CN107332913A - A kind of Optimization deployment method of service function chain in 5G mobile networks - Google Patents

A kind of Optimization deployment method of service function chain in 5G mobile networks Download PDF

Info

Publication number
CN107332913A
CN107332913A CN201710536013.4A CN201710536013A CN107332913A CN 107332913 A CN107332913 A CN 107332913A CN 201710536013 A CN201710536013 A CN 201710536013A CN 107332913 A CN107332913 A CN 107332913A
Authority
CN
China
Prior art keywords
vnf
mapping
server
link
virtual machine
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.)
Granted
Application number
CN201710536013.4A
Other languages
Chinese (zh)
Other versions
CN107332913B (en
Inventor
孙罡
赵东成
廖丹
孙健
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
University of Electronic Science and Technology of China
Original Assignee
University of Electronic Science and Technology of China
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by University of Electronic Science and Technology of China filed Critical University of Electronic Science and Technology of China
Priority to CN201710536013.4A priority Critical patent/CN107332913B/en
Publication of CN107332913A publication Critical patent/CN107332913A/en
Application granted granted Critical
Publication of CN107332913B publication Critical patent/CN107332913B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Classifications

    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04W—WIRELESS COMMUNICATION NETWORKS
    • H04W24/00—Supervisory, monitoring or testing arrangements
    • H04W24/02—Arrangements for optimising operational condition
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F9/00—Arrangements for program control, e.g. control units
    • G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
    • G06F9/44—Arrangements for executing specific programs
    • G06F9/455—Emulation; Interpretation; Software simulation, e.g. virtualisation or emulation of application or operating system execution engines
    • G06F9/45533—Hypervisors; Virtual machine monitors
    • G06F9/45558—Hypervisor-specific management and integration aspects
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L67/00—Network arrangements or protocols for supporting network services or applications
    • H04L67/01—Protocols
    • H04L67/10—Protocols in which an application is distributed across nodes in the network
    • H04L67/1001—Protocols in which an application is distributed across nodes in the network for accessing one among a plurality of replicated servers
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L67/00—Network arrangements or protocols for supporting network services or applications
    • H04L67/50—Network services
    • H04L67/51—Discovery or management thereof, e.g. service location protocol [SLP] or web services
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04W—WIRELESS COMMUNICATION NETWORKS
    • H04W28/00—Network traffic management; Network resource management
    • H04W28/02—Traffic management, e.g. flow control or congestion control
    • H04W28/0252—Traffic management, e.g. flow control or congestion control per individual bearer or channel
    • H04W28/0263—Traffic management, e.g. flow control or congestion control per individual bearer or channel involving mapping traffic to individual bearers or channels, e.g. traffic flow template [TFT]
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F9/00—Arrangements for program control, e.g. control units
    • G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
    • G06F9/44—Arrangements for executing specific programs
    • G06F9/455—Emulation; Interpretation; Software simulation, e.g. virtualisation or emulation of application or operating system execution engines
    • G06F9/45533—Hypervisors; Virtual machine monitors
    • G06F9/45558—Hypervisor-specific management and integration aspects
    • G06F2009/45595—Network integration; Enabling network access in virtual machine instances

Abstract

The invention discloses a kind of Optimization deployment method of service function chain in 5G mobile networks, belong to moving communicating field.The present invention is when the server resource and bandwidth resource allocation of bottom-layer network are asked to each service function chain that dynamic arrives, by considering VNF consolidation strategies, the virtual deployment way for reusing strategy and temporary link strategy in deployment, it is proposed that three kinds of different mapping schemes：Minimize computing resource cost：The maximum mapping cost that every virtual link energy being serviced provider receives is limited during deployment；The most path length of shortization service function chain：The maximum mapping cost that each virtual network function energy being serviced provider receives is limited during deployment；Use sex Great War game theoretical model justice optimization computing resource cost and link circuit resource cost.Three kinds of implementations of the present invention, can reach while mapping success rate, the resource utilization of bottom-layer network of service function chain request is improved, cause total mapping cost minimum as far as possible.

