CN104243258B - A kind of mapping method of virtual network and system based on classification - Google Patents
A kind of mapping method of virtual network and system based on classification Download PDFInfo
- Publication number
- CN104243258B CN104243258B CN201310247047.3A CN201310247047A CN104243258B CN 104243258 B CN104243258 B CN 104243258B CN 201310247047 A CN201310247047 A CN 201310247047A CN 104243258 B CN104243258 B CN 104243258B
- Authority
- CN
- China
- Prior art keywords
- node
- virtual
- equally loaded
- physical
- demand
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Expired - Fee Related
Links
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
- H04L41/12—Discovery or management of network topologies
- H04L41/122—Discovery or management of network topologies of virtualised topologies, e.g. software-defined networks [SDN] or network function virtualisation [NFV]
Landscapes
- Engineering & Computer Science (AREA)
- Computer Networks & Wireless Communication (AREA)
- Signal Processing (AREA)
- Data Exchanges In Wide-Area Networks (AREA)
Abstract
The invention provides a kind of mapping method of virtual network based on classification and system, methods described includes：According to the network topological information in each region, dummy node in virtual network requests is classified, according to classification results, by the virtual network segmentation of request, less virtual net string bag is asked on a large scale, is asked the management entity for being dispatched to each region to carry out parallel mapping virtual net string bag and is handled；The management entity in each region is when maps virtual network is asked, introduce resources balance mechanism, the demand limitation asked respectively virtual net string bag and the available resources of region bottom physical network carry out equilibrium treatment, while coordinating to integrate the mapping of dummy node and the mapping of virtual link；After the management entity in all regions all completes mapping, the mapping result of the management entity in each region is integrated by global management entity and user is fed back to.Compared with prior art, the features such as present invention has low computation complexity, fast mapping and more preferable bottom physical network load balancing.
Description
Technical field
The present invention relates to network virtualization technology, specifically, it is related to a kind of mapping method of virtual network based on classification
And system.
Background technology
Network virtualization technology is the core technology of Future Internet, and it allows in a shared bottom-layer network resource
Multiple virtual networks are supported, there can be the different network architectures between these virtual networks, different types of clothes can also be carried
Business.
One key issue of network virtualization technology is virtual network mapping problems.By by virtual network requests
Dummy node and virtual link are mapped in bottom-layer network resource, to meet the node and link requirements of virtual network requests.It is empty
Intending network mapping problem is proved to be very scabrous technical problem, and its computation complexity is higher.Existing method is not proposed
Effective measures are to reduce computation complexity so that mapping processing time is longer, and the utilization rate of bottom-layer network resource is relatively low.
The content of the invention
It is an object of the present invention to which in order to overcome above mentioned problem, the invention provides a kind of virtual network based on classification
Mapping method and system.
To achieve these goals, the invention provides a kind of mapping method of virtual network based on classification, described side
Method is included：
Step 101）When receiving virtual network requests, to the dummy node involved by virtual network requests according to contingency table
Standard is classified, and the dummy node belonged to a different category is split into different virtual net string bag requests；
Wherein, described criteria for classification is that dummy node is categorized into by the demand information in geographical position according to dummy node
In bottom physical network from its nearest region；All dummy nodes in same virtual net string bag request belong to same point
Class, the classification is the classification belonging to any one dummy node in virtual net string bag request；
Step 102）Each virtual net string bag is asked to send some region into the bottom physical network belonging to it；
Step 103）The equally loaded demand of each node and each node in region in being asked according to virtual net string bag
Equally loaded ability, the request of virtual net string bag is mapped in region on bottom physical layer network；Wherein,
The equally loaded requirement representation is in the case of load balancing, the loading demand of each dummy node；It is described negative
It is the resource of dummy node request and the product of the resource for all link requests for being connected to the dummy node to carry requirement definition；
The equally loaded ability represented in the case of load balancing, the load capacity of each physical node in region；
The load capacity is defined as the available resources of physical node and is connected to the available resources of all links of the physical node
Product.
Geographical position demand when classification analysis is carried out to the dummy node in virtual network requests according to dummy node is believed
Cease and be categorized into dummy node in the bottom physical network in the region nearest from the geographical position that it is required.
Above-mentioned steps 103）Further include：
Step 103-1）Detect current unmapped with the virtual of maximum equally loaded demand in the request of virtual net string bag
Node nv；
Step 103-2）Dummy node n is found in this regionvK candidate's bottom physical node, k described candidate
Node is currently also unmapped in the region and can meet the dummy node nvDemand first k maximum equally loaded
The bottom physical node of ability；
Step 103-3）To each candidate's bottom physical node, calculate candidate's bottom physical node and mapped each
The path length sum between bottom physical node corresponding to dummy node, by described dummy node nvIt is mapped to most
In the both candidate nodes of short path length, and the virtual link between the dummy node that has mapped is mapped.
Above-mentioned equally loaded ability is used and is obtained as below：
Step 201）The physical node of the regional included to bottom physical network and the resource of physical link are entered respectively
Row normalized；
Wherein, the normalized of physical node and the memory capacity of node, magnetic disc i/o, CPU quantity and memory size phase
Close；The normalized of physical link is related to the bandwidth of link, delay and delay jitter；
Step 202）Calculate the load capacity of the node in bottom physical network in regional；
Wherein, the load capacity is defined as the available resources of physical node and is connected to all chains of the physical node
The product of the available resources on road；
Step 203）Calculate the equally loaded matrix of regional in bottom physical network；
Wherein, each element in the equally loaded matrix is：The load between each physical node in some region
The weighted value of the mutual contribution of the ability of equalization, the weighted value of the mutual contribution of the load balance ability is specially：When physics section
Point nsIt is exactly physical node msItself, i.e. ns=ms, or physical node nsWith physical node msBetween when having physical link connection, thing
Manage node nsTo physical node msEqually loaded ability contribute, now weighted value be a non-zero positive number；Otherwise, physics section
Point nsTo physical node msEqually loaded ability contribution weight be 0；nsThe n-th of representing matrixsOK, msThe m of representing matrixs
Row, the line number and columns of matrix are all the total numbers of physical node, and what footmark s was represented is bottom physical network；
Step 204）According to equally loaded matrix, the equally loaded ability of each physical node in region is iterated to calculate out,
That is physical node nsLoad balance ability for the equally loaded ability of all physical nodes in the region contribution sum.
