of the invention
In order to more clearly explain the embodiment of the invention or the technical proposal in the existing technology, to embodiment or will show below
There is attached drawing needed in technical description to be briefly described, it should be apparent that, the accompanying drawings in the following description is only this
Some embodiments of invention for those of ordinary skill in the art without creative efforts, can be with
It obtains other drawings based on these drawings.
Fig. 1 is a kind of flow diagram of the method for combination of cross-domain SFC provided in an embodiment of the present invention;
Fig. 2 is the embodiment flow diagram that a kind of couple of SFC provided in an embodiment of the present invention carries out layout;
Fig. 3 is a kind of embodiment flow diagram of S203 in Fig. 2;
Fig. 4 is a kind of SFC structural schematic diagram provided in an embodiment of the present invention;
Fig. 5 is the embodiment flow diagram that a kind of couple of VNF provided in an embodiment of the present invention is backed up;
Fig. 6 a is VNF expense provided in an embodiment of the present invention and SFC demand magnitude relation analogous diagram;
Fig. 6 b is cross-domain bandwidth cost provided in an embodiment of the present invention and SFC demand magnitude relation analogous diagram;
Fig. 6 c is backup expense provided in an embodiment of the present invention and SFC demand magnitude relation analogous diagram;
Fig. 6 d is overhead provided in an embodiment of the present invention and SFC demand magnitude relation analogous diagram;
Fig. 7 is a kind of structural schematic diagram of the layout device of cross-domain SFC provided in an embodiment of the present invention;
Fig. 8 is the structural schematic diagram of a kind of electronic equipment provided in an embodiment of the present invention.
Specific embodiment
Following will be combined with the drawings in the embodiments of the present invention, and technical solution in the embodiment of the present invention carries out clear, complete
Site preparation description, it is clear that described embodiments are only a part of the embodiments of the present invention, instead of all the embodiments.It is based on
Embodiment in the present invention, it is obtained by those of ordinary skill in the art without making creative efforts every other
Embodiment shall fall within the protection scope of the present invention.
As shown in FIG. 1, FIG. 1 is a kind of flow diagrams of the method for combination of cross-domain SFC provided in an embodiment of the present invention, should
Method may include:
S101 obtains a plurality of business function chain SFC and multiple data center's nodes to layout.
It is available a plurality of in the SFC of layout and multiple data when carrying out layout to SFC in the embodiment of the present invention
Heart node.It wherein, include multiple network function nodes in every SFC, data center's node is used for the network function disposed in SFC
Energy node, for a network function node deployment in a data central node, a data central node indicates a data
Central site network.
S102, generated VNF expense and band when being deployed in different data central node based on heterogeneous networks functional node
Wide expense determines heterogeneous networks functional node being deployed in the probability in different data central node, and based on determining
Multiple probabilities construct probability set.
In practical application, by network function node deployment when data center's node, need to create VNF example, and then will
Network function node deployment can generate VNF expense in the VNF example created.Illustratively, VNF example can be
VFW (Virtualized Fire Wall, virtual firewall), (Virtualized LoadBalance, virtual flow are equal by vLB
Weighing apparatus) etc..When network between different data central node carries out data interaction, need to occupy corresponding bandwidth, so that different
Network function node deployment can generate corresponding bandwidth cost when different data central node.
Because hidden Markov model can the status switch according to known to calculate maximum probability hidden state sequence, and
In the layout process of SFC, the sequence of network function node is it is known that obtained network function node is needed to be disposed in SFC
Position sequence is unknown, thus in the embodiment of the present invention, the Arrangement of cross-domain SFC is modeled as a hidden Markov mould
Type.Generated VNF expense and bandwidth cost when being deployed in different data central node based on heterogeneous networks functional node, determine
Heterogeneous networks functional node is deployed in the probability in different data central node.
As a kind of optional embodiment of the embodiment of the present invention, the first default expression formula can use, determination will be different
Probability of the network function node deployment in different data central node, the first default expression formula can be with are as follows:
In formula, πmIndicate that by network function node deployment, in the probability of than the m-th data central node, M indicates data
The number of central node,Indicate s-th of VNF example in than the m-th data central node,It indicates first network
Functional node is deployed inGenerated expense,It indicates initial network functional node being deployed in initial data centromere
Point, first network function node deployment shift bandwidth cost caused by the than the m-th data central node, and 01 indicates SFC's
Initial network functional node is to the state transfer of first network function node, and σ m expression is from initial data central node to m
The transfer of a data central node.
It, can be with base after determining the probability that heterogeneous networks functional node is deployed in different data central node
Probability set is constructed in identified multiple probabilities.Illustratively, constructed probability set can indicate
Are as follows: Π={ π1,π2,…πM, wherein πMIt indicates network function node deployment in the initial general of m-th data center node
Rate.
S103 is transferred to transition probability determined by the second state based on first state, constructs transition probability matrix.
In the embodiment of the present invention, it can determine that first state is transferred to the transition probability of the second state, then based on determining
Transition probability, construct transition probability matrix.Wherein, first state is that first network functional node in SFC is deployed in first
State corresponding to process in data center's node, the second state are by the second network function node deployment in SFC in second
State corresponding to process in data center's node, the second network function node are the adjacent segments of first network functional node
Point.In the embodiment of the present invention, the sequence of network function node is observable sequence in SFC, by the network function node in SFC
Data center's node of administration cannot observe directly, but can be converted by the intermediate of observable sequence, show as various
State.Illustratively, the process that first network functional node in SFC is deployed in first data center's node can be converted to
A kind of state, the state are expressed as first state.
As a kind of optional embodiment of the embodiment of the present invention, it can use the second default expression formula, determine the first shape
State is transferred to the transition probability of the second state, which can be with are as follows:
In formula, Expression stateIt is transferred to stateTransfer
Probability, stateIt indicates corresponding to the process by (i-1)-th network function node deployment in nth data central node
State, stateIt indicates corresponding to the process by i-th of network function node deployment in than the m-th data central node
State,It indicates (i-1) a network function node deployment in data center node n, i-th of network function node
It is deployed on the transfer bandwidth cost of data center m, (i-1) i indicates (i-1) a network function node to i-th of network function section
The transfer of point, nm indicate from nth data central node to the transfer of than the m-th data central node,Indicate data center
The unit bandwidth expense of node n to data center node m, N indicate the network topology quantity of data center's node,It indicates
Bandwidth on demand amount between (i-1) a network function node of pth article SFC and i-th of network function node.
Illustratively, constructed transition probability matrix can indicate are as follows:
Wherein, A indicates constructed transition probability matrix,Expression stateIt is transferred to state
Transition probability, stateIndicate the process by (i-1)-th network function node deployment in the 1st data central node
Corresponding state, stateIndicate the process institute by i-th of network function node deployment in m-th data center node
Corresponding state.
S104, based under the third state export SFC in the corresponding function type of third network function node output probability,
Construct output probability matrix.
In the embodiment of the present invention, it can determine and export the corresponding function of third network function node in SFC under the third state
The output probability of type constructs output probability matrix further according to identified output probability.Wherein, the third state be will be in SFC
State corresponding to process of the third network function node deployment in third data center node.
As a kind of optional embodiment of the embodiment of the present invention, it can use third and preset expression formula, determine third shape
The output probability of the corresponding function type of third network function node in SFC is exported under state, which presets expression formula can be with are as follows:
In formula,Expression stateLower output network function typeProbability, stateIt indicates i-th
State corresponding to process of a network function node deployment in than the m-th data central node,Indicate the i-th of pth SFC
The function type of a network function node,Expression stateUnder do not export network function typeProbability,Indicate s-th of VNF example in than the m-th data central node,Indicate by i-th of network function node deployment in
Generated expense,It indicatesReliability value,It indicatesFunction type.