Description

A kind of Optimization deployment method of service function chain in 5G mobile networks
Technical field
The invention belongs to moving communicating field, and in particular to the Optimization deployment of service function chain in 5G mobile networks.
Background technology
With wireless flow and increasing substantially for servicing, because forth generation (4G) will can not meet the network demand in future, from And the research of the 5th generation (5G) mobile wireless network is promoted, to tackle the huge challenge of following diversified differentiated service.
With the explosive growth of mobile wireless flow, mobile operator is considering to extend its clothes using cloud computing Business, and tackle the tremendous growth of mobile data flow.In traditional communication network, network function or middle box are (as being grouped Data Network Gateway, gateway, fire wall, content filter, proxy server, wan optimization device, intruding detection system and Intrusion prevention system) realized by some special physical equipments and facility.However, as user is to more different services and new clothes The demand of business is continuously increased, and service supplier must correspondingly buy, store and operate new physicses equipment, to meet user's It is required that.Paid however, the buying of new physicses equipment will produce higher Capital expenditure and operation.And these physical equipments need The personnel for carrying out specialized training are disposed and safeguarded.
To solve above-mentioned problem, network function virtualization (NFV) has been proposed in researcher, it is intended to by the place of packet Manage to be mapped to from box in the middle of hardware and operate in the middle of the software in commercial hardware on box.Operate in the middle of software on box Network function is referred to as virtual network function (VNF).In network function virtualization, multiple virtual network functions are generally according to one The specific composition service function chain that is linked in sequence, to provide different network services.For example, in a mobile network, mobile network Communication between user and service terminal is needed by service function chain：User → gateway → packet data network gateway → anti- Wall with flues → intruding detection system → agency → terminal, this service function chain is generally deployed between user and service terminal and performed The security strategy of traffic filtering.Generally, the type of each virtual network function in service function chain and order are according to business The determinations such as the full-line policy of classification, SLA and operator.
In fact, core network virtualization and network function virtualize two crucial visions for describing following 5G frameworks. As a 5G key technology, network function virtualizes an important directions for having become wireless network architecture evolution.Make For a kind of emerging technology, network function is virtualized by industrial quarters, academia and the extensive concern of standardizing body.For clothes Be engaged in provider, and effective deployment/mapping services functional chain to 5G mobile networks is vital.At present, virtual network function Placement also turn into study hotspot, and there are some researchs on virtual network function Placement Problems.
It is current there has been proposed placement/the deployment scheme asked on virtual network function or service function chain (SFC), still Because the object of processing is virtual network or joint cloud, therefore, it has been suggested that most of deployment way be not suitable for 5G mobile networks. Such as the Capacitated NFV Location algorithms, its main thought is, when placing network function, to reduce as far as possible whole Volume grid cost, while meeting the size constraint of network node.Although which can realize the placement of virtual network function, It is that it is to be virtual network or combine cloud proposition, does not account for the characteristic and related constraint of 5G networks, therefore which is uncomfortable Close 5G mobile networks.And placement/deployment side of the virtual network function or service function chain for those suitable 5G mobile networks Case, the deployment for but only taking into account the virtual network function (e.g., packet data network gateway and gateway) of Radio Access Network is asked Topic, not in view of virtual network function (such as fire wall, content filter, the agency of core network and data center network Server, wan optimization device, intruding detection system and intrusion prevention system) deployment issue.For example pass through the Na Shili that bids By come the gateway quantity for needing playback to put and the path long length of compromising, although can realize that virtual network function is moved in 5G Placement in network, but the deployment issue of the virtual network function of Radio Access Network is only taken into account, however, they do not examine The deployment issue of the core network of worry and the virtual network function of data center network.
The content of the invention
The technical problem to be solved in the present invention is to propose a kind of known 5G mobile networks (bottom-layer network) and online service Under the position precondition of functional chain request, the position of mobile network user and service terminal, find out a kind of in each clothes of consideration Each virtual network function and link connection situation and meet minimum to consume under relevant constraint in business functional chain request Server resource, bandwidth resources and the blocking rate for reducing the request of service function chain are target, and the request of service function chain is put The dispositions method put.The deployment scheme of the present invention has considered the particularity of service function chain request, except to common band Wide resource requirement and server resource requirement are made that outside the configuration of optimization, also directed to the request of service function chain in communication delay side The strict demand in face proposes corresponding resolution policy.
The mapping process that the present invention is asked in service function chain, in order to effectively map/deployment services functional chain asks to arrive 5G mobile networks, introduce three kinds of effective strategies to reduce the cost of computing resource and link resources, so as to improve service function Chain asks acceptance rate.
(1) virtual machine reuses strategy.
In the present invention, in order to improve the resource utilization of server, introduce virtual machine and reuse strategy, i.e. when mapping one During individual virtual network function (VNF), the presence/existing virtual machine for realizing identical VNF can be reused.Virtual machine reuses strategy not The resource utilization of server can be only improved, and when server resource is limited, service function chain can also be improved The acceptance rate of request.In order to improve the resource utilization of server and reduce the computing resource of cost, as one VNF of mapping to clothes During business device, if considering, virtual machine is reused, and first determines whether that the server for placing current VNF whether there is reusable virtual machine, If in the presence of by the reusable virtual machine in current VNF trustships to this server；Otherwise, just current VNF trustships are arrived On new virtual machine on this server.Wherein reusable virtual machine is：(represent that current virtual machine is permitted with identifier is reused Permitted to be reused) trustship at least one and VNFiThe VNF of same type virtual machine, and the VNF and VNF of trustshipiDo not belong to In the request of same service function chain.
When VNF is hosted on reusable virtual machine, within the original service time of the virtual machine, the present invention is not examined The cost for considering computing resource cost, i.e. computing resource is zero, but when VNF service time exceeding the original service of the virtual machine Between, then need to calculate the computing resource cost of extra time.When VNF is using a new virtual machine, the present invention needs to calculate VNF Whole service time computing resource cost.Therefore, any VNF asked a service function chain computing resource cost Calculating process it is as follows：
Cost(VNFi→nk)=p (nk)×ε(VNFi)×Ti p
Wherein, Cost (VNFi→nk) represent i-th of VNF (VNFi) it is mapped to server nkComputing resource cost, p (nk) represent server nkResource unit costs, i.e., the present invention in symbol p () represent bracket in object unit cost, ε (VNFi) represent VNFiTo the resource constraint of server, such as CPU, memory and memory capacity.NV={ VNF1,VNF2,…, VNFnVirtual network function set in the request of service function chain is represented, subscript n represents that a service function chain request is included VNF number, Ti pRepresent VNFiThe time paid is needed, TiRepresent VNFiService time, ToExpression is reused virtual machine Original service time, πiRepresent the VNF of trustship on a virtual machine quantity, πi=1 expression current virtual machine is not reused, πi ＞ 1 represents to be reused by multiple VNF.Other VNF performance can be influenceed by reusing the virtual machine existed, therefore, defined δ and existed as trustship The maximum quantity of VNF on a virtual machine is deposited, i.e.,：
(2) VNF consolidation strategies.
In the present invention, as mapping VNFiWhen, if trustship VNFi-1Server have enough available resources, then can be with Consider mapping VNFiTo trustship VNFi-1Server on, this of the invention strategy is called VNF consolidation strategies.In the present invention, and Do not merge some VNF before the request of this service function chain is mapped, but during mapping, i-th of VNF of mapping to support On the i-th -1 VNF of pipe server, if this server has enough available resources.In VNF consolidation strategies, i-th VNF can only be mapped on the i-th -1 VNF of trustship server, please but trustship this service function chain can not be mapped to On other VNF asked server, routing issue of rattling is avoided with this.For example, as the VNF of service function chain request1It is mapped Physical node B, VNF are arrived2It has been mapped to after physical node F, in mapping VNF3When, it can be mapped to physical node F, such as Fruit physical node F has enough available resources, that is, allows VNF2And VNF3Merge, such VNF2And VNF3Communication is just Bandwidth resources need not be consumed, because VNF2And VNF3Communicated inside physical node F.It does not allow mapping VNF3Arrive Physical node B, because can so cause table tennis to route, this is undesirable.
(3) temporary link mapping policy.