Above-mentioned equally loaded demand is adopted to be obtained with the following method：
Step 301）The demand of dummy node and virtual link in asking each virtual net string bag carries out normalizing respectively
Change is handled；
Wherein, the normalized of dummy node and the memory capacity of node, magnetic disc i/o, CPU quantity and memory size are needed
Ask related；The normalized of virtual link is related to the bandwidth of link, delay and delay jitter demand；
Step 302）Calculate the loading demand of each dummy node in the request of each virtual net string bag；
Wherein, the loading demand is defined as the demand of dummy node and is connected to all links of the dummy node
The product of demand；
Step 303）Calculate the equally loaded matrix of each virtual net string bag request；
Wherein, each element in the equally loaded matrix is：Between each dummy node in the request of virtual net string bag
Equally loaded demand interactional weighted value, the weighted value is specially：As dummy node nvIt is exactly dummy node mvThis
Body, i.e. nv=mv, or dummy node nvWith dummy node mvBetween when having virtual link connection, it is a non-zero to define weighted value
Numerical value；Otherwise nvTo mvEqually loaded demand weighing factor be 0；
nvThe n-th of representing matrixvOK, mvThe m of representing matrixvRow, the line number and columns of matrix are all the total of dummy node
Number, what footmark v was represented is virtual network requests；
Step 304）According to equally loaded matrix, the equally loaded demand of each dummy node in region is iterated to calculate out,
That is dummy node nvLoad balancing demand for the virtual net string bag ask in all dummy nodes equally loaded demand shadow
Ring sum.
In order to realize the above method, the invention provides a kind of virtual network mapped system based on classification, described is
System is included：
Global administration's module, for receiving during virtual network requests, to the dummy node involved by virtual network requests
Classified according to criteria for classification, the dummy node belonged to a different category is split into different virtual net string bag requests, institute
The criteria for classification stated refers to the classification where the region that a dummy node will belong to nearest from it in bottom physical network；And will
The request of each virtual subnetwork sends some region into the bottom physical network belonging to it；With
District management module, for according to each in the equally loaded demand of each node in the request of virtual net string bag and region
The equally loaded ability of individual node, the request of virtual net string bag is mapped in region on bottom physical layer network；
Wherein, the equally loaded requirement representation is in the case of load balancing, the loading demand of each dummy node；Institute
Loading demand is stated to be defined as the resource of dummy node request and be connected to the resource of all link requests of the dummy node
Product；
The equally loaded ability represented in the case of load balancing, the load capacity of each physical node in region；
The load capacity is defined as the available resources of physical node and is connected to the available resources of all links of the physical node
Product.
Geographical position demand when classification analysis is carried out to the dummy node in virtual network requests according to dummy node is believed
Cease and be categorized into dummy node in the region bottom physical network nearest from the geographical position that it is required.
Above-mentioned zone management module is further included：
Equally loaded ability or equally loaded demand obtain submodule, are respectively saved in each region of bottom physical network for obtaining
The equally loaded ability value of point and the equally loaded requirements for obtaining each dummy node that each virtual net string bag is asked；
Detection sub-module, current in the request of virtual net string bag unmapped has maximum equally loaded demand for detecting
Dummy node nv；
Candidate's bottom physical node selects submodule, for finding dummy node n in some regionvK candidate bottom
Layer physical node, k described both candidate nodes are currently also unmapped in the region and can meet the dummy node nv
Demand first k maximum equally loaded ability bottom physical node；With
Node and link maps submodule, for each candidate's bottom physical node, calculating candidate's bottom physics section
The path length sum between bottom physical node corresponding to point and each dummy node mapped, by described virtual section
Point nvIt is mapped in the both candidate nodes with shortest path length, and the virtual link between the dummy node that has mapped is entered
Row mapping.
Above-mentioned equally loaded ability obtains submodule and further included：
First normalized module, for the physical node and physics chain of the regional included to bottom physical network
The resource on road is normalized respectively；
Wherein, the normalized of physical node and the memory capacity of node, magnetic disc i/o, CPU quantity and memory size phase
Close；The normalized of physical link is related to the bandwidth of link, delay and delay jitter；
First node load capacity acquisition module, the load for calculating the node in bottom physical network in regional
Ability；
Wherein, the load capacity is defined as the available resources of physical node and is connected to all chains of the physical node
The product of the available resources on road；
Equally loaded matrix acquisition module, the equally loaded matrix for calculating regional in bottom physical network；
Wherein, the equally loaded matrix defines equal between each physical node in the region of bottom physical network
The mutual contribution weighted value for the load capacity that weighs, and if only if physical node nsIt is exactly physical node msItself, i.e. ns=ms, or nsWith
msBetween have physical link connection, now weighted value be a non-zero positive number；Otherwise physical node nsTo physical node msEquilibrium
The contribution weight of load capacity is 0；With
Equally loaded capacity calculation module, for according to equally loaded matrix, iterating to calculate out each physics section in region
The equally loaded ability of point, i.e. physical node nsLoad balance ability for all physical nodes in the region equally loaded energy
The contribution sum of power.
Above-mentioned equally loaded demand obtains submodule and further included：
Second normalized module, for the dummy node and the need of virtual link in the request of each virtual net string bag
Ask and be normalized respectively；
Wherein, the normalized of dummy node and the memory capacity of node, magnetic disc i/o, CPU quantity and memory size are needed
Ask related；The normalized of virtual link is related to the bandwidth of link, delay and delay jitter demand；
Second loading demand computing module, the load for calculating each dummy node in the request of each virtual net string bag is needed
Ask；
Wherein, the loading demand is defined as the demand of dummy node and is connected to all links of the dummy node
The product of demand；
Second equally loaded matrix, the equally loaded matrix for calculating the request of each virtual net string bag；
Wherein, the equilibrium that the equally loaded matrix is defined between each dummy node in the request of virtual net string bag is born
The interactional weighted value of load demand, and if only if dummy node nvIt is exactly dummy node mvItself, i.e. nv=mv, or virtually
Node nvWith dummy node mvBetween have virtual link connection, weighted value now is the positive number of a non-zero；Otherwise dummy node nv
To dummy node mvEqually loaded demand weighing factor be 0；With
Equally loaded demand acquisition module, for according to equally loaded matrix, iterating to calculate out each virtual section in region
The equally loaded demand of point, i.e. dummy node nvLoad balancing demand asked for the virtual net string bag in all dummy nodes
The influence sum of equally loaded demand.