Illustratively, constructed output probability matrix can indicate are as follows:
Wherein, B indicates constructed output probability matrix,Expression stateLower output network function class
TypeProbability, stateIndicate that the process institute by i-th of network function node deployment in m-th data center node is right
The state answered,Indicate the corresponding function type of i-th of network function node of pth SFC.
S105 constructs hidden horse based on constructed probability set, transition probability matrix and output probability matrix
Er Kefu model.
As a kind of optional embodiment of the embodiment of the present invention, can based on constructed probability set Π, turn
Probability matrix A and output probability matrix B are moved, the trigram models of hidden Markov model are constructed, which can be with
It is expressed as (Π, A, B).
S106, using probability, transition probability and the output probability in hidden Markov model, calculating will be in SFC
Network function node deployment hidden state probability corresponding in data center's node, and based on hidden state probability to a plurality of SFC
Middle each SFC carries out layout, obtains the corresponding hidden state subsequence of each SFC, obtains the corresponding multiple hidden state subgroups of a plurality of SFC
Sequence.
In the embodiment of the present invention, the sequence of network function node is observable sequence in SFC, by the network function in SFC
Data center's node of node deployment cannot observe directly, but can be arrived by observable sequence inspection, each considerable
Examining sequence all is to show as various states by probability density distribution, each observable sequence is that had accordingly generally by one
The status switch of rate Density Distribution generates.In the embodiment of the present invention, calculate network function node deployment in SFC in data center
Corresponding hidden state probability in node, and then layout can be carried out to each SFC in a plurality of SFC based on hidden state probability.
Wherein, the element for including in obtained hidden state subsequence can be with are as follows: the data center that network function node is disposed in SFC
Node.
As a kind of optional embodiment of the embodiment of the present invention, a plurality of SFC to layout can be expressed as SFC set.
Above-mentioned steps S106 is specifically as follows:
Using probability, transition probability and the output probability in hidden Markov model, calculate network function in SFC
Energy node deployment hidden state probability corresponding in data center's node, and based on hidden state probability to each in a plurality of SFC
SFC carries out layout, obtains the corresponding hidden state subsequence of each SFC, obtains SFC and gathers corresponding hidden status switch set.
Illustratively, SFC set can indicate are as follows: SFC={ S1,S2,…Sq, SFC gathers corresponding hidden set of state sequence
Conjunction can indicate are as follows: Q={ Q1,Q2,…Qq, wherein SqIndicate the q articles SFC, QqIndicate the corresponding hidden state subgroup sequence of the q articles SFC
Column.
As a kind of optional embodiment of the embodiment of the present invention, using in hidden Markov model probability, turn
Probability and output probability are moved, the hidden state that network function node deployment in SFC is corresponding in data center's node is calculated
Probability, and layout is carried out to each SFC in a plurality of SFC based on hidden state probability, obtain the corresponding hidden state subgroup sequence of each SFC
Column, obtain SFC and gather the embodiment of corresponding hidden status switch set can be found in Fig. 2, which may include:
S201 judges in SFC set with the presence or absence of to layout SFC.
For the SFC set to layout, judge to whether there is in SFC set to layout SFC, if deposited in SFC set
To layout SFC, then illustrating that treating layout SFC carries out layout, or will start to treat layout SFC progress layout, is then holding
The step of row S202;If illustrating to complete to layout SFC layout, then executing there is no to layout SFC in SFC set
The step of S204.
S202, if existed in SFC set to layout SFC, select SFC gather in longest SFC as current layout
SFC。
As a kind of optional embodiment of the embodiment of the present invention, if existed in SFC set to layout SFC, for simplification
The complexity of SFC layout, select SFC gather in longest SFC as current layout SFC, which can be with are as follows: includes
The at most corresponding SFC of network function number of nodes.Illustratively, which can indicate are as follows:
Wherein,Indicate the k-th network function node of pth SFC.
S203, using probability, transition probability and the output probability in hidden Markov model, calculating will be compiled currently
The hidden state probability that each network function node deployment is corresponding in data center's node in SFC is arranged, and general based on hidden state
Rate carries out layout to current layout SFC, obtains the corresponding hidden state subsequence of current layout SFC.
After determining current layout SFC, layout is carried out to current layout SFC, specific implementation process is described in detail below.
S204, if exporting SFC there is no to layout SFC in SFC set and gathering corresponding hidden status switch set.
There is no to layout SFC in SFC set, then show to complete to layout SFC layout, exports SFC set at this time
Corresponding hidden status switch set.
As a kind of optional embodiment of the embodiment of the present invention, the specific embodiment of above-mentioned steps S203 can be found in figure
3, which may include:
S2031 judges current layout SFC whether complete by layout.
For current layout SFC, it can be determined that current layout SFC whether complete by layout, if it is, executing S2032's
Step;If it is not, then the step of executing S2036.
S2032 judges the current layout network function section of current layout SFC if the non-layout of current layout SFC is completed
Point whether first network function node for being current layout SFC.
When the non-layout of current layout SFC is completed, it can be determined that the current layout network function node of current layout SFC is
No first network function node for current layout SFC, if it is, the step of executing S2033;If it is not, then executing
The step of S2034.
S2033, if the current layout network function node of current layout SFC is first network of current layout SFC
Functional node, then based on the probability in hidden Markov model, calculate by the current layout network function node deployment in
Initial hidden state probability in each data center's node.
As a kind of optional embodiment of the embodiment of the present invention, the 4th default expression formula can use to be based on hidden Ma Er
Can probability in husband's model, calculate the current layout network function node deployment in each data center's node just
Begin hidden state probability, because the current layout network function node is first network function node of current layout SFC, therefore calculates
The corresponding initial hidden state probability of first network function node, be also possible to its hidden state probability of maximum.4th
Default expression formula can be with are as follows:
In formula,Indicate first network function node deployment of the current layout SFC in data center node m
Initial hidden state probability, πmIt indicates network function node deployment in the probability of than the m-th data central node,Expression stateLower output network function typeProbability, stateIt indicates first network function section
State corresponding to the process that point is deployed in than the m-th data central node,Indicate first network function of pth SFC
The function type of node.
S2034, if the current layout network function node of current layout SFC is not first net of current layout SFC
Network functional node is calculated then using the transition probability and output probability in hidden Markov model by the current layout network
Functional node is deployed in the hidden state probability of maximum in each data center's node, and it is corresponding to record the maximum hidden state probability of acquisition
Advance data central node position.
As a kind of optional embodiment of the embodiment of the present invention, in the current layout network function section of current layout SFC
When point is not first network function node of current layout SFC, then can use the 5th default expression formula will deserve to calculate
Maximum hidden state probability of the preceding layout network function node deployment in each data center's node, and record the maximum hidden shape of acquisition
The position of the corresponding advance data central node of state probability.5th default expression formula can be with are as follows:
In formula,It indicates i-th of network function node deployment of the current layout SFC in data center node m
In the hidden state probability of maximum,It indicates (i-1)-th network function node deployment of current layout SFC in data
The hidden state probability of maximum in central node x,Expression stateIt is transferred to stateTransition probability,
StateIndicate shape corresponding to the process by (i-1)-th network function node deployment in x-th of data center's node
State, stateIndicate state corresponding to the process by i-th of network function node deployment in than the m-th data central node,Expression stateLower output network function typeProbability, stateIt indicates i-th of network function section
State corresponding to the process that point is deployed in than the m-th data central node,Indicate i-th of network function section of pth SFC
The function type of point.
S2035, by next network function node of the current layout network function node of current layout SFC, as current
The current layout network function node of layout SFC, execution judge current layout SFC whether layout completion the step of.
Calculating, the hidden state of maximum by the current layout network function node deployment in each data center's node is general
Rate, and after recording the position for obtaining the corresponding advance data central node of maximum hidden state probability, by the current of current layout SFC
Next network function node of layout network function node, as the current layout network function node of current layout SFC, so
Afterwards, the step of executing S2031, until current layout SFC layout is completed.
S2036, if current layout SFC layout is completed, using the end network function node of current layout SFC as working as
Preceding network function node, selection make the maximum corresponding 4th data center's node of the hidden state probability of current network functional node
Current network functional node is affixed one's name to, and the 4th data center's node is stored in the corresponding hidden state subsequence of current layout SFC.
When current layout SFC layout is completed, by the end network function node of current layout SFC, i.e., current layout SFC
The last one network function node as current network functional node, selection makes the hidden state probability of current network functional node
Maximum corresponding 4th data center node deployment current network functional node, and the 4th data center's node is stored in currently
In the corresponding hidden state subsequence of layout SFC.Illustratively, which is pth SFC, the end of current layout SFC
Tail network function node isMake current network functional nodeThe maximum corresponding 4th data center section of hidden state probability
Point is dm, by dmIt is stored in the corresponding hidden state subsequence of current layout SFC.
S2037, the maximum corresponding advance data central node of hidden state probability that will make current network functional node, determines
5th data center's node corresponding to previous network function node for current network functional node is saved in the 5th data center
The previous network function node of current network functional node is disposed in point, and the 5th data center's node is stored in current layout
In the corresponding hidden state subsequence of SFC, using the previous network function node of current network functional node as current network function
Node.
Illustratively, make current network functional nodeThe maximum corresponding 4th data center's node of hidden state probability be
dm, make current network functional nodeThe maximum corresponding advance data central node of hidden state probability, i.e. current network function
The previous network function node of nodeThe 5th corresponding data center node dx, in dxMiddle deploymentBy dxIt is stored in
In the corresponding hidden state subsequence of currently layout SFC, and willAs current network functional node.
S2038 judges whether the previous network function node of current network functional node is first of current layout SFC
Network function node.
Judge current network functional node previous network function node whether first network function for being current layout SFC
Energy node if so, illustrating network function node all deployment completions of current layout SFC, and obtains the current layout SFC
Corresponding hidden state subsequence, then the step of executing S2039, if it is not, illustrating the network function node of current layout SFC simultaneously
It is completed without all deployment, then returns to the step of executing S2037.Illustratively, current layout SFC is Sp, obtained preceding layout
The corresponding hidden state subsequence of SFC can indicate are as follows:Wherein,Layout S before indicatingpIn K
The data center node d that a network function node is disposedK。
S2039, if the previous network function node of current network functional node is first network of current layout SFC
Functional node then deletes current layout SFC from SFC set.
If the previous network function node of current network functional node is first network function section of current layout SFC
Point illustrates that all deployment is completed for the network function node of current layout SFC, and obtains the corresponding hidden shape of the current layout SFC
Current layout SFC is then deleted from SFC set, and continued to next carry out layout to layout SFC by state subsequence.
The method of combination of a kind of cross-domain SFC provided in an embodiment of the present invention, because the hidden state of HMM cannot be observed directly
It arrives, but can be arrived by observable sequence inspection, each observable sequence is to show as various shapes by probability density distribution
State, each observable sequence is generated by a status switch with corresponding probability density distribution, the embodiment of the present invention
In, the layout of cross-domain SFC is modeled as hidden Markov model, hidden Markov model is recycled, to each in a plurality of SFC
SFC carries out layout, obtains the corresponding hidden state subsequence of each SFC, because being based on being calculated network function node in SFC
It is deployed in hidden state probability corresponding in data center's node, the network function node of SFC is disposed, realizes SFC's
Layout can reduce the cross-domain bandwidth cost of SFC after layout, and then reduce consumed bandwidth resources after SFC layout.
As a kind of optional embodiment of the embodiment of the present invention, after above-mentioned steps S106, the embodiment of the present invention
The layout of cross-domain SFC can also include: based on the corresponding reliability value of SFC each after hidden status switch set, layout and hidden shape
In state arrangement set in the corresponding SFC of each hidden state subsequence VNF corresponding to each network function node cost-effectiveness value,
VNF is backed up.
In practical application, user can have reliability requirement to every SFC.Illustratively, as shown in figure 4, there is two
SFC, respectively business function chain 1 and business function chain 2.Business function chain 1 starts from customer end A, and Business Stream needs sequence is worn
It crosses FW (Fire Wall, firewall), LB (LoadBalance, flow equalizer) and NAT (Network Address
Translatio, network address translater), eventually flow to client C.Business function chain 2 starts from customer end B, and Business Stream is suitable
Sequence passes through FW, LB and GW (Gateway, gateway), eventually flows to client D.Wherein, 1 reliability requirement of business function chain is R1
=80%, the reliability requirement of business function chain 2 is R2=85%.
Illustratively, there are 5 data central nodes, be respectively as follows: DC1, DC2, DC3, DC4 and DC5.To business function chain
1 and business function chain 2 carry out layout after, layout result are as follows: 1 Business Stream of business function chain, which sequentially passes through, is deployed in DC1's
VFW is deployed in the vLB of DC3 and is deployed in vNAT (Virtualized Network Address Translatio, the void of DC4
Quasi- network address translater), reach client C.Wherein, it is 0.92 by the vFW reliability value for being deployed in DC1, vLB is deployed in
The reliability value of DC3 is 0.82, is 0.93 by the vNAT reliability value for being deployed in DC4, then the reliability that business function chain 1 is disposed
Value are as follows: 0.92 × 0.82 × 0.93=70.2% < R1=80%.
2 Business Stream of business function chain sequentially passes through the vFW for being deployed in DC2, is deployed in the vLB of DC3 and is deployed in DC5's
VGW (Virtualized Gateway, virtual gateway), eventually arrives at client D.Wherein, vFW is deployed in the reliability of DC2
Value is 0.97, is 0.82 by the vLB reliability value for being deployed in DC3, is 0.74 by the vGW reliability value for being deployed in DC5, business function
The reliability value that energy chain 2 is disposed is 0.97 × 0.82 × 0.74=58.86% < R2=85%.
As it can be seen that after carrying out layout to SFC, the reliability requirement of SFC not necessarily can satisfy the need of user in practical application
Ask, so, in the embodiment of the present invention, can based on above-mentioned hidden status switch set, after layout the corresponding reliability value of each SFC with
And in hidden status switch set in the corresponding SFC of each hidden state subsequence VNF corresponding to each network function node cost
Benefit value backs up VNF, after improving VNF backup, the reliability value of layout SFC.
As a kind of optional embodiment of the embodiment of the present invention, as shown in figure 5, above-mentioned the step of VNF is backed up,
It can specifically include:
S301 traverses each hidden state subsequence in hidden status switch set, by SFC corresponding to the hidden state subsequence
As current SFC.
For every SFC of above-mentioned layout, each hidden state subsequence in hidden status switch set is traversed, by the hidden shape
SFC corresponding to state subsequence is as current SFC.
S302 calculates the reliability value of current SFC.
As a kind of optional embodiment of the embodiment of the present invention, for each current SFC, calculate the current SFC can
It may is that by property value using the 6th default expression formula and calculate the reliability value of the current SFC.6th default expression formula can
With are as follows:
In formula, RpIndicating the reliability value of pth SFC, K indicates the network function number of nodes of pth SFC,It indicates
The reliability value of i-th of network function node of pth SFC.Wherein,By i-th of network in pth SFC layout process
The function type of data center's node location and i-th of network function node that functional node is disposed determines, i.e., when SFC is compiled
It is drained at the reliability value of i-th of network function node deployment VNF in data center's node, data center's node is by VNF's
Function type determine, that is to say, that when SFC layout complete,It is determined.
S303, judges whether the reliability value of current SFC is less than default reliability value.
After the reliability value of all SFC calculates after completing to layout, for every SFC, it can be determined that current SFC's
Whether reliability value is less than default reliability value.The default reliability value is the reliability requirement value that user is directed to this SFC.
Current SFC is placed in first set by S304 when the reliability value of current SFC is less than default reliability value,
And it will be placed in second set by the VNF of current SFC.
When the reliability value of current SFC is less than default reliability value, illustrate that the reliability value of current SFC after layout is discontented
Current SFC is placed in first set by the demand of sufficient user, and will be placed in second set by the VNF of current SFC.
S305 judges whether first set is empty.
Reliability value is placed in first set and is unsatisfactory for the corresponding SFC of user demand, if first set is sky, table
Show that the reliability value of all SFC after SFC layout is all met the needs of users, if first set be not it is empty, after indicating SFC layout
There is the reliability value of SFC to be unsatisfactory for the demand of user.
S306 calculates the cost-effectiveness value of every VNF in second set when first set is not sky.
First set is not demand that is empty, having the reliability value of SFC to be unsatisfactory for user after expression SFC layout, at this point it is possible to
For every VNF in second set, calculated using the above-mentioned 6th default expression formula each in first set after backing up the VNF
The reliability value of SFC.Then, the reliability value based on every SFC in first set after backup VNF, the first default reliability value,
The processing capability requirements amount of backup VNF and the unit capacity expense of backup VNF calculate the using the 7th default expression formula
The cost-effectiveness value of every VNF in two set.7th default expression formula can be with are as follows:
Wherein,
In formula,Indicate s-th of VNF example in than the m-th data central node,Indicate backupCost
Benefit value, p indicate that pth SFC, q indicate the quantity of SFC,Indicate backupTo the lifting values of pth SFC reliability,
RpIndicate backupThe reliability value of pth SFC afterwards, φpIt indicates the first default reliability value, can be the reliable of user setting
Property value,Indicate backupTo the promotion degree of all SFC reliabilities,It indicatesIt generates
Expense, αmIndicate the unit capacity expense of backup VNF,It indicatesProcessing capability requirements amount,It indicates's
Function type.
Work as backupThe reliability value R of pth SFC afterwardspIt, will when greater than the first default reliability valueTaking result is 1,
So that backupAfter generate higher reliability lifting values, and it is lower to generate expenseCost-effectiveness value with higher.
S307, VNF corresponding to maximum cost-effectiveness value are backed up.
In calculating second set after the cost-effectiveness value of every VNF, corresponding VNF can be worth to maximum cost-effectiveness
It is backed up.In practical application, can a VNF corresponding to maximum cost-effectiveness value back up, can also be to most great achievement
The corresponding multiple VNF of this benefit value are backed up, and the specific embodiment of the present invention is not limited thereto.
S308 calculates the reliability value of every SFC in first set after backing up.
In the embodiment of the present invention, after VNF corresponding to maximum cost-effectiveness value is backed up, first set after backup is calculated
In the reliability value of every SFC come pair specifically, reference can be made to the calculation of the reliability value of current SFC in step S302
The reliability value of every SFC is calculated in first set after backup, and this will not be repeated here for the embodiment of the present invention.
S309, if the reliability value of SFC is not less than default reliability value in first set after backup, by the SFC from
It is deleted in first set, and executes and judge whether first set is empty step.
The reliability value of the SFC can be sentenced after the reliability value of every SFC in first set after calculating backup
It is disconnected, judge whether it is not less than default reliability value, if after backup in first set SFC reliability value not less than it is default can
By property value, illustrates that the reliability value of SFC at this time is met the needs of users, which is deleted from first set, and return to execution
The step of S305, until the reliability value of all SFC all meets the needs of user in first set.
The method of combination of a kind of cross-domain SFC provided in an embodiment of the present invention, because the hidden state of HMM cannot be observed directly
It arrives, but can be arrived by observable sequence inspection, each observable sequence is to show as various shapes by probability density distribution
State, each observable sequence is generated by a status switch with corresponding probability density distribution, the embodiment of the present invention
In, the layout of cross-domain SFC is modeled as hidden Markov model, hidden Markov model is recycled, to each in a plurality of SFC
SFC carries out layout, obtains the corresponding hidden state subsequence of each SFC, because being based on being calculated network function node in SFC
It is deployed in hidden state probability corresponding in data center's node, the network function node of SFC is disposed, realizes SFC's
Layout can reduce the cross-domain bandwidth cost of SFC after layout, and then reduce consumed bandwidth resources after SFC layout.In addition,
VNF is backed up, the reliability value of SFC after backup is improved, so that the reliability value of every SFC can meet use after backup
The demand at family.
Illustratively, layout, obtained analogous diagram point are carried out using different method of combination to SFC in the embodiment of the present invention
Not as shown in Fig. 6 a to Fig. 6 d.Wherein, art methods 1 are the prior art described in background technique, prior art side
Method 2 are as follows: SFC Arrangement is modeled as integral linear programming problem, and then the layout realized, specific implementation process can be found in
The realization of the prior art, this will not be repeated here for the embodiment of the present invention.In the embodiment of the present invention, multiple data centers network is from extensive
It is chosen in accurate network topology structure topology-zoo, across data center bandwidth is set as 200Gbps, across data center band
Wide unit costs are the random value being evenly distributed between [0.01,0.02] $/Mbps, the unit IT resource expense of data center
For [0.05,0.10] $/unit, wherein the IT resource of different virtual network function node requests randomly selects between [1,3],
The reliability value of virtual network function node is set as [0.8,0.99], and the length of SFC is evenly distributed between [2,6], SFC's
Bandwidth on demand meets [10,100] Mbps and is uniformly distributed, the reliability value of SFC [0.95,0.98,0.99,0.995,
0.999] it randomly chooses.
Wherein, Fig. 6 a is VNF expense provided in an embodiment of the present invention and SFC demand magnitude relation analogous diagram, when the request of SFC
When quantity is less than 400, art methods 1 handle incoming SFC request using lesser VNF example.But when SFC's
When number of requests is greater than 600, the VNF expense of art methods 1 can significant increase.The embodiment of the present invention is asked in small-scale SFC
More VNF is used in the case of asking, this is because the embodiment of the present invention is handled according to the descending of the length of SFC, has ignored SFC
Other relevances to reduce the complex nature of the problem, lead to more VNF expenses.When handling extensive SFC request, the present invention
The VNF expense of embodiment and art methods 2 is not much different, this is because the embodiment of the present invention SFC layout is converted to it is hidden
The decoding problem of Markov model has fully considered the service efficiency of VNF, realizes the VNF expense of lower cost.
Fig. 6 b is cross-domain bandwidth cost provided in an embodiment of the present invention and SFC demand magnitude relation analogous diagram, identical in layout
When the SFC of quantity, art methods 1 have used more cross-domain bandwidth costs, this is because art methods 1 close SFC
And the use of VNF example is reduced into figure, but this will lead to higher bandwidth consumption.The embodiment of the present invention considers simultaneously
Bandwidth cost and VNF expense, compared with art methods 1, the bandwidth cost used reduces about 26.2%.When SFC is requested
When quantity is less than 400, the bandwidth cost and art methods 2 of the embodiment of the present invention are suitable.It is and existing when quantity is greater than 600
There is technical method 2 to compare, data center's inter-node bandwidth increased costs about 11.3% that the embodiment of the present invention uses, this is because
Output probability in the hidden Markov model of the embodiment of the present invention had both considered VNF expense, it is also considered that the reliability of VNF.
Therefore, the embodiment of the present invention sacrifices the cross-domain bandwidth cost in part to obtain the SFC layout result of high reliability.
Fig. 6 c is backup expense provided in an embodiment of the present invention and SFC demand magnitude relation analogous diagram, when SFC number of requests is small
When 600, use of the embodiment of the present invention and the comparable backup expense of art methods 2.When SFC number of requests is greater than 800
When, the backup expense that the embodiment of the present invention uses reduces 13.2%, this is because the embodiment of the present invention is using hidden Markov
Reliability requirement is considered when model layout SFC, the embodiment of the present invention, which sacrifices the cross-domain bandwidth of deployment, realizes higher reliability
SFC layout.
Fig. 6 d is overhead provided in an embodiment of the present invention and SFC demand magnitude relation analogous diagram, and overhead includes three parts:
VNF expense (as shown in Figure 6 a), cross-domain bandwidth cost (as shown in Figure 6 b) and VNF backup expense (as fig. 6 c).Obviously, with
The embodiment of the present invention is compared with the result that the method for art methods 2 obtains, and art methods 1 consume more expenses.With
Art methods 2 are compared with the result that the method for the embodiment of the present invention obtains, and the cost of art methods 1 is higher by about respectively
20.4% and 15.6%.When SFC quantity is greater than 600, the overhead of the embodiment of the present invention is higher by about than art methods 2
11.4%.
It is standby to VNF progress after SFC layout in the embodiment of the present invention although the expense of the embodiment of the present invention is relatively slightly higher
Part, it can be improved the reliability of SFC, better meet the demand of user.
Corresponding to above method embodiment, the embodiment of the invention provides the layout devices of cross-domain SFC a kind of, such as Fig. 7 institute
Show, the apparatus may include:
Module 401 is obtained, for obtaining a plurality of business function chain SFC to layout and multiple data center's nodes, often
It include multiple network function nodes in SFC, data center's node is used for the network function node disposed in SFC, a net
Network functional node is deployed in a data central node.
First building module 402, is produced when for being deployed in different data central node based on heterogeneous networks functional node
Raw virtual network function VNF expense and bandwidth cost determine heterogeneous networks functional node being deployed in different data centromere
Probability in point, and probability set is constructed based on identified multiple probabilities.
Second building module 403, for being transferred to transition probability determined by the second state based on first state, building turns
Move probability matrix;First network functional node in SFC is is deployed in the process institute in first data center's node by first state
Corresponding state, the second state are the process institute by the second network function node deployment in SFC in second data center's node
Corresponding state, the second network function node are the adjacent node of first network functional node.
Third constructs module 404, and the corresponding function of third network function node in SFC is exported under the third state for being based on
The output probability of type constructs output probability matrix;The third state is by third network function node deployment in SFC in third number
According to state corresponding to the process in central node.
4th building module 405, for general based on constructed probability set, transition probability matrix and output
Rate matrix constructs hidden Markov model.
Orchestration module 406, for utilizing probability, transition probability and output probability in hidden Markov model,
The hidden state probability that network function node deployment in SFC is corresponding in data center's node is calculated, and general based on hidden state
Rate carries out layout to each SFC in a plurality of SFC, obtains the corresponding hidden state subsequence of each SFC, it is corresponding to obtain a plurality of SFC
Multiple hidden state subsequences;The element for including in hidden state subsequence are as follows: the data center that network function node is disposed in SFC
Node.
It should be noted that the device of the embodiment of the present invention is corresponding with a kind of method of combination of cross-domain SFC shown in FIG. 1
Device, a kind of all embodiments of the method for combination of cross-domain SFC shown in FIG. 1 are suitable for the device, and can reach phase
Same or similar beneficial effect.
The layout device of a kind of cross-domain SFC provided in an embodiment of the present invention, because the hidden state of HMM cannot be observed directly
It arrives, but can be arrived by observable sequence inspection, each observable sequence is to show as various shapes by probability density distribution
State, each observable sequence is generated by a status switch with corresponding probability density distribution, the embodiment of the present invention
In, the layout of cross-domain SFC is modeled as hidden Markov model, hidden Markov model is recycled, to each in a plurality of SFC
SFC carries out layout, obtains the corresponding hidden state subsequence of each SFC, because being based on being calculated network function node in SFC
It is deployed in hidden state probability corresponding in data center's node, the network function node of SFC is disposed, realizes SFC's
Layout can reduce the cross-domain bandwidth cost of SFC after layout, and then reduce consumed bandwidth resources after SFC layout.
Optionally, it using the first default expression formula, determines and heterogeneous networks functional node is deployed in different data centromere
Probability in point.
Using the second default expression formula, determine that first state is transferred to the transition probability of the second state.
Expression formula is preset using third, determines and exports the corresponding function of third network function node in SFC under the third state
The output probability of type.
First default expression formula are as follows:
In formula, πmIndicate that by network function node deployment, in the probability of than the m-th data central node, M indicates data
The number of central node,Indicate s-th of VNF example in than the m-th data central node,It indicates first network function
Expense caused by s-th of VNF example of the energy node deployment in than the m-th data central node,It indicates initial network
Functional node is deployed in initial data central node, and first network function node deployment is produced in than the m-th data central node
Raw transfer bandwidth cost, 01 indicates the initial network functional node of SFC to the state transfer of first network function node, σ m
It indicates from initial data central node to the transfer of than the m-th data central node.
Second default expression formula are as follows:
In formula, Expression stateIt is transferred to stateTransfer
Probability, stateIt indicates corresponding to the process by (i-1)-th network function node deployment in nth data central node
State, stateIt indicates corresponding to the process by i-th of network function node deployment in than the m-th data central node
State,It indicates (i-1) a network function node deployment in data center node n, i-th of network function node
It is deployed on the transfer bandwidth cost of data center m, (i-1) i indicates (i-1) a network function node to i-th of network function section
The transfer of point, nm indicate from nth data central node to the transfer of than the m-th data central node,Indicate data center
The unit bandwidth expense of node n to data center node m, N indicate the network topology quantity of data center's node,It indicates
Bandwidth on demand amount between (i-1) a network function node of pth article SFC and i-th of network function node.
Third presets expression formula are as follows:
In formula,Expression stateLower output network function typeProbability, stateIt indicates i-th
State corresponding to process of a network function node deployment in than the m-th data central node,Indicate the i-th of pth SFC
The function type of a network function node,Expression stateUnder do not export network function typeProbability,It indicates to open caused by s-th of VNF example by i-th of network function node deployment in than the m-th data central node
Pin,Indicate the reliability of s-th of VNF example in than the m-th data central node,Indicate than the m-th data centromere
The function type of s-th of VNF example in point.
Optionally, a plurality of SFC to layout is SFC set;It is general using the probability in hidden Markov model, transfer
Rate and output probability calculate the hidden state probability that network function node deployment in SFC is corresponding in data center's node,
And layout is carried out to each SFC in a plurality of SFC based on hidden state probability, the corresponding hidden state subsequence of each SFC is obtained, is obtained
The corresponding multiple hidden state subsequences of a plurality of SFC are as follows:
Using probability, transition probability and the output probability in hidden Markov model, calculate network function in SFC
Energy node deployment hidden state probability corresponding in data center's node, and based on hidden state probability to each in a plurality of SFC
SFC carries out layout, obtains the corresponding hidden state subsequence of each SFC, obtains SFC and gathers corresponding hidden status switch set.
Optionally, orchestration module 406, comprising:
Judging submodule, for judging in SFC set with the presence or absence of to layout SFC.
Select submodule, for SFC set in exist when layout SFC, select SFC set in longest SFC as
Current layout SFC.
Layout submodule, for utilizing probability, transition probability and the output probability in hidden Markov model, meter
Calculate the hidden state probability that network function node deployment each in current layout SFC is corresponding in data center's node, and base
Layout is carried out to current layout SFC in hidden state probability, obtains the corresponding hidden state subsequence of current layout SFC.
Output sub-module, for, there is no when layout SFC, output SFC to gather corresponding hidden state sequence in SFC set
Column set.
Optionally, layout submodule is specifically used for:
Judge current layout SFC whether complete by layout;
If the current non-layout of layout SFC is completed, judge current layout SFC current layout network function node whether
For first network function node of current layout SFC;
If the current layout network function node of current layout SFC is first network function section of current layout SFC
Point is calculated the current layout network function node deployment then based on the probability in hidden Markov model in each number
According to the initial hidden state probability in central node;
If the current layout network function node of current layout SFC is not first network function of current layout SFC
Node is calculated then using the transition probability and output probability in hidden Markov model by the current layout network function section
Point is deployed in the hidden state probability of maximum in each data center's node, and it is corresponding preposition to record the maximum hidden state probability of acquisition
The position of data center's node;
By next network function node of the current layout network function node of current layout SFC, as current layout SFC
Current layout network function node, execution judge current layout SFC whether layout completion the step of;
If current layout SFC layout is completed, using the end network function node of current layout SFC as current network
Functional node, selection keep the maximum corresponding 4th data center's node deployment of the hidden state probability of current network functional node current
Network function node, and the 4th data center's node is stored in the corresponding hidden state subsequence of current layout SFC;
The maximum corresponding advance data central node of hidden state probability that will make current network functional node, is determined as current
5th data center's node corresponding to the previous network function node of network function node, in the middle part of the 5th data center's node
The previous network function node of current network functional node is affixed one's name to, and the 5th data center's node is stored in SFC pairs of current layout
In the hidden state subsequence answered, using the previous network function node of current network functional node as current network functional node;
Judge current network functional node previous network function node whether first network function for being current layout SFC
It can node;
If the previous network function node of current network functional node is not first network function of current layout SFC
Node then executes the maximum corresponding advance data central node of hidden state probability that will make current network functional node, is determined as
The step of 5th data center's node corresponding to the previous network function node of current network functional node;
If the previous network function node of current network functional node is first network function section of current layout SFC
Point then deletes current layout SFC from SFC set.
Optionally, the device of the embodiment of the present invention further include: backup module, for being based on hidden status switch set, layout
Each network function in the corresponding SFC of each hidden state subsequence in the corresponding reliability value of each SFC and hidden status switch set afterwards
The cost-effectiveness value of VNF corresponding to energy node, backs up VNF.
Optionally, backup module is specifically used for:
Traverse each hidden state subsequence in hidden status switch set, using SFC corresponding to the hidden state subsequence as
Current SFC;
Calculate the reliability value of current SFC;
Judge whether the reliability value of current SFC is less than default reliability value;
When the reliability value of current SFC is less than default reliability value, current SFC is placed in first set, and will
It is placed in second set by the VNF of current SFC;
Judge whether first set is empty;
When first set is not sky, the cost-effectiveness value of every VNF in second set is calculated;
VNF corresponding to maximum cost-effectiveness value is backed up;
Calculate the reliability value of every SFC in first set after backing up;
If the reliability value of SFC is not less than default reliability value in first set after backup, which is collected from first
It is deleted in conjunction, and executes and judge whether first set is empty step.
The embodiment of the invention also provides a kind of electronic equipment, as shown in figure 8, include processor 501, communication interface 502,
Memory 503 and communication bus 504, wherein processor 501, communication interface 502, memory 503 are complete by communication bus 504
At mutual communication,
Memory 503, for storing computer program;
Processor 501 when for executing the program stored on memory 503, is realized provided by the embodiment of the present invention
A kind of layout of cross-domain SFC.
An electronic equipment provided in an embodiment of the present invention because the hidden state of HMM cannot observe directly, but can pass through
Observable sequence inspection arrives, and each observable sequence is that various states are shown as by probability density distribution, each can
Observation sequence is generated by a status switch with corresponding probability density distribution, in the embodiment of the present invention, by cross-domain SFC's
Layout is modeled as hidden Markov model, recycles hidden Markov model, carries out layout to each SFC in a plurality of SFC, obtains
The corresponding hidden state subsequence of each SFC, because being based on being calculated network function node deployment in SFC in data center
Corresponding hidden state probability, disposes the network function node of SFC, realizes the layout of SFC, can reduce volume in node
The cross-domain bandwidth cost of SFC after row, and then reduce consumed bandwidth resources after SFC layout.
The communication bus that above-mentioned electronic equipment is mentioned can be Peripheral Component Interconnect standard (Peripheral Component
Interconnect, PCI) bus or expanding the industrial standard structure (Extended Industry Standard
Architecture, EISA) bus etc..The communication bus can be divided into address bus, data/address bus, control bus etc..For just
It is only indicated with a thick line in expression, figure, it is not intended that an only bus or a type of bus.
Communication interface is for the communication between above-mentioned electronic equipment and other equipment.
Memory may include random access memory (Random Access Memory, RAM), also may include non-easy
The property lost memory (Non-Volatile Memory, NVM), for example, at least a magnetic disk storage.Optionally, memory may be used also
To be storage device that at least one is located remotely from aforementioned processor.
Above-mentioned processor can be general processor, including central processing unit (Central Processing Unit,
CPU), network processing unit (Network Processor, NP) etc.;It can also be digital signal processor (Digital Signal
Processing, DSP), it is specific integrated circuit (Application Specific Integrated Circuit, ASIC), existing
It is field programmable gate array (Field-Programmable Gate Array, FPGA) or other programmable logic device, discrete
Door or transistor logic, discrete hardware components.
In another embodiment provided by the invention, a kind of computer readable storage medium is additionally provided, which can
It reads to be stored with computer program in storage medium, the computer program realizes that any of the above-described one kind is cross-domain when being executed by processor
The step of method of combination of SFC.
In another embodiment provided by the invention, a kind of computer program product comprising instruction is additionally provided, when it
When running on computers, so that computer executes the method for combination of any cross-domain SFC in above-described embodiment.
It should be noted that, in this document, relational terms such as first and second and the like are used merely to a reality
Body or operation are distinguished with another entity or operation, are deposited without necessarily requiring or implying between these entities or operation
In any actual relationship or order or sequence.Moreover, the terms "include", "comprise" or its any other variant are intended to
Non-exclusive inclusion, so that the process, method, article or equipment including a series of elements is not only wanted including those
Element, but also including other elements that are not explicitly listed, or further include for this process, method, article or equipment
Intrinsic element.In the absence of more restrictions, the element limited by sentence "including a ...", it is not excluded that
There is also other identical elements in process, method, article or equipment including the element.
Each embodiment in this specification is all made of relevant mode and describes, same and similar portion between each embodiment
Dividing may refer to each other, and each embodiment focuses on the differences from other embodiments.Especially for device/
For electronic equipment embodiment, since it is substantially similar to the method embodiment, so be described relatively simple, related place referring to
The part of embodiment of the method illustrates.
The foregoing is merely illustrative of the preferred embodiments of the present invention, is not intended to limit the scope of the present invention.It is all
Any modification, equivalent replacement, improvement and so within the spirit and principles in the present invention, are all contained in protection scope of the present invention
It is interior.
Claims (10)
1. a kind of method of combination of cross-domain SFC, which is characterized in that the described method includes:
The a plurality of business function chain SFC to layout and multiple data center's nodes are obtained, includes multiple in every SFC
Network function node, data center's node are used for the network function node disposed in the SFC, a network function
Node deployment is in data center's node;
When being deployed in different data central node based on heterogeneous networks functional node generated virtual network function VNF expense and
Bandwidth cost determines heterogeneous networks functional node being deployed in the probability in different data central node, and is based on institute really
Fixed multiple probabilities construct probability set;
It is transferred to transition probability determined by the second state based on first state, constructs transition probability matrix;The first state
For state corresponding to process that first network functional node in SFC is deployed in first data center's node, described second
State be by state corresponding to process of the second network function node deployment in second data center's node in SFC, it is described
Second network function node is the adjacent node of the first network functional node;
Based on the output probability for exporting the corresponding function type of third network function node in SFC under the third state, building output
Probability matrix;The third state be by third network function node deployment in the SFC in third data center node
State corresponding to process;
Based on the constructed probability set, the transition probability matrix and the output probability matrix, construct hidden
Markov model;
Using the probability, the transition probability and the output probability in the hidden Markov model, calculate
By the hidden state probability corresponding in data center's node of network function node deployment in the SFC, and it is based on the hidden shape
State probability carries out layout to each SFC in a plurality of SFC, obtains the corresponding hidden state subsequence of each SFC, obtains institute
State the corresponding multiple hidden state subsequences of a plurality of SFC;The element for including in the hidden state subsequence are as follows: network in the SFC
Data center's node that functional node is disposed.