In the present invention, when mapping i-th of VNF, while needing to map the i-th of connection i-th of VNF and the i-th -1 VNF Bar link ei, (virtual link) ensures to obtain the mapping scheme of near-optimization in i-th VNF.Traditional method is directed to I-th of VNF local optimum mapping solution is found, but this is it cannot be guaranteed that the road of the near-optimization of whole service function chain Footpath.In order to ensure the path of near-optimization, the receptance of whole service function chain request is improved, the present invention is calculating virtual link eiLink circuit resource cost when, generate one connection i-th of VNF and service terminal temporary link tei(virtual link), this The bandwidth demand of temporary link is equal to the bandwidth demand of i+1 bar link, and maps this temporary link teiTo underlying network Network, obtains mapping path.As mapping i-th of VNF and virtual link eiWhen, by using temporary link mapping policy, looked for this To i-th of a VNF near-optimization mapping scheme.In temporary link mapping policy, temporary link need not consume reality Link circuit resource, it is used only to, and i-th of VNF of constraint is too remote without departing from service terminal, is taken with the physics for ensureing i-th of VNF of trustship The adjoining link of business device has enough link circuit resources to map next virtual link, so as to improve service function chain receptance. The calculation of the link circuit resource cost of i-th virtual link is as follows：
Wherein,Represent connection VNFiAnd VNFi-1I-th virtual link (ei) link circuit resource cost, peiTable Show virtual link eiIn the mapping path (bottom path) of bottom-layer network, pteiRepresent temporary link teiIn reflecting for bottom-layer network Rays footpath (bottom path), esRepresent the physical link of bottom-layer network, xiRepresent virtual link eiResource constraint, such as bandwidth, And xi+1Then represent connection VNFi+1And VNFiI+1 bar virtual link resource constraint.In the present invention, acquiescence connection first Individual VNF and user virtual link are virtual link e1.Mapping to service function link request can be divided into two parts.First Part is the VNF for placing and allocating resources to the request of service function chain.Part II is mapping and distribution bandwidth resources to service The virtual link of functional chain request.The mapping process of service function chain is described as follows.
(1) VNF maps：
VNF mapping process can be expressed as：
Wherein NS1The set of the server and router of the bottom-layer network of current service functional chain request is distributed in expression, CN1Represent to distribute to the server resource that current service functional chain is asked, MN={ M (VNF1),M(VNF2),...,M(VNFn) table Show each VNF of current service functional chain request map record.M(VNFi) represent trustship VNFiServer, R (M (VNFi)) Represent server M (VNFi) available resources.CN={ ε (VNF1),ε(VNF2),...,ε(VNFn) represent all virtual network work( The resource constraint set of energy, VMiRepresent trustship VNFiVirtual machine (there is) ε (VMi) represent this exist virtual machine meter Resource is calculated, y ∈ { 0,1,2 ..., Y } represent the numbering of network area, L (M (VNFi)) represent server M (VNFi) where net The numbering in network region, and a server can only belong to a network area,Represent VNFiThis can be mapped to Network area,Represent VNFiThis network area can not be mapped to,Represent server M (VNFi) full Sufficient VNFiPosition constraint；IfThen it is unsatisfactory for.In 5G mobile networks, gateway and packet data gateway category In the function of Radio Access Network, they are generally only deployed in Radio Access Network, and fire wall, content filter, agency Server, wan optimization device, intruding detection system and intrusion prevention system are data center network or the function of core network, They are generally only deployed in core network and data center network.
(2) link maps of service function chain：
The link maps of service function chain are described as follows：
Wherein, ME={ M (e1),M(e2),...,M(e|Ev|) represent every virtual link that current service functional chain is asked Map record,The virtual link set of current service functional chain request is represented, | Ev| represent set EV's The virtual link quantity of element number, i.e. current service functional chain request.Represent that current service functional chain please The resource constraint set for all virtual links asked.P1Represent the bottom road end to end that the request of current service functional chain is mapped Footpath is gathered, and P1Every bottom pathIt is the physical link set E of bottom-layer networkSA subset close.CE1Represent to divide The link circuit resource of dispensing this service function chain request.Represent bottom pathAvailable bandwidth resources, b (es) expression thing Manage link esAvailable bandwidth resources,Represent bottom pathThe path delay of time, d (es) represent physical link esWhen Prolong.
Therefore, each service function chain asks the deployment issue in 5G mobile networks, I) link circuit resource cost minimization； II) the cost minimization of computing resource, can be described according to following linear programming (1)：
s.t.
The purpose of first aim is to reduce the cost of computing resource as much as possible.Increase service function chain is had one by this The probability in longer path.The purpose of second target is to reduce the cost of link circuit resource to greatest extent, i.e. as much as possible Shorten the path of whole service function chain.Meanwhile, the constraints in linear programming (1) is employed to ensure that following constraint：
Constraint 1 is used to ensure that the VNF of trustship on a virtual machine quantity to be no more than the given quantity δ of service supplier.
Constraint 2 gives the time that VNF needs to pay.
Constraint 3 and 4 ensures that the server used meets VNF computational resource requirements.
Constraint 5 and 6 ensures that used physical link meets the constraint of virtual link.
Constraint 7,8 and 9 ensures that the server used meets VNF position constraint.
Because linear programming (1) is multi-objective problem, it is impossible to direct solution, therefore, three solution party are proposed in the present invention Case, to solve multi-objective problem (1).Propose first solution be：The cost of computing resource is reduced to greatest extent；The Two solutions are：Shorten the path of whole service function chain；3rd solution is：By using sex Great War game (BOS) model is that VNF distributes resource and route, and a fair solution is found as computing resource cost and link circuit resource cost Certainly scheme.Above three solution is described in detail below：
(1) cost (abbreviation MC schemes) of computing resource is minimized.
In the solution, definitionFor every virtual link energy being serviced provider in the request of service function chain The maximum mapping cost of receiving, i.e.,Given by service provider, it is typically not greater than the charge of every virtual link.This Individual Optimized model can be described as following linear programming (2) to reduce computing resource cost as target：
s.t.
I.e. please in each service function chain that the server resource and broadband resource of bottom-layer network are distributed to dynamic arrival When asking, the optimal deployment scheme of current service functional chain request is obtained through the following steps：
Step 1：The virtual network function set N asked from service function chain to be mappedV={ VNF1,VNF2,..., VNFnFirst VNF start, be followed successively by current service functional chain request each VNFi(i=1 ..., n) determine alternative mapping Scheme collection：
(1) current VNF to be mapped is determinediAlternate servers collection：
From the available server collection U of bottom-layer networkSIn, position constraint, computational resource requirements will be met and be not aggregatedIn the servers that selected of each VNF as VNFiAlternate servers collection, i.e. VNFiIt is optional Alternate servers include：Server that other VNF had not been selected, VNFi-1Selected server (VNF consolidation strategies), ThereforeIt can also be expressed as：
(2) VNF is determinediAlternative mapping scheme collection：
With M (VNFi) represent to place VNFiAny alternative server, and determine server M (VNFi) on be used for trustship VNFiVirtual machine：Reusable virtual machine is judged whether, if so, then by VNFiTrustship is on reusable virtual machine；It is no Then by VNFiTrustship is on a new virtual machine；
Meeting ei(connection VNFi-1With VNFiVirtual link) link resource needs xiOn the premise of delay requirement, To eiBottom map paths are carried out, e is obtainediMapping pathWherein VNF0Represent user；
One connection VNF of generationiWith the temporary link te of user terminali, and by ei+1(connection VNFiWith VNFi+1It is virtual Link) link resource needs xi+1It is used as teiLink resource needs；Meeting teiLink resource needs on the premise of, it is right teiBottom map paths are carried out, te is obtainediMapping pathPlacement VNF is constrained by temporary link mapping policyi Server it is too remote without departing from user terminal；
Wherein mapping pathMapping pathIt there may be a plurality of.
Condition will be metMost short mapping pathAs to should Preceding M (VNFi) eiFinal mapping path, so as to ensure to eiBottom map paths corresponding to link circuit resource cost will not Exceed
Will the current M (VNF of correspondencei) alternative mapping scheme be saved in alternative mapping scheme and concentrate, wherein alternative mapping side Case includes：M(VNFi), trustship VNFiVirtual machine, eiFinal mapping path；
The combination of all VNF asked by current service functional chain the alternative mapping scheme of difference, can be obtained different Mapping set MN；
Finally, according to formulaCalculate respectively total computing resource of each mapping set into This, the Optimization deployment scheme that the corresponding mapping set of resources costs obtains the request of current service functional chain is calculated by minimum aggregate.
(2) path length (abbreviation SL schemes) of most shortization service function chain.
In the solution, definitionCan the receiving of being serviced provider for each VNF in the request of service function chain Maximum mapping cost, i.e.,Given by service provider, it is typically not greater than each VNF charge.This optimization mould Type is intended to the path length of most shortization service function chain, can be described as following linear programming (3)：
s.t.
I.e. please in each service function chain that the server resource and broadband resource of bottom-layer network are distributed to dynamic arrival When asking, the optimal deployment scheme of current service functional chain request is obtained through the following steps：
Step 1：The virtual network function set N asked from service function chain to be mappedV={ VNF1,VNF2,..., VNFnFirst VNF start, be followed successively by current service functional chain request each VNFi(i=1 ..., n) determine alternative mapping Scheme collection：
101：Concentrated from the available server of bottom-layer network, position constraint, computational resource requirements will be met and do not collected CloseIn the servers that selected of each VNF as VNFiInitial alternate servers collection Ui′；
Determine to be used for trustship VNF on each initial alternate servers respectivelyiVirtual machine：Judge whether reusable Virtual machine, if so, then by VNFiTrustship is on reusable virtual machine；Otherwise by VNFiTrustship is on a new virtual machine；
To each initial alternate servers nk∈Ui', determine to be used for trustship VNF thereon respectivelyiVirtual machine：Judge whether There is reusable virtual machine, if so, then by VNFiTrustship is on reusable virtual machine；Otherwise by VNFiTrustship is new at one Virtual machine on.And by its computing resource cost Cost (VNFi→nk) withCompare, will be less than or equal to's Initial alternate servers nkIt is used as VNFiAlternate servers nm, and record alternate servers nmUpper trustship VNFiVirtual machine；By All alternate servers nmObtain VNFiAlternate servers collection Ui。