In a word, the present invention provides a kind of mapping method of virtual network based on classification, and described method includes：According to each area
The network topological information in domain, classification analysis is carried out to the dummy node in virtual network requests, according to classification results by request
Virtual network segmentation less virtual net string bag request on a large scale, virtual net string bag is asked to be dispatched to the management entity in each region
Carry out parallel mapping processing；The management entity in each region introduces resources balance mechanism when maps virtual network is asked,
The demand limitation asked respectively virtual net string bag and the available resources of region bottom physical network carry out equilibrium treatment, assist simultaneously
The mapping of dummy node and the mapping of virtual link are closed in adjustment；After the management entity in all regions all completes mapping, by complete
The management entity of office integrates the mapping result of the management entity in each region and feeds back to user.Described mapping method of virtual network
Handle the mapping of virtual network requests in a distributed manner on global management entity, centralization is located on the management entity in region
Manage virtual network requests mapping.Described classification is main to utilize the demands such as the geographical position of each dummy node in virtual network requests
Restricted information, calculates the distance between each dummy node and the management entity in each region, and carry out according to described distance
Classification analysis is to determine the region belonging to dummy node.The dummy node for belonging to same category is divided into identical by described segmentation
The request of virtual net string bag in, the virtual link between the request of each virtual net string bag is reduced to a virtual link.Described
Mapping method of virtual network introduces resources balance machine when the management entity in each region carries out the mapping of virtual net string bag request
System.Described mapping method of virtual network is coordinated whole when the management entity in each region carries out the mapping of virtual net string bag request
The mapping of dummy node and the mapping of virtual link are closed.Described resources balance mechanism is born by the resource of aggregators itself
Carry and influence of the resource load to the node of its neighbor node is realized.The load capacity of described node and the resource of node
It is related to the link circuit resource for being connected to the node.
Compared with prior art, the technical advantages of the present invention are that：
By above-mentioned technical proposal, virtual network requests will be divided into the less sub- request of scale, and by regional
Management entity is concomitantly mapped, so as to reduce computation complexity, reduces mapping time.The management entity of regional is reflecting
When penetrating the request of virtual net string bag, by introducing resources balance mechanism, to realize the load balancing of bottom-layer network resource.Dummy node
The coordination that mapping and virtual link map, which is integrated, avoids the widely distributed of node mapping, reduces the bottom physics of link maps
Path length.
Brief description of the drawings
Fig. 1 is the workflow of global management entity in the mapping method of virtual network based on classification that the present invention is provided
Figure；
Fig. 2 is the workflow of district management entity in the mapping method of virtual network based on classification that the present invention is provided
Figure；
Fig. 3 is the bottom physical network framework of the embodiment for the mapping method of virtual network based on classification that the present invention is provided
Schematic diagram；
Fig. 4 is the void of global management entity in the embodiment of the mapping method of virtual network based on classification that the present invention is provided
Intend the schematic diagram of network mapping；
Fig. 5 is that the load-balancing mechanism in the embodiment for the mapping method of virtual network based on classification that the present invention is provided shows
It is intended to；Numeral in the figure in the small square frame of each near nodal represents normalized node resource, by memory capacity, magnetic disc i/o,
CPU, the influence such as internal memory；Numeral in the figure on lines represents normalized link circuit resource, is influenceed by bandwidth, delay, shake；
Fig. 6-1 is virtual net string bag request schematic diagram provided in an embodiment of the present invention；
Fig. 6-2 is the schematic diagram before the bottom physical network mapping in region provided in an embodiment of the present invention；
Fig. 6-3 is the schematic diagram after the mapping of bottom physical network in region provided in an embodiment of the present invention.
Embodiment
Technical scheme is described in detail below in conjunction with the accompanying drawings.
A kind of mapping method of virtual network based on classification of the present invention includes：
According to the network topological information in each region, classification analysis is carried out to the dummy node in virtual network requests, according to
Classification results by the virtual network segmentation of request, asked on a large scale by less virtual net string bag, and the request of virtual net string bag is dispatched to
The management entity in each region carries out parallel mapping processing；The management entity in each region draws when maps virtual network is asked
Resources balance mechanism is entered, the demand limitation asked respectively virtual net string bag and the available resources of region bottom physical network are entered
Row equilibrium treatment, while coordinating to integrate the mapping of dummy node and the mapping of virtual link；All regions management entity all
Complete after mapping, the mapping result of the management entity in each region is integrated by global management entity and user is fed back to.
In order to clearly illustrate present disclosure, the present invention is described as follows to virtual network mapping：
A. bottom physical network
It is a undirected set of graphs G to define bottom physical networks(Ns,Es), the ith zone of bottom physical network is represented
For non-directed graphWhereinThe node set in the i of region is represented,Represent the link set in the i of region.Definition section
Point nsNeighbor node collection be combined into Ns(ns), definition is connected to node nsLink set be Es(ns).In bottom physical network
Node and link are all associated with their resource constraint.We are defined in bottom physical network, node resource restricted information bag
Include geographical position Ls, memory capacity Ss, magnetic disc i/o speed Is, CPU abilities CsWith memory size Ms, link circuit resource, which is limited, includes band
Wide Bs, delay DsWith delay jitter Js（Such as, physical node nsNode resource restricted information：Geographical position represents position coordinates Ls
(ns)=(10,20), memory capacity Ss(ns)=500MB, magnetic disc i/o speed Is(ns)=100bps, CPU abilities Cs(ns)=4
Core and memory size Ms(ns)=2GB, link circuit resource limitation：Bandwidth Bs(ns)=100MB, delay Ds(ns)=200ms and delay
Shake Js(ns)=20ms）.Table 1 has itemized out the resource constraint of midsole layer physical network of the present invention, and is each resource limit
The adopted weights influence factor is formulated, with calculate node and the normalization resource constraint of link.Therefore, bottom physical network nodes ns's
Normalizing resource representation is：Link es's
Normalizing resource representation is：Wherein geographical position is mainly used in
Classification analysis, without considering in the normalization of node resource is represented.