2. the method according to claim 1, wherein being determined using the first default expression formula by the different nets
Network functional node is deployed in the probability in the different data central node;
Using the second default expression formula, determine that the first state is transferred to the transition probability of second state;
Expression formula is preset using third, determines that third network function node described in output SFC is corresponding under the third state
The output probability of function type;
The first default expression formula are as follows:
In formula, πmIndicate that by the network function node deployment, in the probability of than the m-th data central node, M is indicated in data
The number of heart node,Indicate s-th of VNF example in than the m-th data central node,It indicates first network function
Expense caused by s-th of VNF example of the energy node deployment in than the m-th data central node,It indicates initial network function
Energy node deployment is in initial data central node, and first network function node deployment is produced by than the m-th data central node
Transfer bandwidth cost, 01 indicates the initial network functional node of SFC to the state transfer of first network function node, σ m table
Show the transfer from initial data central node to than the m-th data central node;
The second default expression formula are as follows:
In formula, Expression stateIt is transferred to stateTransition probability,
StateIndicate state corresponding to the process by (i-1)-th network function node deployment in nth data central node,
StateIndicate state corresponding to the process by i-th of network function node deployment in than the m-th data central node,Indicate (i-1) a network function node deployment in data center node n, i-th of network function node deployment in
The transfer bandwidth cost of data center m, (i-1) i indicate (i-1) a network function node to i-th network function node
Transfer, nm indicate from nth data central node to the transfer of than the m-th data central node,Indicate data center's node n
To the unit bandwidth expense of data center node m, N indicates the network topology quantity of data center's node,Indicate pth item
Bandwidth on demand amount between (i-1) a network function node of SFC and i-th of network function node;
The third presets expression formula are as follows:
In formula,Expression stateLower output network function typeProbability, stateIt indicates i-th of net
Network functional node is deployed in state corresponding to the process in than the m-th data central node,Indicate i-th of net of pth SFC
The function type of network functional node,Expression stateUnder do not export network function typeProbability,Table
Show expense caused by s-th of VNF example by i-th of network function node deployment in than the m-th data central node,Indicate the reliability of s-th of VNF example in than the m-th data central node,It indicates in than the m-th data central node
S-th of VNF example function type.
3. according to the method described in claim 2, it is characterized in that, a plurality of SFC to layout is SFC set;The benefit
With the probability, the transition probability and the output probability in the hidden Markov model, calculating will be described
Network function node deployment hidden state probability corresponding in data center's node in SFC, and it is based on the hidden state probability
Layout is carried out to each SFC in a plurality of SFC, the corresponding hidden state subsequence of each SFC is obtained, obtains described a plurality of
The corresponding multiple hidden state subsequences of SFC are as follows:
Using the probability, the transition probability and the output probability in the hidden Markov model, calculate
By the hidden state probability corresponding in data center's node of network function node deployment in the SFC, and it is based on the hidden shape
State probability carries out layout to each SFC in a plurality of SFC, obtains the corresponding hidden state subsequence of each SFC, obtains institute
It states SFC and gathers corresponding hidden status switch set.
4. according to the method described in claim 3, it is characterized in that, described using described first in the hidden Markov model
Beginning probability, the transition probability and the output probability are calculated network function node deployment in the SFC in data center
Corresponding hidden state probability in node, and each SFC in a plurality of SFC is compiled based on the hidden state probability
Row obtains the corresponding hidden state subsequence of each SFC, obtains the step of SFC gathers corresponding hidden status switch set,
Include:
Judge in the SFC set with the presence or absence of to layout SFC;
If existed in SFC set to layout SFC, select the SFC gather in longest SFC as current layout
SFC;
Using the probability, the transition probability and the output probability in the hidden Markov model, calculate
By the hidden state probability corresponding in data center's node of each network function node deployment in the current layout SFC, and
Layout is carried out to the current layout SFC based on the hidden state probability, obtains the corresponding hidden state subgroup of the current layout SFC
Sequence;
If exporting the SFC there is no to layout SFC in the SFC set and gathering corresponding hidden status switch set.
5. according to the method described in claim 4, it is characterized in that, described using described first in the hidden Markov model
Beginning probability, the transition probability and the output probability are calculated each network function node in the current layout SFC
Be deployed in data center's node corresponding hidden state probability, and based on the hidden state probability to the current layout SFC into
Row layout, the step of obtaining the current layout SFC corresponding hidden state subsequence, comprising:
Judge the current layout SFC whether complete by layout;
If the non-layout of current layout SFC is completed, the current layout network function node of the current layout SFC is judged
Whether first network function node for being the current layout SFC;
If the current layout network function node of the current layout SFC is the first network function of the current layout SFC
Energy node is calculated then based on the probability in the hidden Markov model by the current layout network function node
The initial hidden state probability being deployed in each data center's node;
If the current layout network function node of the current layout SFC is not first network of the current layout SFC
Functional node, then using the transition probability and the output probability in the hidden Markov model, calculating will be deserved
Maximum hidden state probability of the preceding layout network function node deployment in each data center's node, and record and obtain the maximum
The position of the corresponding advance data central node of hidden state probability;
By next network function node of the current layout network function node of the current layout SFC, as the current volume
Arrange SFC current layout network function node, execution judge the current layout SFC whether layout completion the step of;
If the current layout SFC layout is completed, using the end network function node of the current layout SFC as current
Network function node, selection make the maximum corresponding 4th data center's node of the hidden state probability of the current network functional node
The current network functional node is disposed, and it is corresponding that the 4th data center's node is stored in the current layout SFC
In hidden state subsequence;
The maximum corresponding advance data central node of hidden state probability that will make the current network functional node, is determined as described
5th data center's node corresponding to the previous network function node of current network functional node, in the 5th data center
The previous network function node of the current network functional node is disposed in node, and the 5th data center's node is stored
In the corresponding hidden state subsequence of the current layout SFC, by the previous network function section of the current network functional node
Point is used as current network functional node;
Judge the current network functional node previous network function node whether first net for being the current layout SFC
Network functional node;
If the previous network function node of the current network functional node is not first network of the current layout SFC
Functional node then executes the maximum corresponding advance data centromere of hidden state probability that will make the current network functional node
Point is determined as the step of the 5th data center's node corresponding to the previous network function node of the current network functional node
Suddenly;
If the previous network function node of the current network functional node is the first network function of the current layout SFC
Energy node then deletes the current layout SFC from SFC set.
6. according to the method described in claim 3, it is characterized in that, the method also includes:
Based on each corresponding reliability value of SFC and the hidden set of state sequence after the hidden status switch set, layout
In conjunction in the corresponding SFC of each hidden state subsequence VNF corresponding to each network function node cost-effectiveness value, to described
VNF is backed up.
7. according to the method described in claim 6, it is characterized in that, described based on each after the hidden status switch set, layout
Each net in the corresponding SFC of each hidden state subsequence in the corresponding reliability value of SFC and the hidden status switch set
The cost-effectiveness value of VNF corresponding to network functional node, the step of backup to the VNF, comprising:
Traverse each hidden state subsequence in the hidden status switch set, using SFC corresponding to the hidden state subsequence as
Current SFC;
Calculate the reliability value of the current SFC;
Judge whether the reliability value of the current SFC is less than default reliability value;
When the reliability value of the current SFC is less than default reliability value, the current SFC is placed in first set,
And it will be placed in second set by the VNF of the current SFC;
Judge whether the first set is empty;
When the first set is not sky, the cost-effectiveness value of every VNF in the second set is calculated;
VNF corresponding to maximum cost-effectiveness value is backed up;
Calculate the reliability value of every SFC in the first set after backing up;
If the reliability value of SFC is not less than default reliability value in the first set after backup, by the SFC from described the
One set in delete, and execute it is described judge the first set whether be sky step.
8. a kind of layout device of cross-domain SFC, which is characterized in that described device includes:
Module is obtained, for obtaining a plurality of business function chain SFC to layout and multiple data center's nodes, described in every
It include multiple network function nodes in SFC, data center's node is used for the network function node disposed in the SFC,
One network function node deployment is in data center's node;
First building module, it is generated virtual when for being deployed in different data central node based on heterogeneous networks functional node
Heterogeneous networks functional node is deployed in first in different data central node by network function VNF expense and bandwidth cost, determination
Beginning probability, and probability set is constructed based on identified multiple probabilities;
Second building module constructs transition probability for being transferred to transition probability determined by the second state based on first state
Matrix;The first state is right for the process that first network functional node in SFC is deployed in first data center's node
The state answered, second state are by process of the second network function node deployment in second data center's node in SFC
Corresponding state, the second network function node are the adjacent node of the first network functional node;
Third constructs module, for based on exporting third network function node corresponding function type in SFC under the third state
Output probability constructs output probability matrix;The third state is by third network function node deployment in the SFC in third
State corresponding to process in data center's node;
4th building module, for based on the constructed probability set, the transition probability matrix and described defeated
Probability matrix out constructs hidden Markov model;
Orchestration module, for utilizing the probability, the transition probability and described in the hidden Markov model
Output probability calculates the hidden state probability that network function node deployment in the SFC is corresponding in data center's node, and
Layout is carried out to each SFC in a plurality of SFC based on the hidden state probability, obtains the corresponding hidden state of each SFC
Subsequence obtains the corresponding multiple hidden state subsequences of a plurality of SFC;The element for including in the hidden state subsequence are as follows:
Data center's node that network function node is disposed in the SFC.
9. a kind of electronic equipment, which is characterized in that including processor, communication interface, memory and communication bus, wherein processing
Device, communication interface, memory complete mutual communication by communication bus;
Memory, for storing computer program;
Processor when for executing the program stored on memory, realizes method and step as claimed in claim 1 to 7.
10. a kind of computer readable storage medium, which is characterized in that be stored with computer in the computer readable storage medium
Program, the computer program realize method and step as claimed in claim 1 to 7 when being executed by processor.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910472411.3A CN110166304A (en) | 2019-05-31 | 2019-05-31 | A kind of method of combination, device, electronic equipment and the storage medium of cross-domain SFC |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910472411.3A CN110166304A (en) | 2019-05-31 | 2019-05-31 | A kind of method of combination, device, electronic equipment and the storage medium of cross-domain SFC |
Publications (1)
Publication Number | Publication Date |
---|---|
CN110166304A true CN110166304A (en) | 2019-08-23 |
Family
ID=67630541
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201910472411.3A Pending CN110166304A (en) | 2019-05-31 | 2019-05-31 | A kind of method of combination, device, electronic equipment and the storage medium of cross-domain SFC |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN110166304A (en) |
Cited By (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN111510381A (en) * | 2020-04-23 | 2020-08-07 | 电子科技大学 | Service function chain deployment method based on reinforcement learning in multi-domain network environment |
CN114124818A (en) * | 2021-11-11 | 2022-03-01 | 广东工业大学 | Newly-added function node deployment optimization method for multicast transmission in SDN network |
CN114172820A (en) * | 2021-11-26 | 2022-03-11 | 广东技术师范大学 | Cross-domain SFC dynamic deployment method, device, computer equipment and storage medium |
CN116545877A (en) * | 2023-06-28 | 2023-08-04 | 广东技术师范大学 | Cross-domain SFC dynamic deployment method, device, computer equipment and storage medium |
Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
KR20180083820A (en) * | 2017-01-13 | 2018-07-23 | 주식회사 케이티 | cloud node for network services of virtual customer premises equipment |
CN108833335A (en) * | 2018-04-16 | 2018-11-16 | 中山大学 | A kind of network security function service catenary system based on cloud computing management platform Openstack |
-
2019
- 2019-05-31 CN CN201910472411.3A patent/CN110166304A/en active Pending
Patent Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
KR20180083820A (en) * | 2017-01-13 | 2018-07-23 | 주식회사 케이티 | cloud node for network services of virtual customer premises equipment |
CN108833335A (en) * | 2018-04-16 | 2018-11-16 | 中山大学 | A kind of network security function service catenary system based on cloud computing management platform Openstack |
Non-Patent Citations (1)
Title |
---|
XUXIA ZHONG等: "Cost-aware Service Function Chaining With Reliability Guarantees in NFV-enabled Inter-DC Network", 《2019 IFIP/IEEE INTERNATIONAL SYMPOSIUM ON INTEGRATED NETWORK MANAGEMENT》 * |
Cited By (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN111510381A (en) * | 2020-04-23 | 2020-08-07 | 电子科技大学 | Service function chain deployment method based on reinforcement learning in multi-domain network environment |
CN111510381B (en) * | 2020-04-23 | 2021-02-26 | 电子科技大学 | Service function chain deployment method based on reinforcement learning in multi-domain network environment |
CN114124818A (en) * | 2021-11-11 | 2022-03-01 | 广东工业大学 | Newly-added function node deployment optimization method for multicast transmission in SDN network |
CN114172820A (en) * | 2021-11-26 | 2022-03-11 | 广东技术师范大学 | Cross-domain SFC dynamic deployment method, device, computer equipment and storage medium |
CN114172820B (en) * | 2021-11-26 | 2024-03-05 | 广东技术师范大学 | Cross-domain SFC dynamic deployment method, device, computer equipment and storage medium |
CN116545877A (en) * | 2023-06-28 | 2023-08-04 | 广东技术师范大学 | Cross-domain SFC dynamic deployment method, device, computer equipment and storage medium |
CN116545877B (en) * | 2023-06-28 | 2023-09-05 | 广东技术师范大学 | Cross-domain SFC dynamic deployment method, device, computer equipment and storage medium |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN110166304A (en) | A kind of method of combination, device, electronic equipment and the storage medium of cross-domain SFC | |
CN106294614B (en) | Method and apparatus for accessing business | |
CN109919315B (en) | Forward reasoning method, device, equipment and storage medium of neural network | |
CN104487947B (en) | The unknowable resource allocation frame in domain | |
JP2012181852A (en) | Task processing scheduling method and device for implementing the same | |
CN110362377A (en) | The dispatching method and device of virtual machine | |
CN113033811A (en) | Processing method and device of two-quantum-bit logic gate | |
CN110046101A (en) | Page automated testing method, device and computer storage medium | |
CN109741060A (en) | Information inquiry system, method, device, electronic equipment and storage medium | |
CN114785696B (en) | Importance evaluation method and device for complex network node | |
CN106775493B (en) | A kind of storage control and I/O Request processing method | |
CN110457577A (en) | Data processing method, device, equipment and computer storage medium | |
Muppala et al. | Composite performance and availability analysis using a hierarchy of stochastic reward nets | |
CN1956454B (en) | Method and system for bundling and sending work units to a server based on a weighted cost | |
CN108459792A (en) | A kind of flow switching method, device and computer equipment | |
CN106934027A (en) | Distributed reptile realization method and system | |
CN108875090A (en) | A kind of song recommendations method, apparatus and storage medium | |
CN108132831A (en) | The processing method and processing unit of task | |
CN106339802A (en) | Task allocation method, task allocation device and electronic equipment | |
CN106886864A (en) | A kind of business construction method, device and electronic equipment | |
CN106155822A (en) | A kind of disposal ability appraisal procedure and device | |
CN108694413A (en) | Adaptively sampled unbalanced data classification processing method, device, equipment and medium | |
CN108259195A (en) | The determining method and system of the coverage of anomalous event | |
CN107526636A (en) | Resource identification method and device | |
CN108595149A (en) | Restructural multiply-add operation device |
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 | ||
WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20190823 |
|
WD01 | Invention patent application deemed withdrawn after publication |