102：Determine VNFiAlternative mapping scheme collection：
To each alternate servers nm∈Ui, meeting eiLink resource needs xiOn the premise of delay requirement, from clothes Be engaged in device nmTo VNFi-1Alternate servers bottom path in, search a shortest path be used as eiMapping pathWherein VNF0User is represented, i.e., for same server nmFor, VNFi-1There are how many alternate servers, then there is how many It is on eiMapping path
One connection VNF of generationiWith the temporary link te of user terminali, and by ei+1Link resource needs xi+1As teiLink resource needs；Meeting teiLink resource needs on the premise of, from server nmTo the bottom road of user terminal In footpath, search a shortest path and be used as teiMapping path
Will the current n of correspondencemAlternative mapping scheme be saved in VNFiAlternative mapping scheme concentrate, wherein alternative mapping side Case includes：nm, trustship VNFiVirtual machine, mapping path
Step 2：The combination of all VNF asked by current service functional chain the alternative mapping scheme of difference, obtains difference Mapping set；
According to formulaThe total link corresponding to each mapping set is calculated respectively Resources costs, Optimization Dept.'s management side of current service functional chain request is obtained by the corresponding mapping set of minimum total link resources costs Case.
(3) using sex Great War game theoretical model justice optimization computing resource cost and link circuit resource cost (abbreviation FOCL Scheme).
Sex Great War game theoretical model describes such a game scene：In game, two players have some common Interests, but common interests have different results, and have conflicting preference.For example, this would rather see same TV to Mr. and Mrs Program, but be not desired to individually see their respective TV programme, and this also prefers to see their favorite programs to Mr. and Mrs.
Therefore in the present invention, using computing resource cost and link circuit resource cost as in sex Great War game theoretical model Two players, it is based on two strategies：I the virtual machine existed, II) are reused when mapping VNF) use one when mapping VNF New virtual machine.The virtual machine of one presence is used to map VNF, and it can reduce the cost of computing resource, but it may lead Service function chain is caused to have a longer path.One new virtual machine is used to map VNF, and it is easier to find service function One of chain is close to optimal path, but it may result in the cost of higher computing resource.So each VNF mapping Journey is the process of a game, and computing resource cost and link circuit resource cost the two players carry out game to decide whether to use One already present virtual machine.In the present invention, a virtual machine existed will be reused as first strategy (i.e. 1-S), will Using a new virtual machine as second strategy (that is, 2-S), i.e., it regard computing resource cost as first player (i.e. 1- P), it regard link circuit resource cost as second player (i.e. 2-P).This game strategies is as represented by table 1.
The game strategies of table 1
The annotation of involved expression formula and parameter is as follows in table 1：
As the virtual machine mapping VNF using a presenceiWhen computing resource income；
As the virtual machine mapping VNF using a presenceiWhen link circuit resource income；
Cost(ME(VNFi))：As the virtual machine mapping VNF using a presenceiWhen computing resource cost；
Cost(pE(ei))：As the virtual machine mapping VNF using a presenceiWhen link circuit resource cost；
ME(VNFi)：As the virtual machine mapping VNF using a presenceiWhen VNFiMapping scheme；
pE(ei)：As the virtual machine mapping VNF using a presenceiWhen virtual link eiMapping path；
VNF is mapped when using a new virtual machineiWhen computing resource income；
VNF is mapped when using a new virtual machineiWhen link circuit resource income；
Cost(MN(VNFi))：VNF is mapped when using a new virtual machineiWhen computing resource cost；
Cost(pN(ei))：VNF is mapped when using a new virtual machineiWhen link circuit resource cost；
MN(VNFi)：VNF is mapped when using a new virtual machineiWhen VNFiMapping scheme；
pN(ei)：VNF is mapped when using a new virtual machineiWhen virtual link eiMapping path；
Wherein Cost (ME(VNFi)), Cost (MN(VNFi)), Cost (pE(ei)) and Cost (pN(ei)) according to following public affairs Formula is calculated：
Cost(Mλ(VNFi))=p (Mλ(VNFi))×ε(VNFi)×Ti p,
Wherein, subscript λ ∈ { E, N }, pE(tei)、pN(tei) represent respectively when the virtual machine, new virtual for using a presence Machine maps VNFiWhen corresponding temporary link teiMapping path.
In the model, there are two Pure strategy nash equilibria points, i.e., WithIn order to obtain VNFiNear-optimization mapping scheme, in this hair Bright middle selection total revenue highest Pure strategy nash equilibria point is focusing equilibrium point.It is VNF in the present invention to focus on equilibrium pointiReflect Penetrate scheme.Mapping process with each VNF is the process of a repeated game.It is above-mentioned on justice optimization computing resource cost and Link circuit resource cost, can be described by following linear programming (4)：
s.t.
λ={ E, N } (4)
Wherein,Represent trustship VNFiVirtual machineComputing resource, subscript λ is used for that to distinguish virtual machine to be Exist, or new, wherein E represents existing, and N represents new, similarly hereinafter.Bottom path is represented respectivelyTime delay, available bandwidth resources.
Step 1：The virtual network function set N asked from service function chain to be mappedV={ VNF1,VNF2,..., VNFnFirst VNF start, be followed successively by current service functional chain request each VNFi(i=1 ..., what is n) determined finally reflects Penetrate scheme：
101：It is determined that current VNF to be mappediAlternate servers collection
Concentrated from the available server of bottom-layer network, position constraint, computational resource requirements will be met and be not aggregatedIn the servers that selected of each VNF as VNFiAlternate servers collection
102：To alternate servers collectionIn each server nk, by server nkOn a new virtual machine conduct Trustship VNFiVirtual machine；
Meeting eiLink resource needs xiOn the premise of delay requirement, from placement VNFi-1Server (i.e. M (VNFi-1)) arrive server nkBottom path in, search a shortest path be used as eiMapping pathWherein M (VNF0) Represent the physical node where user terminal, i.e. user；
One connection VNF of generationiWith the temporary link te of user terminali, and by ei+1Link resource needs xi+1As teiLink resource needs；Meeting teiLink resource needs on the premise of, from server nkTo the bottom of user terminal In path, search a shortest path and be used as teiMapping path
By corresponding server nkComputing resource cost CostVNFN(VNFi→nk), link circuit resource costSum Obtain corresponding server nkTotal mapping cost, will total mapping cost minimization server nkIt is designated as
103：To alternate servers collectionIn each server nj, calculation server njTotal mapping cost, search total Map the n of cost minimizationjAnd be designated as
Wherein server njTotal mapping cost calculation be：
Judge server njIt is upper to whether there is reusable virtual machine, if it is not, then by server njTotal mapping cost set For infinity；
If so, then by VNFiTrustship connects VNF on reusable virtual machine, and meetingi-1With VNFiVirtual link eiLink resource needs xiOn the premise of delay requirement, from placement VNFi-1Server-to-server nkBottom path In, search a shortest path and be used as eiMapping pathWherein VNF0Represent user；And one connection VNF of generationiWith The temporary link te of user terminali, and VNF will be connectediWith VNFi+1Virtual link ei+1Link resource needs xi+1It is used as tei Link resource needs；Meeting teiLink resource needs on the premise of, from server njTo the bottom path of user terminal In, search a shortest path and be used as teiMapping pathBy server njComputing resource cost CostVNFE(VNFi→ nj), link circuit resource costSum obtains corresponding server njTotal mapping cost；
104：WillIn the server of total mapping cost minimization be used as and place VNFiServer M (VNFi), base In server M (VNFi) determine VNFiFinal mapping scheme, including server M (VNFi), server M (VNFi) on trustship VNFiVirtual machine, corresponding server M (VNFi) eiMapping path；
CauseFor definite value, so right Ask it is maximum then can directly take to be converted to take Cost (Mλ(VNFi)) withThe minimum mapping scheme of sum, wherein λ ∈ {N,E}。
Step 2：All VNF asked by current service functional chain final mapping scheme obtains current service functional chain please The Optimization deployment scheme asked.
In summary, by adopting the above-described technical solution, the beneficial effects of the invention are as follows：
(1) it is applied widely.Traditional virtual network function or service function chain mapping algorithm are to be directed to virtual net mostly What network and data center network were proposed, or do not account for the complete service functional chain in complete 5G networks.The present invention is carried Method can be suitably used for the complete service functional chain request in 5G networks, therefore compared with traditional mapping algorithm, of the invention is suitable It is wider with scope.
(2) mapping cost is low.Because the present invention proposes three kinds of schemes of the multi-objective problem for linear programming (1), Service function chain is asked to obtain on the basis of reusing strategy and temporary link mapping policy with reference to VNF consolidation strategies, virtual machine Low mapping cost deployment scheme, especially the third scheme considers the mode of computing resource and link circuit resource cost, The cost of its mapping scheme is lower.
(3) resource utilization is high.Because the present invention is when carrying out mapping processing, used virtual machine reuses strategy, VNF Consolidation strategy and temporary link mapping policy can reduce the consumption of resource, thus it is possible to improve the utilization rate of resource.
(4) mapping blocking rate is small.Because the present invention is when carrying out mapping processing, used virtual machine reuses strategy, VNF Consolidation strategy and temporary link mapping policy can reduce the consumption of resource, so, the successful possibility of mapping is also bigger, obstruction Rate is also just smaller.
Brief description of the drawings
Fig. 1 is the schematic diagram of a service function chain request, wherein, the numeral in rectangle frame above virtual network function Represent that the numeral above server resource demand, virtual link represents virtual link resource requirement, delay；
Fig. 2 is the schematic diagram and comparison diagram of a temporary link mapping policy, and wherein Fig. 2-a are not consider that temporary link reflects The mapping scheme of strategy is penetrated, Fig. 2-b, 2-c are that A~G in the mapping scheme for considering temporary link mapping policy, figure represents different Physical node (i.e. server), empty short-term represents the mapping path of virtual link, and imaginary point line represents the mapping road of temporary link Footpath.