The resource constraint of the bottom physical network of table 1
B. virtual network requests
Virtual networks request of the present invention is a non-directed graph Gv(Nv,Ev), i-th of son request of virtual network requests
It is expressed as non-directed graphWhereinThe node set of i-th of virtual net string bag request is represented,Represent i-th of void
Intend the link set of network request.Definition node nvNeighbor node collection be combined into Nv(nv), definition is connected to node nvLink
Collection is combined into Ev(nv).Node and link in virtual network requests is associated all with their demand limitation.We define virtual net
In network request, the demand restricted information that dummy node is proposed includes：Geographical position Lv, memory capacity Sv, magnetic disc i/o speed Iv、
CPU quantity CvWith memory size Mv, the demand of virtual link, which is limited, includes bandwidth Bv, delay DvWith delay jitter Jv（Such as, virtual section
Point nvNode resource restricted information：Geographical position represents position coordinates Lv(nv)=(30,20), memory capacity Sv(nv)=5MB,
Magnetic disc i/o speed Iv(nv)=50bps, CPU abilities Cv(nvThe core of)=2 and memory size Mv(nv)=200MB, link circuit resource limit
System：Bandwidth Bv(nv)=10MB, delay Dv(nv)=200ms and delay jitter Jv(nv)=20ms）.Table 2 is virtual in the present invention
The weights influence factor and the power of resource constraint in bottom physical network of the demand limitation of network request, the wherein limitation of each demand
Weight factor of influence is identical.Therefore, dummy node n in virtual network requestsvNormalization resource representation be：Virtual link evNormalization resource representation
For：Wherein geographical position is mainly used in classification analysis, without examining
Consider in the normalization of node resource demand is represented.
The demand limitation of the virtual network requests of table 2
C. the place-centric of region bottom physical network
In order to which preferably virtual network requests are mapped in bottom physical network, the present invention makes full use of virtual network please
The topology information asked, limits according to the geographical position demand of dummy node, the dummy node in virtual network requests is divided
Alanysis, so as to which dummy node is mapped in the region bottom physical network nearest from it.Define each in bottom physical network
The place-centric collection in individual region is combined into Os, whereinFor region i place-centric, it is defined as all nodes in the i of region
Geometric center, i.e.,
Wherein,Represent the node number in the i of region.
D. load-balancing mechanism
In order to improve the resource utilization and load balancing of bottom physical network, the present invention is abundant using load-balancing mechanism
Utilize the resource and the resource of its neighbor node of node itself.In bottom physical network, load capacity and the node of node are provided
Source is relevant with the link circuit resource for being connected to node, and link circuit resource is shared by its two end points.Therefore, bottom thing is defined
Manage nodes nsLoad capacity Rs(ns) as follows：
On this basis, bottom physical network interior joint n is definedsEqually loaded abilityIt is as follows：
That is, the neighbor node of the equally loaded ability of the equally loaded ability and node of node itself and node is balanced negative
Loading capability is relevant, wherein w1,w2It is the related weights influence factor, meets w1+w2=1,For node
msTo node nsInfluence.Thus, definable region i equally loaded matrixIt is as follows：
The equally loaded ability of region i interior nodes is expressed as follows：
Similarly, the equally loaded demand of the node in definable virtual network requestsIt is wherein virtual
The loading demand R of dummy node in network requestv(nv) meet：
Dummy node nvEqually loaded demandMeet：
That is, the equally loaded of itself of the equally loaded demand and node of dummy node needs the equal of the neighbor node of summing junction
The loading demand that weighs is relevant, wherein,For node mvTo node nvInfluence.
The equally loaded matrix A of virtual network requestsvIt is defined as follows：
Thus, i-th of virtual net string bag of definable asks the equally loaded demand of interior joint
In order to more clearly illustrate the object, technical solutions and advantages of the present invention, below in conjunction with the embodiment of the present invention
Accompanying drawing, the technical scheme in the embodiment of the present invention is clearly and completely described, it is clear that described embodiment is this
A part of embodiment of invention, rather than whole embodiments.
As shown in figure 1, the present invention provides a kind of work of global management entity in mapping method of virtual network based on classification
Make flow as follows：1）During beginning, global management entity waits the arrival of virtual network requests；2）When virtual network requests are reached
When, described global management entity carries out classification analysis to the dummy node in virtual network requests, determines each dummy node institute
The classification of category and region；3）According to described classification results, virtual network requests are divided into multiple virtual subnets and asked；4）By institute
The management entity that each virtual net string bag stated requests assignment to regional is handled respectively；5）When the management of regional
When entity all handles completion, merge the result of the management entity of regional and feed back to user.
When the dummy node in virtual network requests carries out classification analysis, the main geographical position for considering dummy node
Etc. demand limitation, dummy node is categorized into the region bottom physical network nearest from the geographical position that it is required, therefore, can
Define dummy node nvAffiliated classification
According to the classification results of virtual network requests, the dummy node for belonging to different classifications is divided into different virtual nets
In string bag request, and simplify the link between each height request.Belong to i-th of virtual net string bag requestIt is virtual
Node nvMeet φ (nv)=i, the virtual link between all leap virtual net string bag request i and virtual net string bag request j is simple
A link e (i, j) is turned to, it meets condition：That is the virtual link of the simplification
Normalization demand be across virtual net string bag request i and virtual net string bag request j all link normalization demands in most
Big value, if the result of virtual network mapping can meet the demand of simplified link e (i, j), then for being asked across virtual net string bag
I and virtual net string bag is asked to ask j all virtual link demands to be all met.Virtual network requests are divided into multiple void
Intend after network request, the request of each virtual net string bag is assigned in suitable region bottom physical network, for virtual net
String bag asks i, i.e.,It will be assigned in i-th of bottom physical network and carry out further mapping processing.
Fig. 3 is a kind of bottom physical network framework signal of mapping method of virtual network embodiment based on classification of the present invention
Figure, as shown in figure 3, the bottom physical network in the present embodiment is double-layer structure, i.e. management level and service layer.Wherein, management level
Including global management entity and district management entity.Nodal information and Link State in the district management entity assembles region,
And the place-centric in the region and the logical topology of regional nodes are calculated, then send the result for updating calculating to global pipe
Manage entity.Global management entity directly and district management entity communication, the last state of the whole physical network of holding.Wherein, area
The information of each node and link in the management entity periodic collection region of domain, only when the network state in region changes,
District management entity actively transmits fresh information to global management entity, to reduce communication overhead.In addition, global management entity one
As in the case of belong to passive receiving area network state information, but still can active interrogation management entity, and ask for related letter
Breath.