Embodiment
To make the object, technical solutions and advantages of the present invention clearer, with reference to embodiment and accompanying drawing, to this hair It is bright to be described in further detail.
Using the 5G networks based on SDN as objective for implementation, Virtual network operator can be by Optimization deployment proposed by the invention Scheme (method of mapping services functional chain) is deployed on the key-course in SDN control router, and SDN control routers can be with Dispatch the control management function that itself carries and collect the whole network information, obtain all node resource situations in network, and link Resource, the information such as time delay.The topological and corresponding of the whole network can just be obtained by this centralized control mode router Resource information.When there is the request arrival of service function chain, the whole network letter that SDN control routers can be grasped according to oneself Breath, mapping method based on service function chain of the scheduling deployment on its key-course calculates the keys such as mapping cost, reject rate Parameter, and feed back to operator.
At the online mapping that the SDN for deploying the FOCL schemes of the present invention controls router to ask multiple service function chains During reason, it is ExpiredSFC that the service function chain request set left away is defined first, defines the arrival team of service function chain request It is classified as ArrivedSFC.In queue ArrivedSFC, each service function chain request is mapped one by one.Definition by The service function chain request set being blocked in bottom-layer network inadequate resource is SFCblo(the mapping request collection being referred to as blocked Close SFCblo).It is to the online processing procedure that multiple service function chains are asked then：
Input：
1st, bottom-layer network resource, including bottom-layer network GS=(NS,ES), the resource constraint SC=(C of bottom-layer networkE,CN, LN), all set of paths of bottom end to end P.Wherein NSRepresent server set, the set of routers, E of bottom-layer networkSRepresent The physical link set of bottom-layer network, CERepresent the attribute of physical link, such as bandwidth, time delay, unit cost, CNRepresent bottom The server of network and the attribute of router, the wherein attribute of server include the unit cost of server resource, server money Source (such as CPU, memory and storage volume), the attribute of router refers mainly to the disposal ability of router, LNRepresent underlying network The position of network server, router；
2nd, the service function chain request queue ArrivedSFC that reaches, wherein the request of each service function chain includes it Virtual network (virtual network function set NV, virtual link set EV) and placement constraint (each VNF resource constraint set CN、 Virtual link resource constraint set CE, virtual link maximum delay constraint set CD, VNF placement location constraint set LN, use The position L at familyU, service terminal position LT)。
Output：Map costWith the mapping request set SFC being blockedblo。
Step 1：Initially
Step 2：IfIt is carried out step 3；Otherwise, step 10 is gone to.
Step 3：IfBottom-layer network resource is just updated, is madeOtherwise, go to step 4。
Step 4：The service function chain request SFC of head of the queue is taken out from ArrivedVDCk, wherein subscript k is service function The identifier of chain request.
Step 5：Call SFCM Algorithm mappings SFCk。
Step 6：If finding SFCkA mapping schemeIt is carried out step 7；Otherwise, 8 are gone to step.
Step 7：OrderAnd bottom-layer network resource is updated, 9 are then gone to step, whereinExpression is reflected Penetrate schemeMapping cost, i.e. SFCkAll VNF total mapping cost (computing resource cost+link circuit resource cost) Sum.
Step 8：Update SFCblo=SFCblo∪{SFCk}。
Step 9：Update ArrivedSFC=ArrivedSFC-SFCk, then go to step 2.
Step 10：ReturnSFCblo。
Involved SFCM algorithms be used to find for each VNF in a service function chain request in above-mentioned steps 5 One mapping scheme, while finding the mapping path of service function chain request, and allocates resources to each VNF and every virtual chain Road.SFCM algorithms find each VNF using a mapping scheme that there is virtual machine and a new virtual machine, then pass through sex Great War betting model determines final mapping scheme.The mapping scheme that SFCM algorithms find out mapping cost minimization reflects as final Penetrate schemeThis mapping schemeIncluding mapping cost, VNF map record (server of placement, the void of trustship Plan machine) and service functional chain link map record (mapping path of a current VNF and upper VNF of connection virtual link).
Input：
1st, bottom-layer network GS=(NS,ES), the resource constraint SC=(C of bottom-layer networkE,CN,LN), all bottoms end to end Set of paths P.
2nd, a service function chain request GV=(NV,EV) and placement constraint PC=(CN,CE,CD,LN,LU,LT), such as Fig. 1 institutes Show.
Output：Mapping scheme
Step 501：All available servers are stored in USIn.
Step 502：Since first VNF, N is traveled through successivelyVIn each VNFi, next step is performed, if NVIn Each VNF has had stepped through, and just goes to step 512；
Step 503：Travel through each server n in bottom-layer networkk∈US, next step is performed, if the clothes in bottom-layer network Business device has had stepped through, and just goes to step 507；
Step 504：IfAnd nkOther VNF in not asked by this service function chain are used or met VNF consolidation strategies, perform step 505 to step 506；
Step 505：By current VNFiIt is mapped to server nkOn a new virtual machine on, and calculate and record CostVNFN(VNFi→nk) according to formula (5)；
Step 506：Based on current VNFiThe server n being placedk, one is generated from nkServer where to user terminal (by the position L of service terminalTLearn) temporary link tei, temporary link teiVirtual link resource constraint be xi+1；
Dijkstra's algorithm (shortest path first) is used again, based on bottom set of paths P, is meeting bandwidth and time delay about Link e is found under conditions of beami(connection VNFiWith VNFi-1Virtual link) and temporary link teiMost short bottom path WithCalculate and recordAccording to formula (6), calculate and record VNFiTotal mapping cost TCostVNFN(VNFi→ nk) according to formula (9), then go to step 50 3；
Step 507：Travel through each server n in bottom-layer networkj∈US, step 508 is performed, if in bottom-layer network Server has had stepped through, and just goes to step 511；
Step 508：IfAnd njOther VNF in not asked by this service function chain are used or met VNF consolidation strategies, perform step 509 to step 510；
Step 509：Judge server njIt is upper to whether there is reusable virtual machine, if it is not, then by server njTotal mapping Cost is set to infinity, and goes to step 507；If so, then performing step 510；
Step 510：By VNFiTrustship is calculated and record CostVNF on reusable virtual machineE(VNFi→nj) root According to formula (7)；
Meanwhile, based on current VNFiThe server n being placedj, one is generated from njServer where to user terminal (by The position L of service terminalTLearn) temporary link tei, temporary link teiVirtual link resource constraint be xi+1；
Reuse dijkstra's algorithm and find link eiWith temporary link teiMost short bottom pathWithCalculate and RecordAccording to formula (8), calculate and record VNFiTotal mapping cost TCostVNFE(VNFi→nj) according to formula (10) step 507, is gone to；
Step 511：New virtual machine mapping VNF is used alliMapping scheme in find a total mapping cost TCostVNFN(VNFi→nk) minimum mapping scheme；The virtual machine existed mapping VNF is used alliMapping scheme in look for To a total mapping cost TCostVNFE(VNFi→nj) minimum mapping scheme；Looked for according to formula (11) from the two schemes Scheme to mapping cost minimization is used as VNFiFinal mapping scheme M*(VNFi), and update the request of current service functional chain Mapping schemeReturn to step 502, whereinInitial value be empty set；
Step 512：Return
As use server nkOn a new virtual machine mapping VNFiWhen, VNFiBe mapped to instinct according to formula (5) count Calculate, virtual link eiMapping cost can be calculated according to formula (6)：
CostVNFN(VNFi→nk)=P (nk)ε(VNFi)Ti (5)
As use server njOn one presence virtual machine mapping VNFiWhen, VNFiBe mapped to instinct according to formula (7) calculate, virtual link eiMapping cost can be calculated according to formula (8)：
CostVNFE(VNFi→nj)=P (nj)ε(VNFi)max{Ti-To,0} (7)
As use server nkOn a new virtual machine mapping VNFiWhen, minimum is always mapped to instinct according to formula (9) Calculate：
As use server njOn one presence virtual machine mapping VNFiWhen, minimum is always mapped to instinct according to public affairs Formula (10) is calculated：
VNFiMinimum always be mapped to instinct according to formula (11) calculating：
TCostVNF(VNFi→nm)=min { TCostVNFN(VNFi→nk),TCostVNFE(VNFi→nj)} (11)
Wherein nmFor nkOr njIf, TCostVNFN(VNFi→nk)、TCostVNFE(VNFi→nj) result of calculation is identical, then Appoint and take one.
With reference to such as Fig. 2, in the service function chain request including 2 VNF, user (User) is placed on physical node A On, service terminal (Terminal) is placed on physical node G, as placement VNF1When, it is assumed that optional physical node have B, D, if only considering local optimum mapping, two alternatively all meet, and when selecting physical node B (Fig. 2-a), then can cause VNF1Away from service terminal, therefore the present invention passes through temporary link teiTo constrain VNF1It is too remote without departing from service terminal, calculating VNF1During link circuit resource cost between user, by temporary link teiMapping path be also contemplated for into, to e1Link money For the cost of source, it is：T1× ((physical pathway A → D unit cost) × x1+ (physical pathway D → G unit cost) × x2).So that in the case of same units cost, selection physical node D (Fig. 2-b) link circuit resource cost is smaller.So The rule determination VNF of total mapping cost minimization is considered in conjunction with VNF consolidation strategies, virtual reuse strategy afterwards2Physical node, So as to obtain the mapping scheme of service function chain request, as shown in fig. 2-c, its corresponding bottom path is：A→D→E→G. If not passing through temporary link tei, then the bottom path such as Fig. 2-a may be obtained, i.e. A → B → C → F → I → H → G will be bright It is aobvious to increase its link circuit resource cost.
Therefore, the present invention is in each service that the server resource and bandwidth resource allocation of bottom-layer network arrive to dynamic When functional chain is asked, by considering VNF consolidation strategies, the virtual deployment side for reusing strategy and temporary link strategy in deployment Formula, reaches while mapping success rate, the resource utilization of bottom-layer network of service function chain request is improved, causes as far as possible Total mapping cost is minimum.
The foregoing is only a specific embodiment of the invention, any feature disclosed in this specification, except non-specifically Narration, can alternative features equivalent by other or with similar purpose replaced；Disclosed all features or all sides Method or during the step of, in addition to mutually exclusive feature and/or step, can be combined in any way.