Fig. 4 is the virtual of global management entity in a kind of embodiment of the mapping method of virtual network based on classification of the present invention
The schematic diagram of network mapping.As shown in figure 4, when virtual network requests VNR is reached, global management entity is first to virtual network
The topology of request carries out classification analysis, and the virtual network requests are divided into the less virtual net of scale according to the result of classification
String bag asks VNRsub1And VNRsub2.They are assigned to suitable region by the classification according to belonging to asking each virtual net string bag
Parallel centralized virtual network mapping is carried out in bottom physics.As shown in figure 4, virtual net string bag asks VNRsub1It is allocated
Mapped to region 1, virtual net string bag request VNRsub2Region 3 is assigned to be mapped.
As shown in Fig. 2 the present invention provides a kind of work of district management entity in mapping method of virtual network based on classification
Make flow as follows：1）During beginning, the equally loaded energy of each node in the bottom-layer network node in district management entity zoning
Power；2）Wait the arrival of virtual net string bag request；3）When virtual net string bag asks to reach, calculate each in the request of virtual net string bag
The equally loaded demand of dummy node；4）On the bottom physical network that the request of virtual net string bag is mapped in region.
The equally loaded of the equally loaded ability of bottom physical network nodes and the request of virtual net string bag in zoning
During demand, it will be calculated using the equally loaded mechanism being previously mentioned.For the bottom physical network nodes in the i of region, its
Equally loaded ability is expressed asIts pseudo-code of the algorithm is as follows：
The present invention calculates the equally loaded demand that i-th of virtual net string bag is asked with same algorithm
Fig. 5 is the load balancing machine in a kind of embodiment for mapping method of virtual network based on classification that the present invention is provided
Schematic diagram processed, the present invention calculates equally loaded demand and equilibrium respectively to bottom physical network in the request of virtual net string bag and region
Load capacity.As shown in figure 5, the present invention calculates equally loaded ability to bottom physical network with load-balancing mechanism, to void
The calculating for intending the equally loaded demand of network request is similar.The calculating of equally loaded demand and equally loaded ability is relied on
In the equally loaded matrix of each network topology（As shown in table 4）With the load capacity of each node（As shown in table 3）.From such as table
The node B that 5 result of calculation can be seen that in the region has maximum equally loaded ability, and node G has minimum load
The ability of equalization.
Table 3, the load capacity of each node
| A | B | C | D | E | F | G |
| 175 | 387 | 85 | 247.5 | 90 | 160 | 72 |
Table 4, equally loaded matrix
| A | B | C | D | E | F | G | |
| A | 0.70 | 0.17 | 0 | 0 | 0 | 0.09 | 0.04 |
| B | 0.09 | 0.70 | 0.04 | 0.12 | 0.05 | 0 | 0 |
| C | 0 | 0.16 | 0.70 | 0.14 | 0 | 0 | 0 |
| D | 0 | 0.19 | 0.06 | 0.70 | 0.06 | 0 | 0 |
| E | 0 | 0.12 | 0 | 0.11 | 0.70 | 0.07 | 0 |
| F | 0.16 | 0 | 0 | 0 | 0.08 | 0.70 | 0.06 |
| G | 0.16 | 0 | 0 | 0 | 0 | 0.14 | 0.70 |
Table 5, the load balance ability of each node
| A | B | C | D | E | F | G |
| 182.5 | 344.3 | 91.19 | 229.5 | 125.5 | 108.4 | 48.48 |
In the present invention, the equally loaded of each node needs range of summation during district management entity is asked according to virtual net string bag
The equally loaded ability of each node in interior bottom physical network, region is mapped to using Greedy strategy by the request of virtual net string bag
On interior bottom physical network.The present invention is during mapping, and the mapping of dummy node and virtual link is to carry out simultaneously, i.e.,
When mapping dummy node, with reference to the demand characteristic of virtual link to avoid opening for longer path in bottom physical network
Pin.In mapping, the current unmapped virtual section with maximum equally loaded demand in the request of virtual net string bag is detected first
Point nv, dummy node n is then found in this regionvK candidate's bottom physical node, k described both candidate nodes are the areas
It is currently also unmapped and the dummy node n can be met in domainvDemand first k maximum equally loaded ability bottom
Physical node.To each candidate's bottom physical node, candidate's bottom physical node and each dummy node mapped are calculated
Path length sum between corresponding bottom physical node, by described dummy node nvIt is mapped to shortest path path length
In the both candidate nodes of degree, and the virtual link between the dummy node that has mapped is mapped.
District management entity is described as follows to the mapping algorithm false code that virtual net string bag is asked：
Fig. 6-1,6-2 and 6-3 are region pipes in a kind of embodiment of the mapping method of virtual network based on classification of the present invention
The schematic diagram of the virtual network mapping of entity is managed, the numerical value in small square frame in above-mentioned three width figure represents the normalization money of node
Source, by memory capacity, magnetic disc i/o, the influence of CPU and internal memory；Numerical value on lines represents the normalization resource of link, tape
Width, delay, the influence of shake；At the time of on wave（T1, t2 ... ...）Represent pointed node or link in the map by
The time sequencing of processing；Numerical value in icon represents equally loaded ability/equally loaded demand of node.
In the present embodiment, the number k=3 of both candidate nodes is taken.As in fig. 6-2, before mapping is started, have in the request of virtual net string bag
The node of maximum equally loaded demand is a, meanwhile, there is maximum equally loaded ability in region in bottom physical network
Node is B, wherein, node a resource requirement is 12, and node B available resources are 14, therefore node B can meet node a need
Ask, node a is mapped in node B.Now, the unmapped section with maximum equally loaded demand in the request of virtual net string bag
Point is dummy node c, and both candidate nodes of the dummy node c in bottom physical network are D, A, E（Reason is：Take both candidate nodes
Number k=3, are removed after the bottom physical node B mapped, the maximum 3 candidates section of remaining k=3 equally loaded ability
Point is D, A, E）.Wherein, tri- nodes of D, A, E to the node B mapped path length be 1, and D equally loaded energy
Power is maximum, therefore dummy node c is mapped on D, then by the virtual link between the dummy node a, c that have mapped<a,c>
It is mapped to physical link<B,D>On.Now, dummy node to be mapped is b, and dummy node b both candidate nodes are A, E, F.Its
In, both candidate nodes A to the node mapped path length sum is 4, both candidate nodes E to the path of the node mapped
Length is 2, and both candidate nodes F to the node mapped path length is 4, therefore, dummy node b is mapped into both candidate nodes
On E, to reduce the expense of longer path mapping, then again by virtual link<a,b>It is mapped to bottom physical pathway<B,E>, will
Virtual link<c,b>It is mapped to bottom physical pathway<D,E>.
In a kind of mapping method of virtual network based on classification of the present invention, when all virtual net string bag requests are all successfully reflected
During the bottom physical network being mapped in each region, whole virtual network requests mapping just successfully have mapped.Otherwise, refusal is reflected
Penetrate and resource allocation, and return result to user.
In summary, a kind of mapping method of virtual network based on classification of the present invention passes through to the void in virtual network requests
Intend node and carry out classification analysis, less virtual net string bag is asked on a large scale by the virtual network segmentation of request according to classification results
And the parallel mapping processing of the management entity progress in each region is dispatched to, so that computation complexity is reduced, when reducing mapping
Between.The management entity of regional calculates virtual network when maps virtual network is asked by introducing resources balance mechanism
Son request equally loaded demand and bottom physical network equally loaded ability come aid in mapping, it is achieved thereby that bottom-layer network
The load balancing of resource.In district management entity, the coordination that dummy node mapping and virtual link map, which is integrated, avoids section
It is widely distributed that point maps, and reduces the bottom physical path length of link maps.
It should be noted last that, the above embodiments are merely illustrative of the technical solutions of the present invention and it is unrestricted.Although ginseng
The present invention is described in detail according to embodiment, it will be understood by those within the art that, to the technical side of the present invention
Case is modified or equivalent substitution, and without departure from the spirit and scope of technical solution of the present invention, it all should cover in the present invention
Right among.
Claims (6)
1. a kind of mapping method of virtual network based on classification, described method is included：
Step 101) when receiving virtual network requests, the dummy node involved by virtual network requests is entered according to criteria for classification
Row classification, the dummy node belonged to a different category is split into different virtual net string bag requests；
Wherein, described criteria for classification is that dummy node is categorized into from it by the demand information in geographical position according to dummy node
In the bottom physical network in nearest region；All dummy nodes in same virtual net string bag request belong to same classification,
The classification is the classification belonging to any one dummy node in virtual net string bag request；
Step 102) each virtual net string bag is asked to send some region into the bottom physical network belonging to it；
Step 103) according to virtual net string bag ask in each node equally loaded demand it is balanced with the node of each in region
Load capacity, the request of virtual net string bag is mapped in region on bottom physical layer network；Wherein,
The equally loaded requirement representation is in the case of load balancing, the loading demand of each dummy node；The load is needed
Seek the resource for being defined as dummy node request and be connected to the product of the resource of all link requests of the dummy node；
The equally loaded ability represented in the case of load balancing, the load capacity of each physical node in region；It is described
Load capacity is defined as multiplying for the available resources of all links of the available resources of physical node with being connected to the physical node
Product；
The equally loaded ability is used and is obtained as below：
Step 201) physical node of regional that is included to bottom physical network and the resource of physical link return respectively
One change is handled；
Wherein, the normalized of physical node is related to the memory capacity of node, magnetic disc i/o, CPU quantity and memory size；
The normalized of physical link is related to the bandwidth of link, delay and delay jitter；
Step 202) calculate the load capacity of node in bottom physical network in regional；
Wherein, the load capacity is defined as the available resources of physical node and is connected to all links of the physical node
The product of available resources；
Step 203) calculate bottom physical network in regional equally loaded matrix；
Wherein, each element in the equally loaded matrix is：The load balancing between each physical node in some region
The weighted value of the mutual contribution of ability, the weighted value of the mutual contribution of the load balance ability is specially：As physical node ns
It is exactly physical node msItself, i.e. ns=ms, or physical node nsWith physical node msBetween when having physical link connection, physics section
Point nsTo physical node msEqually loaded ability contribute, now weighted value be a non-zero positive number；Otherwise, physical node ns
To physical node msEqually loaded ability contribution weight be 0；nsThe n-th of representing matrixsOK, msThe m of representing matrixsRow,
The line number and columns of matrix are all the total numbers of physical node, and what footmark s was represented is bottom physical network；
Step 204) according to equally loaded matrix, iterate to calculate out the equally loaded ability of each physical node in region, i.e. thing
Manage node nsLoad balance ability for the equally loaded ability of all physical nodes in the region contribution sum.
2. the mapping method of virtual network according to claim 1 based on classification, it is characterised in that the step 103) enter
One step is included：
Step 103-1) detect the current unmapped dummy node with maximum equally loaded demand in the request of virtual net string bag
nv；
Step 103-2) dummy node n is found in this regionvK candidate's bottom physical node, k described both candidate nodes
It is currently also unmapped in the region and can meets the dummy node nvDemand first k maximum equally loaded ability
Bottom physical node；
Step 103-3) to each candidate's bottom physical node, calculate candidate's bottom physical node with mapped each is virtual
The path length sum between bottom physical node corresponding to node, by described dummy node nvIt is mapped to shortest path
In the both candidate nodes of electrical path length, and the virtual link between the dummy node that has mapped is mapped.
3. the mapping method of virtual network according to claim 1 based on classification, it is characterised in that the equally loaded is needed
Ask to adopt and obtain with the following method：
Step 301) place is normalized respectively to the dummy node and the demand of virtual link in the request of each virtual net string bag
Reason；
Wherein, the normalized of dummy node and the memory capacity of node, magnetic disc i/o, CPU quantity and memory size demand phase
Close；The normalized of virtual link is related to the bandwidth of link, delay and delay jitter demand；
Step 302) calculate each virtual net string bag request in each dummy node loading demand；
Wherein, the loading demand is defined as the demand of dummy node and is connected to the demand of all links of the dummy node
Product；
Step 303) calculate the equally loaded matrix that each virtual net string bag is asked；
Wherein, each element in the equally loaded matrix is：It is equal between each dummy node in the request of virtual net string bag
Weigh the interactional weighted value of loading demand, and the weighted value is specially：As dummy node nvIt is exactly dummy node mvItself, i.e.,
nv=mv, or dummy node nvWith dummy node mvBetween when having virtual link connection, define the numerical value that weighted value is a non-zero；
Otherwise nvTo mvEqually loaded demand weighing factor be 0；
nvThe n-th of representing matrixvOK, mvThe m of representing matrixvRow, the line number and columns of matrix are all the total numbers of dummy node,
What footmark v was represented is virtual network requests；
Step 304) according to equally loaded matrix, iterate to calculate out the equally loaded demand of each dummy node in region, i.e., it is empty
Intend node nvLoad balancing demand for the virtual net string bag ask in all dummy nodes equally loaded demand influence it
With.
4. a kind of virtual network mapped system based on classification, it is characterised in that described system is included：
Global administration's module, for receiving during virtual network requests, to the dummy node involved by virtual network requests according to
Criteria for classification is classified, and the dummy node belonged to a different category is split into different virtual net string bag requests, described
Criteria for classification refers to the classification where the region that a dummy node will belong to nearest from it in bottom physical network；And by each
Virtual subnetwork request sends some region into the bottom physical network belonging to it；With
District management module, for being saved according to the equally loaded demand of each node in the request of virtual net string bag with each in region
The equally loaded ability of point, the request of virtual net string bag is mapped in region on bottom physical layer network；
Wherein, the equally loaded requirement representation is in the case of load balancing, the loading demand of each dummy node；It is described negative
It is the resource of dummy node request and the product of the resource for all link requests for being connected to the dummy node to carry requirement definition；
The equally loaded ability represented in the case of load balancing, the load capacity of each physical node in region；It is described
Load capacity is defined as multiplying for the available resources of all links of the available resources of physical node with being connected to the physical node
Product；
The equally loaded ability obtains submodule and further included：
First normalized module, for the physical node of regional that is included to bottom physical network and physical link
Resource is normalized respectively；
Wherein, the normalized of physical node is related to the memory capacity of node, magnetic disc i/o, CPU quantity and memory size；
The normalized of physical link is related to the bandwidth of link, delay and delay jitter；
First node load capacity acquisition module, the load energy for calculating the node in bottom physical network in regional
Power；
Wherein, the load capacity is defined as the available resources of physical node and is connected to all links of the physical node
The product of available resources；
Equally loaded matrix acquisition module, the equally loaded matrix for calculating regional in bottom physical network；
Wherein, the equilibrium that the equally loaded matrix is defined between each physical node in the region of bottom physical network is born
The mutual contribution weighted value of loading capability, and if only if physical node nsIt is exactly physical node msItself, i.e. ns=ms, or nsAnd msIt
Between have physical link connection, now weighted value be a non-zero positive number；Otherwise physical node nsTo physical node msEqually loaded
The contribution weight of ability is 0；With
Equally loaded capacity calculation module, for according to equally loaded matrix, iterating to calculate out each physical node in region
Equally loaded ability, i.e. physical node nsLoad balance ability be the equally loaded ability of all physical nodes in the region
Contribute sum.
5. the virtual network mapped system according to claim 4 based on classification, it is characterised in that the district management mould
Block is further included：
Equally loaded ability or equally loaded demand obtain submodule, for obtaining each node in each region of bottom physical network
Equally loaded ability value and the equally loaded requirements for obtaining each dummy node that each virtual net string bag is asked；
Detection sub-module, it is current unmapped with the virtual of maximum equally loaded demand in the request of virtual net string bag for detecting
Node nv；
Candidate's bottom physical node selects submodule, for finding dummy node n in some regionvK candidate's bottom physics
Node, k described both candidate nodes are currently also unmapped in the region and can meet the dummy node nvDemand
First k maximum equally loaded ability bottom physical node；With
Node and link maps submodule, for each candidate's bottom physical node, calculate candidate's bottom physical node with
Path length sum between the bottom physical node corresponding to each dummy node mapped, by described dummy node nv
It is mapped in the both candidate nodes with shortest path length, and the virtual link between the dummy node that has mapped is reflected
Penetrate.
6. the virtual network mapped system according to claim 5 based on classification, it is characterised in that the equally loaded is needed
Acquisition submodule is asked further to include：
Second normalized module, for the dummy node and the demand point of virtual link in the request of each virtual net string bag
It is not normalized；
Wherein, the normalized of dummy node and the memory capacity of node, magnetic disc i/o, CPU quantity and memory size demand phase
Close；The normalized of virtual link is related to the bandwidth of link, delay and delay jitter demand；
Second loading demand computing module, the loading demand for calculating each dummy node in the request of each virtual net string bag；
Wherein, the loading demand is defined as the demand of dummy node and is connected to the demand of all links of the dummy node
Product；
Second equally loaded matrix, the equally loaded matrix for calculating the request of each virtual net string bag；
Wherein, the equally loaded that the equally loaded matrix is defined between each dummy node in the request of virtual net string bag is needed
The interactional weighted value asked, and if only if dummy node nvIt is exactly dummy node mvItself, i.e. nv=mv, or dummy node nv
With dummy node mvBetween have virtual link connection, weighted value now is the positive number of a non-zero；Otherwise dummy node nvTo virtual
Node mvEqually loaded demand weighing factor be 0；With
Equally loaded demand acquisition module, for according to equally loaded matrix, iterating to calculate out each dummy node in region
Equally loaded demand, i.e. dummy node nvLoad balancing demand for the virtual net string bag ask in all dummy nodes equilibrium
The influence sum of loading demand.
Priority Applications (3)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201310247047.3A CN104243258B (en) | 2013-06-20 | 2013-06-20 | A kind of mapping method of virtual network and system based on classification |
| PCT/CN2014/080372 WO2014202016A1 (en) | 2013-06-20 | 2014-06-20 | Classification-based virtual network mapping method and system |
| SG11201510407UA SG11201510407UA (en) | 2013-06-20 | 2014-06-20 | Classification-based virtual network embedding method and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201310247047.3A CN104243258B (en) | 2013-06-20 | 2013-06-20 | A kind of mapping method of virtual network and system based on classification |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN104243258A CN104243258A (en) | 2014-12-24 |
| CN104243258B true CN104243258B (en) | 2017-10-31 |
Family
ID=52103972
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201310247047.3A Expired - Fee Related CN104243258B (en) | 2013-06-20 | 2013-06-20 | A kind of mapping method of virtual network and system based on classification |
Country Status (3)
| Country | Link |
|---|---|
| CN (1) | CN104243258B (en) |
| SG (1) | SG11201510407UA (en) |
| WO (1) | WO2014202016A1 (en) |
Families Citing this family (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108243044B (en) * | 2016-12-27 | 2021-08-24 | 山东财兜信息科技有限公司 | Service deployment method and device |
| CN110445630A (en) * | 2018-05-03 | 2019-11-12 | 中兴通讯股份有限公司 | Divide method and device, the storage medium, electronic device of network topology |
| CN112148206A (en) * | 2019-06-28 | 2020-12-29 | 北京金山云网络技术有限公司 | Data reading and writing method and device, electronic equipment and medium |
| CN112636961B (en) * | 2020-12-15 | 2022-11-08 | 国网河南省电力公司信息通信公司 | Virtual network resource allocation method based on reliability and distribution strategy under network slice |
| CN113163498B (en) * | 2021-05-26 | 2022-09-30 | 广东电网有限责任公司电力调度控制中心 | Virtual network resource allocation method and device based on genetic algorithm under 5G network slice |
| CN113630328B (en) * | 2021-07-09 | 2022-09-16 | 西安电子科技大学 | Data center virtual network mapping method and system |
| WO2023123171A1 (en) * | 2021-12-30 | 2023-07-06 | 西安电子科技大学 | Pattern complementation-based virtual network mapping method and apparatus, and computer readable storage medium |
| CN114697222B (en) * | 2022-03-22 | 2023-06-20 | 西安交通大学 | Survivability virtual network mapping algorithm based on space-time-ground fusion network |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102075402A (en) * | 2011-02-12 | 2011-05-25 | 华为技术有限公司 | Virtual network mapping processing method and system |
| CN102868733A (en) * | 2012-08-29 | 2013-01-09 | 北京邮电大学 | Method for remapping virtual network resources |
Family Cites Families (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US9106540B2 (en) * | 2009-03-30 | 2015-08-11 | Amazon Technologies, Inc. | Providing logical networking functionality for managed computer networks |
| JP5190084B2 (en) * | 2010-03-30 | 2013-04-24 | 株式会社日立製作所 | Virtual machine migration method and system |
| CN102075429B (en) * | 2011-01-21 | 2014-12-17 | 北京邮电大学 | Virtual network mapping method based on principle of proximity |
| CN102364899A (en) * | 2011-11-16 | 2012-02-29 | 北京邮电大学 | Particle-swam-optimization-based virtual network mapping method and system |
| CN102904794A (en) * | 2012-09-27 | 2013-01-30 | 北京邮电大学 | Method and device for mapping virtual network |
-
2013
- 2013-06-20 CN CN201310247047.3A patent/CN104243258B/en not_active Expired - Fee Related
-
2014
- 2014-06-20 SG SG11201510407UA patent/SG11201510407UA/en unknown
- 2014-06-20 WO PCT/CN2014/080372 patent/WO2014202016A1/en active Application Filing
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102075402A (en) * | 2011-02-12 | 2011-05-25 | 华为技术有限公司 | Virtual network mapping processing method and system |
| CN102868733A (en) * | 2012-08-29 | 2013-01-09 | 北京邮电大学 | Method for remapping virtual network resources |
Non-Patent Citations (2)
| Title |
|---|
| 一种启发式网络虚拟化资源分配算法;罗娟等;《中国科学：信息科学》;20120831;第42卷(第8期);正文第1-14页，图1-11 * |
| 虚拟网络映射问题研究及其进展;李小玲等;《软件学报 ISSN 1000-9825》;20121130;第23卷(第11期);正文第1-20页，图1-12 * |
Also Published As
| Publication number | Publication date |
|---|---|
| SG11201510407UA (en) | 2016-01-28 |
| CN104243258A (en) | 2014-12-24 |
| WO2014202016A1 (en) | 2014-12-24 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN104243258B (en) | A kind of mapping method of virtual network and system based on classification | |
| CN104995870B (en) | Multiple target server arrangement determines method and apparatus | |
| CN102149158B (en) | Method for fusing sensor grid data based on grid clustering | |
| CN104104621B (en) | A kind of virtual network resource dynamic self-adapting adjusting method based on Nonlinear Dimension Reduction | |
| CN107426332B (en) | A kind of load-balancing method and system of web server cluster | |
| CN113615137B (en) | CDN optimization platform | |
| US20180232840A1 (en) | Geospatial clustering for service coordination systems | |
| CN101873224A (en) | Cloud computing load balancing method and equipment | |
| WO2014208661A1 (en) | Device, method, system, and program for designing placement of virtual machine | |
| CN108768716A (en) | A kind of micro services routing resource and device | |
| Wolfson et al. | Fairness versus optimality in ridesharing | |
| Saraswat et al. | Energy efficient data forwarding scheme in fog-based ubiquitous system with deadline constraints | |
| CN106534318A (en) | OpenStack cloud platform resource dynamic scheduling system and method based on flow affinity | |
| Martinez et al. | Scalable design and dimensioning of fog-computing infrastructure to support latency-sensitive IoT applications | |
| CN104917659B (en) | A kind of mapping method of virtual network based on virtual network connection performance | |
| CN109831524A (en) | A kind of load balance process method and device | |
| Misra et al. | Traffic-aware efficient mapping of wireless body area networks to health cloud service providers in critical emergency situations | |
| CN106454948B (en) | Wireless network virtualizes interior joint redistribution method | |
| Gao et al. | Virtual network reconfiguration with load balancing and migration cost considerations | |
| CN102932271A (en) | Method and device for realizing load balancing | |
| US20210390502A1 (en) | Determining a fulfillment location for an expedited package request | |
| CN111327708B (en) | Optical network virtual mapping method based on topology perception | |
| Lin et al. | An optimal routing strategy for transport networks with minimal transmission cost and high network capacity | |
| Kaur et al. | Latency and network aware placement for cloud-native 5G/6G services | |
| Liu et al. | An optimized human resource management model for cloud-edge computing in the internet of things |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20171031 |
|
| CF01 | Termination of patent right due to non-payment of annual fee |