Claims (3)

1. a kind of Optimization deployment method of service function chain in 5G mobile networks, it is characterised in that comprise the following steps：
Step 1：The virtual network function set N asked from service function chain to be mappedV={ VNF1,VNF2,...,VNFn First VNF starts, and is followed successively by each VNF of current service functional chain requestiIt is determined that alternative mapping scheme collection, wherein i= 1 ..., n, n represent the VNF quantity of current service functional chain request：
101：It is determined that current VNF to be mappediAlternate servers collection：
Concentrated from the available server of bottom-layer network, position constraint, computational resource requirements will be met and be not aggregatedIn the servers that selected of each VNF as VNFiAlternate servers collection, i.e. VNFiIt is optional Alternate servers include：Server that other VNF had not been selected, VNFi-1Selected server (VNF consolidation strategies)；
102：Determine VNFiAlternative mapping scheme collection：
With M (VNFi) represent to place VNFiAny alternative server, determine server M (VNFi) on be used for trustship VNFiIt is virtual Machine：Reusable virtual machine is judged whether, if so, then by VNFiTrustship is on reusable virtual machine；Otherwise by VNFiSupport On Guan Yi new virtual machines；The reusable virtual machine is：With reuse identifier trustship at least one with VNFiThe VNF of same type virtual machine, and the VNF and VNF of trustshipiIt is not belonging to same service function chain request；
VNF is connected meetingi-1With VNFiVirtual link eiLink resource needs xiOn the premise of delay requirement, to eiEnter Row bottom map paths, obtain eiMapping pathWherein VNF0Represent user；
One connection VNF of generationiWith the temporary link te of user terminali, and VNF will be connectediWith VNFi+1Virtual link ei+1's Link resource needs xi+1It is used as teiLink resource needs；Meeting teiLink resource needs on the premise of, to teiCarry out Bottom map paths, obtain teiMapping path
Condition will be metMost short mapping pathIt is used as the current M of correspondence (VNFi) eiFinal mapping path, wherein esThe physical link of bottom-layer network is represented, symbol p () represents object in bracket Unit cost, TiRepresent VNFiService time,Represent eiDefault maximum mapping cost；
Will the current M (VNF of correspondencei) alternative mapping scheme be saved in alternative mapping scheme and concentrate, wherein alternative mapping scheme bag Include：M(VNFi), trustship VNFiVirtual machine, eiFinal mapping path；
Step 2：The combination of all VNF asked by current service functional chain the alternative mapping scheme of difference, obtains different reflect Penetrate set；
According to formulaTotal calculating resources costs of each mapping set are calculated respectively, by minimum It is total to calculate the Optimization deployment scheme that the corresponding mapping set of resources costs obtains the request of current service functional chain；
Wherein VNFiPaid-for time Ti pFor：If VNFiThe VNF of trustship quantity is equal to 1 on the virtual machine at place, then Ti p=Ti； If VNFiThe VNF of trustship quantity is more than 1 on the virtual machine at place, then Ti p=max { Ti-To, 0 }, wherein ToRepresent trustship VNFi Virtual machine the original service time.
2. the method as described in claim 1, it is characterised in that replace with step 101~102, step 2：
101：Concentrated from the available server of bottom-layer network, position constraint, computational resource requirements will be met and be not aggregatedIn the servers that selected of each VNF as VNFiInitial alternate servers collection；
Determine to be used for trustship VNF on each initial alternate servers respectivelyiVirtual machine：Judge whether reusable virtual Machine, if so, then by VNFiTrustship is on reusable virtual machine；Otherwise by VNFiTrustship is on a new virtual machine；It is described can The virtual machine of reuse is：With reusing the trustship of identifier at least one and VNFiThe VNF of same type virtual machine, and The VNF and VNF of trustshipiIt is not belonging to same service function chain request；
And each will correspond to VNF by initial alternate serversiComputing resource cost and VNFiDefault maximum mapping costCompare, will be less than or equal toInitial alternate servers be used as VNFiAlternate servers, and preserve institute State trustship VNF on alternate serversiVirtual machine；
Wherein each initial alternate servers correspondence VNFiComputing resource cost be：The unit cost of initial alternate servers × VNFiComputational resource requirements × VNFiPaid-for time T on initial alternate serversi p；
Paid-for time Ti pFor：If VNFiThe VNF of trustship quantity is equal to 1 on the virtual machine at place, then Ti p=Ti；If VNFiPlace Virtual machine on the VNF quantity of trustship be more than 1, then Ti p=max { Ti-To, 0 }, wherein TiRepresent VNFiService time, To Represent trustship VNFiVirtual machine the original service time；
102：Determine VNFiAlternative mapping scheme collection：
With M (VNFi) represent to place VNFiAny alternative server, meet VNF in satisfactioni-1With VNFiVirtual link eiLink Resource requirement xiOn the premise of delay requirement, from server M (VNFi-1) arrive M (VNFi) bottom path in, search one most Short path is used as eiMapping pathWherein M (VNF0) represent user terminal；
One connection VNF of generationiWith the temporary link te of user terminali, and VNF will be connectediWith VNFi+1Virtual link ei+1's Link resource needs xi+1It is used as teiLink resource needs；Meeting teiLink resource needs on the premise of, from server M (VNFi) into the bottom path of user terminal, search a shortest path and be used as teiMapping path
Will the current M (VNF of correspondencei) alternative mapping scheme be saved in alternative mapping scheme and concentrate, wherein alternative mapping scheme bag Include：M(VNFi), trustship VNFiVirtual machine, mapping path
Step 2：The combination of all VNF asked by current service functional chain the alternative mapping scheme of difference, obtains different reflect Penetrate set；
According to formulaThe total link resource corresponding to each mapping set is calculated respectively Cost, the Optimization deployment scheme of current service functional chain request is obtained by the corresponding mapping set of minimum total link resources costs, Wherein EVRepresent the virtual link set of current service functional chain request, esRepresent the physical link of bottom-layer network.
3. a kind of Optimization deployment method of service function chain in 5G mobile networks, it is characterised in that comprise the following steps：
Step 1：The virtual network function set N asked from service function chain to be mappedV={ VNF1,VNF2,...,VNFn First VNF starts, and is followed successively by each VNF of current service functional chain requestiThe final mapping scheme determined, wherein i= 1 ..., n, n represent the VNF quantity of current service functional chain request：
101：It is determined that current VNF to be mappediAlternate servers collection
Concentrated from the available server of bottom-layer network, position constraint, computational resource requirements will be met and be not aggregatedIn the servers that selected of each VNF as VNFiAlternate servers collection
102：To alternate servers collectionIn each server nk, by server nkOn a new virtual machine be used as trustship VNFiVirtual machine；
VNF is connected meetingi-1With VNFiVirtual link eiLink resource needs xiOn the premise of delay requirement, from placement VNFi-1Server-to-server nkBottom path in, search a shortest path be used as eiMapping pathWherein VNF0Represent user；
One connection VNF of generationiWith the temporary link te of user terminali, and VNF will be connectediWith VNFi+1Virtual link ei+1's Link resource needs xi+1It is used as teiLink resource needs；Meeting teiLink resource needs on the premise of, from server nkInto the bottom path of user terminal, search a shortest path and be used as teiMapping path
By corresponding server nkComputing resource cost CostVNFN(VNFi→nk), link circuit resource costSum is obtained Corresponding server nkTotal mapping cost, will total mapping cost minimization server nkIt is designated as
Wherein CostVNFN(VNFi→nm)=P (nm)ε(VNFi)Ti,Symbol p () represents the unit cost of object in bracket, ε (VNFi) represent VNFiComputational resource requirements, TiRepresent VNFiService when Between, esRepresent the physical link of bottom-layer network；
103：To alternate servers collectionIn each server nj, calculation server njTotal mapping cost, lookup be always mapped to This minimum njAnd be designated as
Wherein server njTotal mapping cost calculation be：
Judge server njIt is upper to whether there is reusable virtual machine, if it is not, then by server njTotal mapping cost be set to nothing It is poor big；The reusable virtual machine is：With reusing the trustship of identifier at least one and VNFiThe VNF of same type Virtual machine, and the VNF and VNF of trustshipiIt is not belonging to same service function chain request；
If so, then by VNFiTrustship connects VNF on reusable virtual machine, and meetingi-1With VNFiVirtual link ei's Link resource needs xiOn the premise of delay requirement, from placement VNFi-1Server-to-server nkBottom path in, look into A shortest path is looked for as eiMapping pathWherein VNF0Represent user；One connection VNF of generationiWith user terminal Temporary link tei, and VNF will be connectediWith VNFi+1Virtual link ei+1Link resource needs xi+1It is used as teiLink circuit resource Demand；Meeting teiLink resource needs on the premise of, from server njInto the bottom path of user terminal, one is searched Shortest path is used as teiMapping pathBy server njComputing resource cost CostVNFE(VNFi→nj), link money Source costSum obtains corresponding server njTotal mapping cost；
Wherein CostVNFE(VNFi→nj)=P (nj)ε(VNFi)max{Ti-To, 0 },esRepresent the physical link of bottom-layer network, ToRepresent trustship VNFiVirtual machine The original service time；
104：WillIn the server of total mapping cost minimization be used as and place VNFiServer M (VNFi), based on service Device M (VNFi) determine VNFiFinal mapping scheme, including server M (VNFi), server M (VNFi) on trustship VNFiVoid Plan machine, corresponding server M (VNFi) eiMapping path；
Step 2：All VNF asked by current service functional chain final mapping scheme obtains the request of current service functional chain Optimization deployment scheme.
CN201710536013.4A 2017-07-04 2017-07-04 Optimized deployment method of service function chain in 5G mobile network Active CN107332913B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201710536013.4A CN107332913B (en) 2017-07-04 2017-07-04 Optimized deployment method of service function chain in 5G mobile network

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201710536013.4A CN107332913B (en) 2017-07-04 2017-07-04 Optimized deployment method of service function chain in 5G mobile network

Publications (2)

Publication Number Publication Date
CN107332913A true CN107332913A (en) 2017-11-07
CN107332913B CN107332913B (en) 2020-03-27

Family

ID=60197782

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201710536013.4A Active CN107332913B (en) 2017-07-04 2017-07-04 Optimized deployment method of service function chain in 5G mobile network

Country Status (1)

Country Link
CN (1) CN107332913B (en)

Cited By (47)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN108134843A (en) * 2018-01-26 2018-06-08 重庆邮电大学 A kind of service function chain dispositions method under 5G-C-RAN scenes
CN108200202A (en) * 2018-02-06 2018-06-22 电子科技大学 A kind of service function chain security deployment method that network is calculated applied to cloud and mist
CN108259238A (en) * 2018-01-10 2018-07-06 重庆邮电大学 The method that network function service chaining maps in data center
CN108322333A (en) * 2017-12-28 2018-07-24 广东电网有限责任公司电力调度控制中心 A kind of laying method of the virtual network function based on genetic algorithm
CN108400940A (en) * 2018-02-27 2018-08-14 西南交通大学 A kind of multicast virtual network function dispositions method based on Estimation of Distribution Algorithm
CN108429685A (en) * 2018-04-28 2018-08-21 电子科技大学 A kind of service function chain method for searching based on Segment routing technology
CN108471389A (en) * 2018-03-12 2018-08-31 电子科技大学 A kind of switch system based on service function chain
CN108494596A (en) * 2018-03-23 2018-09-04 西安电子科技大学 The cooperating type structure relied between multiple VNF and mapping SFC methods
CN108616425A (en) * 2018-04-28 2018-10-02 电子科技大学 A kind of alleviation method of service function chain cascading failure risk
CN108684046A (en) * 2018-04-23 2018-10-19 重庆邮电大学 A kind of access net service function chain dispositions method based on incidental learning
CN108696596A (en) * 2018-05-29 2018-10-23 南京财经大学 The resource provider method of Services Composition problem in cloud integrated wireless-fiber hybrid access network network
CN108923961A (en) * 2018-06-25 2018-11-30 电子科技大学 A kind of multiple entry network service function chain optimization method
CN108965020A (en) * 2018-07-27 2018-12-07 北京邮电大学 Cross-domain virtual network mapping method and device thereof, computer-readable medium
CN109547241A (en) * 2018-11-14 2019-03-29 南京大学 A kind of virtual network function dispositions method towards NUMA architecture based on Markov algorithm
CN109617811A (en) * 2019-01-29 2019-04-12 电子科技大学 The quick migration method of mobile application in a kind of cloud network
CN109831319A (en) * 2018-12-27 2019-05-31 南京大学 A kind of network function dispositions method considering multi dimensional resource constraint
CN109842528A (en) * 2019-03-19 2019-06-04 西安交通大学 A kind of dispositions method of the service function chain based on SDN and NFV
CN110018832A (en) * 2019-02-20 2019-07-16 中国人民解放军海军工程大学 A kind of Radar Software deployment of components strategy based on improvement Dynamic Programming
CN110022230A (en) * 2019-03-14 2019-07-16 北京邮电大学 The parallel dispositions method of service chaining and device based on deeply study
CN110049130A (en) * 2019-04-22 2019-07-23 北京邮电大学 A kind of service arrangement and method for scheduling task and device based on edge calculations
CN110120978A (en) * 2019-05-17 2019-08-13 电子科技大学 A kind of method for security protection of elasticity user's cloud computing resources
CN110190987A (en) * 2019-05-08 2019-08-30 南京邮电大学 Based on backup income and the virtual network function reliability dispositions method remapped
CN110298381A (en) * 2019-05-24 2019-10-01 中山大学 A kind of cloud security service functional tree Network Intrusion Detection System
CN110311816A (en) * 2019-06-28 2019-10-08 上海交通大学 The VNF laying method of the co-located interference perception of VNF under a kind of NFV environment
CN110311862A (en) * 2019-06-14 2019-10-08 北京邮电大学 A kind of service chaining mapping method and device
CN110380898A (en) * 2019-07-08 2019-10-25 紫光云技术有限公司 Multi-operator edge service chain layout strategy based on Hyperledger Fabric
CN110446121A (en) * 2019-04-07 2019-11-12 西安电子科技大学 Virtual network function service chaining mapping method based on betweenness center degree
CN110535705A (en) * 2019-08-30 2019-12-03 西安邮电大学 A kind of service function chain building method of adaptive user delay requirement
CN110545228A (en) * 2019-07-26 2019-12-06 北京邮电大学 service function chain request processing method and system
CN110851235A (en) * 2019-11-04 2020-02-28 中国人民解放军战略支援部队信息工程大学 Virtual network function deployment method suitable for multidimensional resource optimization configuration
CN111147435A (en) * 2018-11-05 2020-05-12 南宁富桂精密工业有限公司 Reverse playback processing method
CN111385202A (en) * 2020-03-17 2020-07-07 重庆邮电大学 Route distribution method based on virtual network function
CN111526070A (en) * 2020-04-29 2020-08-11 重庆邮电大学 Service function chain fault detection method based on prediction
WO2020181496A1 (en) * 2019-03-12 2020-09-17 Nokia Shanghai Bell Co., Ltd. Method, device and computer readable medium for service chain
CN111800291A (en) * 2020-05-27 2020-10-20 北京邮电大学 Service function chain deployment method and device
CN112083933A (en) * 2020-08-27 2020-12-15 重庆大学 Service function chain deployment method based on reinforcement learning
WO2020255025A1 (en) * 2019-06-18 2020-12-24 Telefonaktiebolaget Lm Ericsson (Publ) Method and system for deploying a new service function chain (sfc) based on similarities with previously deployed sfcs
CN113032096A (en) * 2021-03-17 2021-06-25 西安电子科技大学 SFC mapping method based on node importance and user demand dual perception
CN113225211A (en) * 2021-04-27 2021-08-06 中国人民解放军空军工程大学 Fine-grained service function chain extension method
CN113348651A (en) * 2019-01-24 2021-09-03 威睿公司 Dynamic inter-cloud placement of sliced virtual network functions
CN113535388A (en) * 2021-06-25 2021-10-22 广西综合交通大数据研究院 Task-oriented service function aggregation method
CN114124796A (en) * 2021-10-26 2022-03-01 清华大学 Service chain deployment processing method and device, computer storage medium and terminal
CN114205317A (en) * 2021-10-21 2022-03-18 北京邮电大学 Service function chain SFC resource allocation method based on SDN and NFV and electronic equipment
CN114650225A (en) * 2022-03-14 2022-06-21 西安电子科技大学 Service function chain SFC mapping method based on particle swarm optimization algorithm
US11588733B2 (en) 2019-05-14 2023-02-21 Vmware, Inc. Slice-based routing
US11595315B2 (en) 2019-05-14 2023-02-28 Vmware, Inc. Quality of service in virtual service networks
US11902080B2 (en) 2019-05-14 2024-02-13 Vmware, Inc. Congestion avoidance in a slice-based network

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN105141617A (en) * 2015-09-14 2015-12-09 上海华为技术有限公司 Deploying and adjusting method for service functions among data centers and deploying and adjusting device for service functions among data centers
CN105430051A (en) * 2015-10-30 2016-03-23 浙江工商大学 Service function chaining construction method for SDN
CN106411678A (en) * 2016-09-08 2017-02-15 清华大学 Bandwidth guarantee type virtual network function (VNF) deployment method
CN106487556A (en) * 2015-08-28 2017-03-08 中兴通讯股份有限公司 The dispositions method of business function SF and device
US20170104847A1 (en) * 2015-10-12 2017-04-13 Fujitsu Limited Vertex-centric service function chaining in multi-domain networks

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN106487556A (en) * 2015-08-28 2017-03-08 中兴通讯股份有限公司 The dispositions method of business function SF and device
CN105141617A (en) * 2015-09-14 2015-12-09 上海华为技术有限公司 Deploying and adjusting method for service functions among data centers and deploying and adjusting device for service functions among data centers
US20170104847A1 (en) * 2015-10-12 2017-04-13 Fujitsu Limited Vertex-centric service function chaining in multi-domain networks
CN105430051A (en) * 2015-10-30 2016-03-23 浙江工商大学 Service function chaining construction method for SDN
CN106411678A (en) * 2016-09-08 2017-02-15 清华大学 Bandwidth guarantee type virtual network function (VNF) deployment method

Non-Patent Citations (3)

* Cited by examiner, † Cited by third party
Title
MICHAEL TILL BECK等: "Scalable and coordinated allocation of service function chains", 《ELSEVIER》 *
孙罡: "虚拟网络的映射技术研究", 《中国博士学位论文全文数据库 信息科技辑（月刊）》 *
赵东成: "数据中心虚拟机的迁移方法研究", 《中国优秀硕士学位论文全文数据库 信息科技辑（月刊）》 *

Cited By (79)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN108322333A (en) * 2017-12-28 2018-07-24 广东电网有限责任公司电力调度控制中心 A kind of laying method of the virtual network function based on genetic algorithm
CN108322333B (en) * 2017-12-28 2021-03-02 广东电网有限责任公司电力调度控制中心 Virtual network function placement method based on genetic algorithm
CN108259238A (en) * 2018-01-10 2018-07-06 重庆邮电大学 The method that network function service chaining maps in data center
CN108259238B (en) * 2018-01-10 2021-06-18 重庆邮电大学 Method for mapping network function service chain in data center
CN108134843A (en) * 2018-01-26 2018-06-08 重庆邮电大学 A kind of service function chain dispositions method under 5G-C-RAN scenes
CN108134843B (en) * 2018-01-26 2020-07-31 重庆邮电大学 Service function chain deployment method under 5G-C-RAN scene
CN108200202A (en) * 2018-02-06 2018-06-22 电子科技大学 A kind of service function chain security deployment method that network is calculated applied to cloud and mist
CN108200202B (en) * 2018-02-06 2019-11-12 电子科技大学 A kind of service function chain security deployment method calculating network applied to cloud and mist
CN108400940A (en) * 2018-02-27 2018-08-14 西南交通大学 A kind of multicast virtual network function dispositions method based on Estimation of Distribution Algorithm
CN108471389A (en) * 2018-03-12 2018-08-31 电子科技大学 A kind of switch system based on service function chain
CN108471389B (en) * 2018-03-12 2020-06-12 电子科技大学 Switch system based on service function chain
CN108494596A (en) * 2018-03-23 2018-09-04 西安电子科技大学 The cooperating type structure relied between multiple VNF and mapping SFC methods
CN108494596B (en) * 2018-03-23 2020-02-21 西安电子科技大学 Collaborative construction and SFC (Small form-factor computing) mapping method for dependency among multiple VNFs (virtual network configuration functions)
CN108684046B (en) * 2018-04-23 2021-07-06 重庆邮电大学 Random learning-based access network service function chain deployment method
CN108684046A (en) * 2018-04-23 2018-10-19 重庆邮电大学 A kind of access net service function chain dispositions method based on incidental learning
CN108616425A (en) * 2018-04-28 2018-10-02 电子科技大学 A kind of alleviation method of service function chain cascading failure risk
CN108616425B (en) * 2018-04-28 2021-06-01 电子科技大学 Method for relieving cascade failure risk of service function chain
CN108429685A (en) * 2018-04-28 2018-08-21 电子科技大学 A kind of service function chain method for searching based on Segment routing technology
CN108429685B (en) * 2018-04-28 2020-08-04 电子科技大学 Service function chain routing method based on segmented routing technology
CN108696596B (en) * 2018-05-29 2021-03-02 南京财经大学 Resource providing method for service combination problem in cloud integrated wireless-optical fiber hybrid access network
CN108696596A (en) * 2018-05-29 2018-10-23 南京财经大学 The resource provider method of Services Composition problem in cloud integrated wireless-fiber hybrid access network network
CN108923961A (en) * 2018-06-25 2018-11-30 电子科技大学 A kind of multiple entry network service function chain optimization method
CN108923961B (en) * 2018-06-25 2020-02-21 电子科技大学 Multi-entry network service function chain optimization method
CN108965020A (en) * 2018-07-27 2018-12-07 北京邮电大学 Cross-domain virtual network mapping method and device thereof, computer-readable medium
CN111147435A (en) * 2018-11-05 2020-05-12 南宁富桂精密工业有限公司 Reverse playback processing method
CN109547241A (en) * 2018-11-14 2019-03-29 南京大学 A kind of virtual network function dispositions method towards NUMA architecture based on Markov algorithm
CN109547241B (en) * 2018-11-14 2021-07-30 南京大学 Virtual network function deployment method facing NUMA (non Uniform memory Access) architecture based on Markov algorithm
CN109831319B (en) * 2018-12-27 2021-10-26 南京大学 Network function deployment method considering multidimensional resource constraints
CN109831319A (en) * 2018-12-27 2019-05-31 南京大学 A kind of network function dispositions method considering multi dimensional resource constraint
CN113348651A (en) * 2019-01-24 2021-09-03 威睿公司 Dynamic inter-cloud placement of sliced virtual network functions
CN113348651B (en) * 2019-01-24 2023-06-09 威睿公司 Dynamic inter-cloud placement of sliced virtual network functions
CN109617811A (en) * 2019-01-29 2019-04-12 电子科技大学 The quick migration method of mobile application in a kind of cloud network
CN110018832A (en) * 2019-02-20 2019-07-16 中国人民解放军海军工程大学 A kind of Radar Software deployment of components strategy based on improvement Dynamic Programming
CN110018832B (en) * 2019-02-20 2022-07-12 中国人民解放军海军工程大学 Radar software component deployment strategy based on improved dynamic programming
WO2020181496A1 (en) * 2019-03-12 2020-09-17 Nokia Shanghai Bell Co., Ltd. Method, device and computer readable medium for service chain
CN110022230A (en) * 2019-03-14 2019-07-16 北京邮电大学 The parallel dispositions method of service chaining and device based on deeply study
CN109842528A (en) * 2019-03-19 2019-06-04 西安交通大学 A kind of dispositions method of the service function chain based on SDN and NFV
CN110446121A (en) * 2019-04-07 2019-11-12 西安电子科技大学 Virtual network function service chaining mapping method based on betweenness center degree
CN110446121B (en) * 2019-04-07 2021-05-18 西安电子科技大学 Virtual network function service chain mapping method based on betweenness centrality
CN110049130A (en) * 2019-04-22 2019-07-23 北京邮电大学 A kind of service arrangement and method for scheduling task and device based on edge calculations
CN110049130B (en) * 2019-04-22 2020-07-24 北京邮电大学 Service deployment and task scheduling method and device based on edge computing
CN110190987A (en) * 2019-05-08 2019-08-30 南京邮电大学 Based on backup income and the virtual network function reliability dispositions method remapped
CN110190987B (en) * 2019-05-08 2022-02-01 南京邮电大学 Virtual network function reliability deployment method based on backup income and remapping
US11588733B2 (en) 2019-05-14 2023-02-21 Vmware, Inc. Slice-based routing
US11902080B2 (en) 2019-05-14 2024-02-13 Vmware, Inc. Congestion avoidance in a slice-based network
US11595315B2 (en) 2019-05-14 2023-02-28 Vmware, Inc. Quality of service in virtual service networks
CN110120978A (en) * 2019-05-17 2019-08-13 电子科技大学 A kind of method for security protection of elasticity user's cloud computing resources
CN110298381B (en) * 2019-05-24 2022-09-20 中山大学 Cloud security service function tree network intrusion detection system
CN110298381A (en) * 2019-05-24 2019-10-01 中山大学 A kind of cloud security service functional tree Network Intrusion Detection System
CN110311862A (en) * 2019-06-14 2019-10-08 北京邮电大学 A kind of service chaining mapping method and device
WO2020255025A1 (en) * 2019-06-18 2020-12-24 Telefonaktiebolaget Lm Ericsson (Publ) Method and system for deploying a new service function chain (sfc) based on similarities with previously deployed sfcs
CN110311816A (en) * 2019-06-28 2019-10-08 上海交通大学 The VNF laying method of the co-located interference perception of VNF under a kind of NFV environment
CN110311816B (en) * 2019-06-28 2021-08-13 上海交通大学 VNF (virtual network function) placement method for VNF co-location interference sensing in NFV (network virtual function) environment
CN110380898A (en) * 2019-07-08 2019-10-25 紫光云技术有限公司 Multi-operator edge service chain layout strategy based on Hyperledger Fabric
CN110380898B (en) * 2019-07-08 2022-11-15 紫光云技术有限公司 Hyperhedger Fabric-based multi-operator edge service chain arrangement strategy
CN110545228A (en) * 2019-07-26 2019-12-06 北京邮电大学 service function chain request processing method and system
CN110545228B (en) * 2019-07-26 2020-11-20 北京邮电大学 Service function chain request processing method and system
CN110535705B (en) * 2019-08-30 2022-04-26 西安邮电大学 Service function chain construction method capable of adapting to user time delay requirement
CN110535705A (en) * 2019-08-30 2019-12-03 西安邮电大学 A kind of service function chain building method of adaptive user delay requirement
CN110851235B (en) * 2019-11-04 2022-06-03 中国人民解放军战略支援部队信息工程大学 Virtual network function deployment method suitable for multidimensional resource optimization configuration
CN110851235A (en) * 2019-11-04 2020-02-28 中国人民解放军战略支援部队信息工程大学 Virtual network function deployment method suitable for multidimensional resource optimization configuration
CN111385202A (en) * 2020-03-17 2020-07-07 重庆邮电大学 Route distribution method based on virtual network function
CN111385202B (en) * 2020-03-17 2022-03-11 重庆邮电大学 Route distribution method based on virtual network function
CN111526070A (en) * 2020-04-29 2020-08-11 重庆邮电大学 Service function chain fault detection method based on prediction
CN111526070B (en) * 2020-04-29 2022-06-03 重庆邮电大学 Service function chain fault detection method based on prediction
CN111800291B (en) * 2020-05-27 2021-07-20 北京邮电大学 Service function chain deployment method and device
CN111800291A (en) * 2020-05-27 2020-10-20 北京邮电大学 Service function chain deployment method and device
CN112083933B (en) * 2020-08-27 2023-07-04 重庆大学 Service function chain deployment method based on reinforcement learning
CN112083933A (en) * 2020-08-27 2020-12-15 重庆大学 Service function chain deployment method based on reinforcement learning
CN113032096B (en) * 2021-03-17 2023-03-31 西安电子科技大学 SFC mapping method based on node importance and user demand dual perception
CN113032096A (en) * 2021-03-17 2021-06-25 西安电子科技大学 SFC mapping method based on node importance and user demand dual perception
CN113225211A (en) * 2021-04-27 2021-08-06 中国人民解放军空军工程大学 Fine-grained service function chain extension method
CN113225211B (en) * 2021-04-27 2022-09-02 中国人民解放军空军工程大学 Fine-grained service function chain extension method
CN113535388B (en) * 2021-06-25 2023-09-29 广西综合交通大数据研究院 Task-oriented service function aggregation method
CN113535388A (en) * 2021-06-25 2021-10-22 广西综合交通大数据研究院 Task-oriented service function aggregation method
CN114205317A (en) * 2021-10-21 2022-03-18 北京邮电大学 Service function chain SFC resource allocation method based on SDN and NFV and electronic equipment
CN114205317B (en) * 2021-10-21 2023-07-21 北京邮电大学 SDN and NFV-based service function chain SFC resource allocation method and electronic equipment
CN114124796A (en) * 2021-10-26 2022-03-01 清华大学 Service chain deployment processing method and device, computer storage medium and terminal
CN114650225A (en) * 2022-03-14 2022-06-21 西安电子科技大学 Service function chain SFC mapping method based on particle swarm optimization algorithm

Also Published As

Publication number Publication date
CN107332913B (en) 2020-03-27

Similar Documents

Publication Publication Date Title
CN107332913A (en) A kind of Optimization deployment method of service function chain in 5G mobile networks
CN104702522B (en) Computer implemented method, device, the controller of software defined network routing data
CN103827825B (en) Virtual resource object component
Wang et al. Algorithms for a facility location problem with stochastic customer demand and immobile servers
CN104363159B (en) A kind of opening virtual network constructing system and method based on software defined network
CN103369027B (en) Location aware Virtual Service in mixing cloud environment is equipped with
CN104798356B (en) Method and apparatus for the utilization rate in controlled level expanding software application
CN107710237A (en) Deep neural network divides on server
CN104503826B (en) The virtual machine mapping method and device of a kind of cloud computation data center
CN108809857A (en) A method of the traffic monitoring based on SDN and service quality securing strategy
CN103475947B (en) A kind of flow media distribution system based on software defined network and method
CN105337834B (en) A kind of mapping method under wireless network virtualized environment
CN107836001A (en) Convolutional neural networks on hardware accelerator
CN104202183B (en) The method and apparatus that a kind of solution SDN stream ranks configuration conformance updates
CN104125092A (en) Matching method and device of network resource
CN106533806A (en) Method for providing cross-layer quality of service (QoS) based on application awareness in multi-tenant software defined network (SDN)
CN105282191B (en) SiteServer LBS, controller and method
CN106537824B (en) Method and apparatus for the response time for reducing information centre's network
CN114090244B (en) Service arrangement method, device, system and storage medium
CN102868542B (en) The control method and system of service quality in a kind of service delivery network
CN105791151B (en) A kind of dynamic flow control method and device
CN107454019A (en) Software defined network distribution method of dynamic bandwidth, device, equipment and storage medium
CN110087250A (en) A kind of network slice layout scheme and its method based on multiple target combined optimization model
Zhang et al. DSCD: Delay sensitive cross-domain virtual network embedding algorithm
Dräxler et al. Specification, composition, and placement of network services with flexible structures

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant