CN104869044A - Mapping method and mapping device for virtual network - Google Patents
Mapping method and mapping device for virtual network Download PDFInfo
- Publication number
- CN104869044A CN104869044A CN201510344602.3A CN201510344602A CN104869044A CN 104869044 A CN104869044 A CN 104869044A CN 201510344602 A CN201510344602 A CN 201510344602A CN 104869044 A CN104869044 A CN 104869044A
- Authority
- CN
- China
- Prior art keywords
- physical
- node
- physical node
- link
- virtual
- 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
Links
Landscapes
- Data Exchanges In Wide-Area Networks (AREA)
Abstract
The invention provides a mapping method and a mapping device for a virtual network. The mapping method comprises the following steps: performing safety capacity assessment on physical nodes in a bottom network shared by the virtual network to obtain a safety capacity matrix; based on the safety capacity matrix, determining the physical node corresponding to each virtual node in the virtual network; mapping the virtual link of each virtual node in the virtual network to a plurality of loop-free physical links corresponding to each physical node; randomly selecting one loop-free physical link from the loop-free physical links of the physical nodes, wherein the selected physical link is used for transmitting a virtual network request, that is, according to the mapping method provided by the scheme of the invention, one loop-free physical link can be randomly selected for transmitting the virtual network request during transmission of the virtual network request, the manner of randomly selecting the physical link can improve the unpredictability of link transmission, so as to improve the data transmission safety.
Description
Technical field
The invention belongs to technical field of the computer network, in particular, particularly relate to a kind of virtual net mapping method and device.
Background technology
To ossify problem for solving network, network virtualization technology is subject to the extensive concern of academia, wherein network virtualization technology allows the virtual net simultaneously running multiple isomery on the bottom-layer network shared, and each virtual net is equivalent to the resource burst of bottom-layer network, carrying particular type business.
In network virtualization technology, virtual net mapping problems is the key content of network virtualization research, mainly completes virtual net request being mapped on bottom-layer network idling-resource of task.Because the bottom-layer network topological structure in virtual net mapping is in dynamic change (being mainly reflected in node/link failure), so it is current research problem demanding prompt solution that the reliability how realizing virtual net under dynamic topology environment maps.
Reliability at present for virtual net maps research weakness relatively, mainly be divided into active-standby switch and online migration two kinds of modes, wherein active-standby switch refers in bottom-layer network to be every bar virtual link preparation two physical links, article one, physical link is main path, another physical link is backup path, when main path breaks down, switch to the request of backup path transfer of virtual net; Online migration is then when physical link breaks down, and calculates another physical link in real time and carrys out the request of transfer of virtual net.But these two kinds of modes all belong to Passive Defence, reduce the fail safe of transfer of data.
Summary of the invention
In view of this, the object of the present invention is to provide a kind of virtual net mapping method and device, every bar virtual link can be mapped to many physical links, and any physical link of random selecting transmits when transfer of data, improve the unpredictability of link transmission, thus improve the fail safe of transfer of data.
The invention provides a kind of virtual net mapping method, described method comprises:
In the bottom-layer network share virtual net, each physical node carries out safe capacity assessment, obtains safe capacity matrix;
Based on described safe capacity matrix, determine the physical node that each dummy node in described virtual net is corresponding;
The virtual link of dummy node each in described virtual net is mapped to many acyclic physical links of each self-corresponding physical node;
Random selecting acyclic physical link from many acyclic physical links of described physical node, selected acyclic physical link is used for the request of transfer of virtual net.
Preferably, described safe capacity assessment is carried out to each physical node in described bottom-layer network, obtains safe capacity matrix, comprising:
To obtain in described bottom-layer network any physical node to the maximum flow valuve c between u and v
u,v;
Change the rebuilding of the link weight of physical node each in described bottom-layer network into each self-corresponding message influenced proportionality coefficient
inverse, the influenced proportionality coefficient of message and the respective Prevention-Security capacity factor of wherein said each physical node are inversely proportional to, and described Prevention-Security capacity factor is known coefficient;
To obtain in amended bottom-layer network any physical node to the maximum flow valuve f between u and v
u,v;
Based on described maximum flow valuve f
u,v, obtain the first calculating parameter e of each element in described safe capacity matrix
u,v;
Based on described maximum flow valuve c
u,vwith described first calculating parameter e
u,v, obtain the element m in described safe capacity matrix
u,v, wherein 1≤u≤N, 1≤v≤N, N represents total number of physical node in described bottom-layer network, and N be greater than 1 integer.
Preferably, described based on described safe capacity matrix, determine to comprise the physical node that each dummy node in described virtual net is corresponding:
To the dummy node that network resource requirement is maximum from described virtual net, based on breadth first traversal principle, determine the mapping order of each dummy node in described virtual net;
Based on described mapping order, each described dummy node is mapped, wherein the mapping process of an xth dummy node is comprised: based on the network resource requirement of an xth dummy node, determine the first both candidate nodes set of i-th dummy node in bottom-layer network; Based on the first average security capacity between the physical node that each physical node and xth-1 dummy node in the first both candidate nodes set described in described safe capacity matrix computations is corresponding; An xth dummy node is mapped to the physical node that in described first both candidate nodes set, the first average security capacity is maximum, and 2≤x≤M, M represents total number of dummy node in described virtual net, and M be greater than 2 integer;
The mapping process of the 1st dummy node is comprised: based on the network resource requirement of the 1st dummy node, determine the second both candidate nodes set of the 1st dummy node in bottom-layer network; Based on the second average security capacity of each physical node in the second both candidate nodes set described in described safe capacity matrix computations; 1st dummy node is mapped to the physical node that in described second both candidate nodes set, the second average security capacity is maximum.
Preferably, the described virtual link by dummy node each in described virtual net is mapped to many acyclic physical links of each self-corresponding physical node, comprising:
The virtual link mapping problems of dummy node each in described virtual net is converted to linear programming problem, and described linear programming problem is the linear function problem of the stochastic flow that each virtual link is corresponding;
Described linear programming problem is solved, obtains the optimal stochastic Flow Policy P of each virtual link
i, P
k=(p
1, k, p
2, k..., p
t,k), t is the number of the physical link that i-th virtual link is mapped to, p
j,krepresent that virtual net request from a kth virtual link is through the stream probability of physical link j, 1≤k≤M;
Remove the loop in the optimal stochastic Flow Policy of each virtual link, obtain the acyclic stochastic flow strategy of each virtual link;
Described acyclic stochastic flow strategy is converted to acyclic random routing strategy R
k, R
kthe virtual link being used to indicate each dummy node is mapped to many acyclic physical links of each self-corresponding physical node, R
k=(r
1, k..., r
j,k..., r
z,k)
t, r
j,kfor virtual net request is from physical link
a physical node u be forwarded to the probability of another physical node v, z is total number of the physical node that each dummy node is mapped to.
Preferably, the loop in the optimal stochastic Flow Policy of each virtual link of described removal, obtains the acyclic stochastic flow strategy of each virtual link, comprising:
For optimal stochastic Flow Policy P
k, build by p
j,kthe network topology G of the physical link composition of >0
a;
If at described network topology G
ain find most short transmission path B between physical node u to physical node v, the minimum stream probability f on most short transmission path B described in acquisition; Wherein u and v is virtual link
be mapped to two end points after bottom-layer network, and u for described in the initial physical node in most short transmission path, v for described in the termination physical node in most short transmission path;
Upgrade network topology based on each described minimum stream probability obtained, wherein renewal process comprises: from network topology G
ain most short transmission path on remove stream probability and minimum stream probability difference be the physical link of zero, obtain the network topology G after upgrading
a, and to the network topology G after renewal
aperform at described network topology G
ain search most short transmission path B between physical node u to physical node v, and until there is not most short transmission path in minimum stream probability f on most short transmission path B described in obtaining, forms acyclic stochastic flow strategy for the stream probability that the physical link removed in each virtual link is corresponding between physical node u and physical node v;
If at described network topology G
ain find most short transmission path B between physical node u to physical node v, determine that the optimal stochastic Flow Policy of each virtual link is acyclic stochastic flow strategy.
The present invention also provides a kind of virtual net mapping device, it is characterized in that, described device comprises:
Assessment unit, for virtual net share bottom-layer network in each physical node carry out safe capacity assessment, obtain safe capacity matrix;
Determining unit, for based on described safe capacity matrix, determines the physical node that each dummy node in described virtual net is corresponding;
Map unit, for being mapped to many acyclic physical links of each self-corresponding physical node by the virtual link of dummy node each in described virtual net;
Choose unit, for random selecting acyclic physical link from many acyclic physical links of described physical node, selected acyclic physical link is used for the request of transfer of virtual net.
Preferably, described assessment unit comprises:
First obtains subelement, for obtaining in described bottom-layer network any physical node to the maximum flow valuve c between u and v
u,v;
Amendment subelement, for changing the rebuilding of the link weight of physical node each in described bottom-layer network into each self-corresponding message influenced proportionality coefficient
inverse, the influenced proportionality coefficient of message and the respective Prevention-Security capacity factor of wherein said each physical node are inversely proportional to, and described Prevention-Security capacity factor is known coefficient;
Second obtains subelement, for obtaining in amended bottom-layer network any physical node to the maximum flow valuve f between u and v
u,v;
First computation subunit, for based on described maximum flow valuve f
u,v, obtain the first calculating parameter e of each element in described safe capacity matrix
u,v;
Second computation subunit, for based on described maximum flow valuve c
u,vwith described first calculating parameter e
u,v, obtain the element m in described safe capacity matrix
u,v, wherein 1≤u≤N, 1≤v≤N, N represents total number of physical node in described bottom-layer network, and N be greater than 1 integer.
Preferably, described determining unit comprises:
Determine subelement, for from described virtual net to the dummy node that network resource requirement is maximum, based on breadth first traversal principle, determine the mapping order of each dummy node in described virtual net;
Map subelement, for based on described mapping order, each described dummy node is mapped, wherein the mapping process of an xth dummy node is comprised: based on the network resource requirement of an xth dummy node, determine the first both candidate nodes set of i-th dummy node in bottom-layer network; Based on the first average security capacity between the physical node that each physical node and xth-1 dummy node in the first both candidate nodes set described in described safe capacity matrix computations is corresponding; An xth dummy node is mapped to the physical node that in described first both candidate nodes set, the first average security capacity is maximum, and 2≤x≤M, M represents total number of dummy node in described virtual net, and M be greater than 2 integer;
The mapping process of the 1st dummy node is comprised: based on the network resource requirement of the 1st dummy node, determine the second both candidate nodes set of the 1st dummy node in bottom-layer network; Based on the second average security capacity of each physical node in the second both candidate nodes set described in described safe capacity matrix computations; 1st dummy node is mapped to the physical node that in described second both candidate nodes set, the second average security capacity is maximum.
Preferably, described map unit comprises:
First conversion subelement, for the virtual link mapping problems of dummy node each in described virtual net is converted to linear programming problem, described linear programming problem is the linear function problem of the stochastic flow that each virtual link is corresponding;
Solving subelement, for solving described linear programming problem, obtaining the optimal stochastic Flow Policy P of each virtual link
i, P
k=(p
1, k, p
2, k..., p
t,k), t is the number of the physical link that i-th virtual link is mapped to, p
j,krepresent that virtual net request from a kth virtual link is through the stream probability of physical link j, 1≤k≤M;
Removing subelement, for removing the loop in the optimal stochastic Flow Policy of each virtual link, obtaining the acyclic stochastic flow strategy of each virtual link;
Second conversion subelement, for being converted to acyclic random routing strategy R by described acyclic stochastic flow strategy
k, R
kthe virtual link being used to indicate each dummy node is mapped to many acyclic physical links of each self-corresponding physical node, R
k=(r
1, k..., r
j,k..., r
z,k)
t, r
j,kfor virtual net request is from physical link
a physical node u be forwarded to the probability of another physical node v, z is total number of the physical node that each dummy node is mapped to.
Preferably, described removal subelement comprises:
Build subelement, for for optimal stochastic Flow Policy P
k, build by p
j,kthe network topology G of the physical link composition of >0
a;
Obtain subelement, if at described network topology G
ain find most short transmission path B between physical node u to physical node v, the minimum stream probability f on most short transmission path B described in acquisition; Wherein u and v is virtual link
be mapped to two end points after bottom-layer network, and u for described in the initial physical node in most short transmission path, v for described in the termination physical node in most short transmission path;
Upgrade subelement, for upgrading network topology based on each described minimum stream probability obtained, wherein renewal process comprises: from network topology G
ain most short transmission path on remove stream probability and minimum stream probability difference be the physical link of zero, obtain the network topology G after upgrading
a, and to the network topology G after renewal
aperform at described network topology G
ain search most short transmission path B between physical node u to physical node v, and until there is not most short transmission path in minimum stream probability f on most short transmission path B described in obtaining, forms acyclic stochastic flow strategy for the stream probability that the physical link removed in each virtual link is corresponding between physical node u and physical node v;
Strategy determines subelement, if at described network topology G
ain find most short transmission path B between physical node u to physical node v, determine that the optimal stochastic Flow Policy of each virtual link is acyclic stochastic flow strategy.
Compared with prior art, technique scheme tool provided by the invention has the following advantages:
Technique scheme provided by the invention, by to virtual net share bottom-layer network in each physical node carry out safe capacity assessment, obtain safe capacity matrix, the physical node that in virtual net, each dummy node is corresponding can be determined based on safe capacity matrix, and the virtual link of each dummy node is mapped in many acyclic physical links of each self-corresponding physical node, the request of transfer of virtual net can be carried out by random selecting acyclic physical link like this, the mode of this random selecting physical link improves the unpredictability of link transmission, thus improve the fail safe of transfer of data.
Accompanying drawing explanation
In order to be illustrated more clearly in the embodiment of the present invention or technical scheme of the prior art, be briefly described to the accompanying drawing used required in embodiment or description of the prior art below, apparently, accompanying drawing in the following describes is some embodiments of the present invention, for those of ordinary skill in the art, under the prerequisite not paying creative work, other accompanying drawing can also be obtained according to these accompanying drawings.
Fig. 1 is the flow chart of the virtual net mapping method that the embodiment of the present invention provides;
Fig. 2 is the schematic diagram of the virtual net that the embodiment of the present invention provides;
Fig. 3 is the schematic diagram of the bottom-layer network that the embodiment of the present invention provides;
Fig. 4 is the schematic diagram that virtual net that the embodiment of the present invention provides is mapped to bottom-layer network;
Fig. 5 is the flow chart obtaining safe capacity matrix in the virtual net mapping method that provides of the embodiment of the present invention;
Fig. 6 is the flow chart that in the virtual net mapping method that provides of the embodiment of the present invention, virtual link maps;
Fig. 7 is the flow chart removing loop in the virtual net mapping method that provides of the embodiment of the present invention;
Fig. 8 is the schematic diagram that the embodiment of the present invention provides network topology;
Fig. 9 is the schematic diagram of the network topology after upgrading network topology shown in Fig. 8;
Figure 10 is the structural representation of the virtual net mapping device that the embodiment of the present invention provides;
Figure 11 is the structural representation of assessment unit in the virtual net mapping device that provides of the embodiment of the present invention;
Figure 12 is the structural representation of determining unit in the virtual net mapping device that provides of the embodiment of the present invention;
Figure 13 is the structural representation of map unit in the virtual net mapping device that provides of the embodiment of the present invention;
Figure 14 is the structural representation removing subelement in the virtual net mapping device that provides of the embodiment of the present invention.
Embodiment
For making the object of the embodiment of the present invention, technical scheme and advantage clearly, below in conjunction with the accompanying drawing in the embodiment of the present invention, technical scheme in the embodiment of the present invention is clearly and completely described, obviously, described embodiment is the present invention's part embodiment, instead of whole embodiments.Based on the embodiment in the present invention, those of ordinary skill in the art, not making the every other embodiment obtained under creative work prerequisite, belong to the scope of protection of the invention.
Refer to Fig. 1, it illustrates a kind of flow chart of the virtual net mapping method that the embodiment of the present invention provides, the virtual link of dummy node each in virtual net can be mapped to many acyclic physical links of physical node in bottom-layer network, can transmit by random selecting physical link when the request of transfer of virtual net like this.Virtual net mapping method shown in above-mentioned Fig. 1 can comprise the following steps:
101: in the bottom-layer network share virtual net, each physical node carries out safe capacity assessment, obtains safe capacity matrix.
In embodiments of the present invention, virtual net is the virtual subnet for particular demands service carried on bottom-layer network, can be expressed as a two-dimensional plot
wherein v is used to indicate mark has the symbol of v to be parameter in virtual net,
represent dummy node set in virtual net,
represent virtual link set in virtual net,
represent the cpu resource needed for each dummy node,
represent the bandwidth resources that virtual link needs, as shown in Figure 2, lowercase a to c represents the dummy node in virtual net, each lowercase goes out the cpu resource needed for numeral dummy node with square frame, arrow between two dummy nodes represents virtual link, the bandwidth resources that the numeral virtual link on it needs.
The physical network that bottom-layer network is shared as multiple virtual net, can be expressed as a two-dimensional plot G
s=(N
s, L
s, C
s, B
s), wherein s is used to indicate mark has the symbol of s to be parameter in bottom-layer network, N
srepresent the physical node set in bottom-layer network, L
srepresent the physical link set in bottom-layer network, C
scomprise C
s(n
s) and C
r(n
s), C
s(n
s) represent physical node n in bottom-layer network
scPU (Central ProcessingUnit, central processing unit) total resources, C
r(n
s) represent physical node n in bottom-layer network
scpu resource surplus, B
scomprise
with
represent the physical link between physical node u and v in bottom-layer network
bandwidth resources total amount,
represent the physical link between physical node u and v in bottom-layer network
bandwidth resources surplus, the schematic diagram of bottom-layer network as shown in Figure 3, wherein capitalization A to H represents the physical node in bottom-layer network, the cpu resource surplus of the numeral physical node of each physical node place square frame, arrow between two physical nodes represents physical link, the numeral bandwidth left amount on it.
Can obtain safe capacity matrix by carrying out safe dose assessment to each physical node in bottom-layer network, each element wherein in safe capacity matrix is for representing that corresponding physical node and physical link carry out the fail safe of transfer of data.For given bottom-layer network G
s=(N
s, L
s, C
s, B
s), its safe capacity matrix is M=[m
u,v]
n × N, N=|N
s| be physical node number in bottom-layer network, m
u,v=c
u,ve
u,v.
102: based on safe capacity matrix, determine the physical node that each dummy node in virtual net is corresponding.
103: many acyclic physical links virtual link of dummy node each in virtual net being mapped to each self-corresponding physical node.
In embodiments of the present invention, above-mentioned steps 102 and 103 is the node mapping in virtual net mapping and these two stages of link maps, and wherein virtual net maps is at bottom-layer network G
sin be virtual net
virtual net request find meet node resource and link circuit resource constraint optimum subgraph.
Node mapping can be expressed as:
dummy node in virtual net is mapped to the physical node of bottom-layer network, needs in the map to ensure that the cpu resource surplus of physical node is greater than the cpu resource needed for dummy node, i.e. the following formula of demand fulfillment:
Corresponding link maps can be expressed as:
virtual link each in virtual net is mapped in the acyclic physical link set in bottom-layer network, and determines the forwarding probability of every bar physical link, i.e. random routing strategy.Wherein, R is the set of all random routing strategies, and L is all acyclic physical link subset set.
The implication of random routing strategy is in embodiments of the present invention: for given virtual link
the down hop of the physical node of its virtual net request in each bottom-layer network forwards probability distribution, is
random routing strategy R
k=(r
1, k..., r
j,k..., r
z,k)
t, r
j,kfor virtual net request is from the physical link of bottom-layer network
a physical node u be forwarded to the probability of another physical node v, and the following condition of random routing strategy demand fulfillment:
From above-mentioned introducing random routing strategy, obtaining the forwarding probability between physical node, and after determining the physical node that dummy node is corresponding, can determine based on random routing strategy the multiple acyclic physical link that virtual link is corresponding.First the stochastic flow strategy of each dummy node will be obtained in embodiments of the present invention, the corresponding random routing strategy of each stochastic flow strategy in order to obtain random routing strategy.
Wherein each random routing strategy R
k=(r
1, k..., r
j,k..., r
z,k)
tcorresponding stochastic flow strategy is: P
k=(p
1, k..., p
j,k..., p
m, k)t, p
j,kto represent from the virtual net request of i-th virtual link through the stream probability of physical link j, add the numeral stream probability of sign as shown by the arrows in Figure 4, when stochastic flow direction indication is arrow direction indication be
(as+0.2 between physical node B and C), when stochastic flow direction indication is arrow direction indication is
(as-0.2 between physical node B and C), the relation between random routing strategy and stochastic flow strategy is as follows:
Therefore after obtaining stochastic flow strategy, stochastic flow strategy can be converted to random routing strategy based on formula (1) and (2), the virtual link of each dummy node to be mapped to many acyclic physical links of each self-corresponding physical node
104: random selecting acyclic physical link from many acyclic physical links of physical node, selected acyclic physical link is used for the request of transfer of virtual net.
As can be seen from such scheme, by to virtual net share bottom-layer network in each physical node carry out safe capacity assessment, obtain safe capacity matrix, the physical node that in virtual net, each dummy node is corresponding can be determined based on safe capacity matrix, and the virtual link of each dummy node is mapped in many acyclic physical links of each self-corresponding physical node, the request of transfer of virtual net can be carried out by random selecting acyclic physical link like this, the mode of this random selecting physical link improves the unpredictability of link transmission, thus improve the fail safe of transfer of data.
Below in conjunction with flow chart, each step above-mentioned is described in detail, refers to Fig. 5, it illustrates the flow chart of step 101 in the virtual net mapping method that the embodiment of the present invention provides, can comprise the following steps:
1011: in acquisition bottom-layer network, any physical node is to the maximum flow valuve c between u and v
u,v.
If bottom-layer network is regarded as an oil pipeline net, physical node u represents the sending point in oil pipeline net, physical node v represents the acceptance point in oil pipeline net, other physical nodes represent terminal, the bandwidth left amount of physical link instruction can represent the maximum delivery of this segment pipe, so how to arrange oil transportation circuit just can make maximum to the total gross traffic of acceptance point v from sending point u, such problem is called maximum flow problem, therefore maximum flow valuve c
u,vrefer to from physical node u, the max-flow of physical node v can be arrived.Maximum flow valuve c in embodiments of the present invention
u,vhao-Orlin Algorithm for Solving can be utilized to obtain.Wherein Hao and Orlin is the surname of Hao-Orlin algorithm paper two authors respectively.
1012: change the rebuilding of the link weight of physical node each in bottom-layer network into each self-corresponding message influenced proportionality coefficient
inverse, wherein the influenced proportionality coefficient of the message of each physical node and respective Prevention-Security capacity factor are inversely proportional to, and Prevention-Security capacity factor is known coefficient.In embodiments of the present invention, Prevention-Security ability is used to indicate the fail safe of the physical link of each physical node, can be obtained by safe class test.
1013: to obtain in amended bottom-layer network any physical node to the maximum flow valuve f between u and v
u,v.Maximum flow valuve f
u,vbe used to indicate from physical node u in amended bottom-layer network equally, the max-flow of physical node v can be arrived, Hao-Orlin Algorithm for Solving can be utilized to obtain.
1014: based on maximum flow valuve f
u,v, obtain the first calculating parameter e of each element in safe capacity matrix
u,v, wherein e
u,v=1-1/f
u,v, the first calculating parameter e of each element in safe capacity matrix namely can be obtained based on this formula
u,v.
1015: based on the first maximum flow valuve c
u,vwith the first calculating parameter e
u,v, obtain the element m in safe capacity matrix
u,v, wherein m
u,v=c
u,ve
u,v, 1≤u≤N, 1≤v≤N, N represents total number of physical node in bottom-layer network, and N be greater than 1 integer.
Namely the safe capacity matrix of the bottom-layer network that virtual net is shared can be obtained by above-mentioned steps, because each element in safe capacity matrix is for representing that corresponding physical node and physical link carry out the fail safe of transfer of data, so when determining the physical node that in virtual net, each dummy node is corresponding, can the higher physical node of preferred security, to improve the fail safe of transfer of data.
In embodiments of the present invention, based on safe capacity matrix, determine that the process of the physical node that each dummy node in virtual net is corresponding is as follows:
First, to the dummy node that network resource requirement is maximum from virtual net, based on breadth first traversal principle, determine the mapping order of each dummy node in virtual net, then based on mapping order, each dummy node is mapped, also different according to its mapping process of the order of dummy node when mapping dummy node.
Mapping process comprises in embodiments of the present invention: to the mapping process of the 1st dummy node and the mapping process to the dummy node after the 1st dummy node.Wherein the mapping process of the 1st dummy node is comprised: based on the network resource requirement of the 1st dummy node, determine the second both candidate nodes set of the 1st dummy node in bottom-layer network; Based on the second average security capacity of each physical node in the set of safe capacity matrix computations second both candidate nodes; 1st dummy node is mapped to the physical node that in the second both candidate nodes set, the second average security capacity is maximum.Wherein in determined second both candidate nodes set, the cpu resource surplus of each physical node is more than or equal to the 1st cpu resource needed for dummy node.Second average security capacity is then the mean value of the safe capacity between other all second both candidate nodes, such as the second both candidate nodes set comprises four physical nodes, be respectively physical node 1, physical node 2, physical node 3 and physical node 4, then the second average security capacity of physical node 1 is: (safe capacity between the safe capacity+physical node 1 and 4 between the safe capacity+physical node 1 and 3 between physical node 1 and 2)/3, the safe capacity wherein between physical node 1 and 2 is m
1.2.
Mapping process accordingly for other dummy nodes after the 1st dummy node comprises: based on the network resource requirement of an xth dummy node, determine the first both candidate nodes set of i-th dummy node in bottom-layer network; Based on each physical node and the 1st in the set of safe capacity matrix computations first both candidate nodes, 2 ..., the first average security capacity between the physical node that x-1 dummy node is corresponding; An xth dummy node is mapped to the physical node that in the first both candidate nodes set, the first average security capacity is maximum, 2≤x≤M, M represents total number of dummy node in virtual net, and M be greater than 2 integer, wherein in determined first both candidate nodes set, the cpu resource surplus of each physical node is more than or equal to the cpu resource needed for an xth dummy node.
Such as the 3rd the determined first both candidate nodes set of dummy node comprises four physical nodes, be respectively physical node 5, physical node 6, physical node 7 and physical node 8, and the safe capacity of physical node 5 and the 1st and the 2nd determined first both candidate nodes of dummy node is a and b respectively, then the first average security capacity of physical node 5 is: (a+b)/2.
Here it should be noted is that: above-mentioned breadth-first variable principle is the mapping order going out each dummy node based on the order computation that the bandwidth resources of dummy node needs are descending.In addition, when the number of the first maximum average security capacity and the second maximum average security capacity is multiple, can map by one of them physical node of random selecting.
Above-mentioned node mapping of having set forth in virtual net mapping, then the link maps in virtual net mapping is introduced below, namely the detailed process of step 103 in the virtual net mapping method shown in Fig. 1 is described, specifically can consults shown in Fig. 6, comprise the following steps:
1031: the virtual link mapping problems of dummy node each in virtual net is converted to linear programming problem, linear programming problem is the linear function problem of the stochastic flow that each virtual link is corresponding.
In embodiments of the present invention, virtual link mapping problems being converted to linear programming problem is: suppose that variable is stochastic flow p
j,k(λ is newly-increased variable with λ, without particular meaning), and solve min λ, be bandwidth constraint condition to the constraints of this Two Variables in solution procedure, namely for every bar virtual link, bandwidth for its physical link distribution mapped is less than or equal to the remaining bandwidth of this physical link, and formula is:
it can be reduced to:
Variable p
j,kthe condition met is as follows:
1032: linear programming problem is solved, obtain the optimal stochastic Flow Policy P of each virtual link
k=(p
1, k, p
2, k..., p
t,k), t is the number of the physical link that i-th virtual link is mapped to, p
j,krepresent that virtual net request from a kth virtual link is through the stream probability of physical link j, 1≤k≤M.
After converting linear programming problem to, utilize lp_solve function library to solve above-mentioned linear programming problem, obtain the set of optimal stochastic Flow Policy
wherein
target function min λ (P when obtaining minimum value
1..., P
m) value, m is the total number of dummy node, and wherein lp_solve function library is an open source software program, is specifically designed to and solves linear programming problem.
1033: remove the loop in the optimal stochastic Flow Policy of each virtual link, obtain the acyclic stochastic flow strategy of each virtual link.Wherein so-called loop refers to that the start node in the path of many physical link compositions and terminal node are same node, but in the transmitting procedure of reality, there is not the situation passing again start node through multiple physical node back, therefore the embodiment of the present invention needs to remove the loop in the optimal stochastic strategy of each virtual link.
With optimal stochastic Flow Policy P
kfor example, the process removing loop can be consulted shown in Fig. 7, comprises the following steps:
201: build by p
j,kthe network topology G of the physical link composition of >0
a.The schematic diagram that virtual net as shown in Figure 4 maps, arrow wherein with sign numeral place is the stochastic flow of each dummy node, positive sign represents that stochastic flow is forward transmission, and the transmission direction as physical node B to C is forward transmission, can construct p by the numeral on it
j,kthe network topology G of the physical link composition of >0
a, as shown in Figure 8, Fig. 8 is the network topology G built based on Fig. 4
a.
202: if at network topology G
ain find most short transmission path B between physical node u to physical node v, obtain the minimum stream probability f on most short transmission path B.Wherein u and v is virtual link
be mapped to two end points after bottom-layer network, and u is the initial physical node in most short transmission path, v is the termination physical node in most short transmission path.
In embodiments of the present invention, most short transmission path B is from the minimum path of the transmission range physical node u to physical node v, it can be obtained by dijkstra (Di Jiesitela) algorithm, wherein dijkstra algorithm is typical signal source shortest path algorithm, for calculating the shortest path of a node to other all nodes.Most short transmission path B comprises at least one stochastic flow, therefore after finding most short transmission path, from the stochastic flow that this most short transmission path comprises, chooses minimum stream probability as minimum stream probability f.
203: upgrade network topology based on each minimum stream probability obtained, wherein renewal process comprises: from network topology G
ain most short transmission path on remove stream probability and minimum stream probability difference be the physical link of zero, obtain the network topology G after upgrading
a, and to the network topology G after renewal
aperform at network topology G
ain search most short transmission path B between physical node u to physical node v, and the minimum stream probability f obtained on most short transmission path B is until can not find the path between physical node u to physical node v, forms acyclic stochastic flow strategy for the stream probability that the physical link removed in each virtual link is corresponding.
To suppose in Fig. 8 that most short transmission path is D → C → G → A, minimum stream probability is 0.1, then the network topology G after upgrading
aas shown in Figure 9, to the network topology G shown in Fig. 9
aexecution step 202 searches the network topology G after renewal
ain most short transmission path B, and obtain the minimum stream probability f on it, then utilize minimum stream probability f to the network topology G shown in Fig. 9
aagain carry out upgrading to obtain the most short transmission path of the next one and minimum stream probability.
As can be seen from said process, often find a most short transmission path and get a minimum stream probability, can by these two parameters to network topology G
aupgrade.
204: if at network topology G
ain find most short transmission path B between physical node u to physical node v, determine that the optimal stochastic Flow Policy of each virtual link is acyclic stochastic flow strategy.
1034: acyclic stochastic flow strategy is converted to acyclic random routing strategy R
k, R
kthe virtual link being used to indicate each dummy node is mapped to many acyclic physical links of each self-corresponding physical node, R
k=(r
1, k..., r
j,k..., r
z,k)
t, r
j,kfor virtual net request is from physical link
a physical node u be forwarded to the probability of another physical node v, z is total number of the physical node that each dummy node is mapped to.
Above-mentioned formula (1) and (2) are the corresponding relation between random routing strategy and stochastic flow strategy, therefore, after obtaining acyclic stochastic flow strategy, acyclic random routing strategy R can be obtained based on formula (1) and (2)
k.
For aforesaid each embodiment of the method, in order to simple description, therefore it is all expressed as a series of combination of actions, but those skilled in the art should know, the present invention is not by the restriction of described sequence of movement, because according to the present invention, some step can adopt other orders or carry out simultaneously.Secondly, those skilled in the art also should know, the embodiment described in specification all belongs to preferred embodiment, and involved action and module might not be that the present invention is necessary.
Corresponding with said method embodiment, the embodiment of the present invention also provides a kind of virtual net mapping device, and its structural representation as shown in Figure 10, can comprise: assessment unit 11, determining unit 12, map unit 13 and choose unit 14.
Assessment unit 11, for virtual net share bottom-layer network in each physical node carry out safe capacity assessment, obtain safe capacity matrix.Each element wherein in safe capacity matrix is for representing that corresponding physical node and physical link carry out the fail safe of transfer of data.For given bottom-layer network G
s=(N
s, L
s, C
s, B
s), its safe capacity matrix is M=[m
u,v]
n × N, N=|N
s| be physical node number in bottom-layer network, m
u,v=c
u,ve
u,v.
Determining unit 12, for based on safe capacity matrix, determines the physical node that each dummy node in virtual net is corresponding.
Map unit 13, for being mapped to many acyclic physical links of each self-corresponding physical node by the virtual link of dummy node each in virtual net.
In embodiments of the present invention, the function that above-mentioned determining unit 12 and map unit 13 possess be virtual net map in node mapping and these two stages of link maps, wherein virtual net map be at bottom-layer network G
sin be virtual net
virtual net request find meet node resource and link circuit resource constraint optimum subgraph, comprise node mapping and link maps two processes, specifically refer to the structure of following determining unit 12 and map unit 13.
Choose unit 14, for random selecting acyclic physical link from many acyclic physical links of physical node, selected acyclic physical link is used for the request of transfer of virtual net.
As can be seen from such scheme, by to virtual net share bottom-layer network in each physical node carry out safe capacity assessment, obtain safe capacity matrix, the physical node that in virtual net, each dummy node is corresponding can be determined based on safe capacity matrix, and the virtual link of each dummy node is mapped in many acyclic physical links of each self-corresponding physical node, the request of transfer of virtual net can be carried out by random selecting acyclic physical link like this, the mode of this random selecting physical link improves the unpredictability of link transmission, thus improve the fail safe of transfer of data.
In embodiments of the present invention, the structural representation of assessment unit 11 can as Figure 11 so, comprising: first obtains subelement 111, amendment subelement 112, second obtains subelement 113, first computation subunit 114 and the second computation subunit 115.
First obtains subelement 111, for obtaining in bottom-layer network any physical node to the maximum flow valuve c between u and v
u,v.If bottom-layer network is regarded as an oil pipeline net, physical node u represents the sending point in oil pipeline net, physical node v represents the acceptance point in oil pipeline net, other physical nodes represent terminal, the bandwidth left amount of physical link instruction can represent the maximum delivery of this segment pipe, so how to arrange oil transportation circuit just can make maximum to the total gross traffic of acceptance point v from sending point u, such problem is called maximum flow problem, therefore maximum flow valuve c
u,vrefer to from physical node u, the max-flow of physical node v can be arrived.Maximum flow valuve c in embodiments of the present invention
u,vhao-Orlin Algorithm for Solving can be utilized to obtain.Wherein Hao and Orlin is the surname of Hao-Orlin algorithm paper two authors respectively.
Amendment subelement 112, for changing the rebuilding of the link weight of physical node each in bottom-layer network into each self-corresponding message influenced proportionality coefficient
inverse, wherein the influenced proportionality coefficient of the message of each physical node and respective Prevention-Security capacity factor are inversely proportional to, and Prevention-Security capacity factor is known coefficient.In embodiments of the present invention, Prevention-Security ability is used to indicate the fail safe of the physical link of each physical node, can be obtained by safe class test.
Second obtains subelement 113, for obtaining in amended bottom-layer network any physical node to the maximum flow valuve f between u and v
u,v.Maximum flow valuve f
u,vbe used to indicate from physical node u in amended bottom-layer network equally, the max-flow of physical node v can be arrived, Hao-Orlin Algorithm for Solving can be utilized to obtain.
First computation subunit 114, for based on maximum flow valuve f
u,v, obtain the first calculating parameter e of each element in safe capacity matrix
u,v.Wherein e
u,v=1-1/f
u,v, the first calculating parameter e of each element in safe capacity matrix namely can be obtained based on this formula
u,v.
Second computation subunit 115, for based on maximum flow valuve c
u,vwith the first calculating parameter e
u,v, obtain the element m in safe capacity matrix
u,v, wherein m
u,v=c
u,ve
u,v, 1≤u≤N, 1≤v≤N, N represents total number of physical node in bottom-layer network, and N be greater than 1 integer.
Before address, determining unit 12 is for being mapped as the physical node in bottom-layer network by the dummy node in virtual net, its structure can as shown in figure 12, comprise accordingly: determine subelement 121 and map subelement 122.
Determine subelement 121, for from virtual net to the dummy node that network resource requirement is maximum, based on breadth first traversal principle, determine the mapping order of each dummy node in virtual net.
Map subelement 122, for based on mapping order, each dummy node is mapped, wherein the mapping process of an xth dummy node is comprised: based on the network resource requirement of an xth dummy node, determine the first both candidate nodes set of i-th dummy node in bottom-layer network.Based on the first average security capacity between the physical node that each physical node and xth-1 dummy node in the set of safe capacity matrix computations first both candidate nodes is corresponding.An xth dummy node is mapped to the physical node that in the first both candidate nodes set, the first average security capacity is maximum, and 2≤x≤M, M represents total number of dummy node in virtual net, and M be greater than 2 integer.
The mapping process of the 1st dummy node is comprised: based on the network resource requirement of the 1st dummy node, determine the second both candidate nodes set of the 1st dummy node in bottom-layer network.Based on the second average security capacity of each physical node in the set of safe capacity matrix computations second both candidate nodes.1st dummy node is mapped to the physical node that in the second both candidate nodes set, the second average security capacity is maximum.
Here it should be noted is that: above-mentioned breadth-first variable principle is the mapping order going out each dummy node based on the order computation that the bandwidth resources of dummy node needs are descending.In addition, when the number of the first maximum average security capacity and the second maximum average security capacity is multiple, can map by one of them physical node of random selecting.
The structural representation of corresponding map unit 13 as shown in figure 13, can comprise: the first conversion subelement 131, solve subelement 132, remove subelement 133 and the second conversion subelement 134.
First conversion subelement 131, for the virtual link mapping problems of dummy node each in virtual net is converted to linear programming problem, linear programming problem is the linear function problem of the stochastic flow that each virtual link is corresponding.
Solving subelement 132, for solving linear programming problem, obtaining the optimal stochastic Flow Policy P of each virtual link
i, P
k=(p
1, k, p
2, k..., p
t,k), t is the number of the physical link that i-th virtual link is mapped to, p
j,krepresent that virtual net request from a kth virtual link is through the stream probability of physical link j, 1≤k≤M.
Removing subelement 133, for removing the loop in the optimal stochastic Flow Policy of each virtual link, obtaining the acyclic stochastic flow strategy of each virtual link.Wherein remove subelement 133 to comprise: build subelement 1331, acquisition subelement 1332, renewal subelement 1333 and strategy and determine subelement 1334, as shown in figure 14.
Build subelement 1331, for for optimal stochastic Flow Policy P
k, build by p
j,kthe network topology G of the physical link composition of >0
a.
Obtain subelement 1332, if at network topology G
ain find most short transmission path B between physical node u to physical node v, obtain the minimum stream probability f on most short transmission path B.Wherein u and v is virtual link
be mapped to two end points after bottom-layer network, and u is the initial physical node in most short transmission path, v is the termination physical node in most short transmission path.
Upgrade subelement 1333, for upgrading network topology based on each minimum stream probability obtained, wherein renewal process comprises: from network topology G
ain most short transmission path on remove stream probability and minimum stream probability difference be the physical link of zero, obtain the network topology G after upgrading
a, and to the network topology G after renewal
aperform at network topology G
ain search most short transmission path B between physical node u to physical node v, and until there is not most short transmission path in the minimum stream probability f obtained on most short transmission path B, forms acyclic stochastic flow strategy for the stream probability that the physical link removed in each virtual link is corresponding between physical node u and physical node v.
Strategy determines subelement 1334, if at network topology G
ain find most short transmission path B between physical node u to physical node v, determine that the optimal stochastic Flow Policy of each virtual link is acyclic stochastic flow strategy.
Second conversion subelement 134, for being converted to acyclic random routing strategy R by acyclic stochastic flow strategy
k, R
kthe virtual link being used to indicate each dummy node is mapped to many acyclic physical links of each self-corresponding physical node, R
k=(r
1, k..., r
j,k..., r
z,k)
t, r
j,kfor virtual net request is from physical link
a physical node u be forwarded to the probability of another physical node v, z is total number of the physical node that each dummy node is mapped to.
It should be noted that, each embodiment in this specification all adopts the mode of going forward one by one to describe, and what each embodiment stressed is the difference with other embodiments, between each embodiment identical similar part mutually see.For device class embodiment, due to itself and embodiment of the method basic simlarity, so description is fairly simple, relevant part illustrates see the part of embodiment of the method.
Finally, also it should be noted that, in this article, the such as relational terms of first and second grades and so on is only used for an entity or operation to separate with another entity or operating space, and not necessarily requires or imply the relation that there is any this reality between these entities or operation or sequentially.And, term " comprises ", " comprising " or its any other variant are intended to contain comprising of nonexcludability, thus make to comprise the process of a series of key element, method, article or equipment and not only comprise those key elements, but also comprise other key elements clearly do not listed, or also comprise by the intrinsic key element of this process, method, article or equipment.When not more restrictions, the key element limited by statement " comprising ... ", and be not precluded within process, method, article or the equipment comprising described key element and also there is other identical element.
To the above-mentioned explanation of the disclosed embodiments, those skilled in the art are realized or uses the present invention.To be apparent for a person skilled in the art to the multiple amendment of these embodiments, General Principle as defined herein can without departing from the spirit or scope of the present invention, realize in other embodiments.Therefore, the present invention can not be restricted to these embodiments shown in this article, but will meet the widest scope consistent with principle disclosed herein and features of novelty.
The above is only the preferred embodiment of the present invention; it should be pointed out that for those skilled in the art, under the premise without departing from the principles of the invention; can also make some improvements and modifications, these improvements and modifications also should be considered as protection scope of the present invention.
Claims (10)
1. a virtual net mapping method, is characterized in that, described method comprises:
In the bottom-layer network share virtual net, each physical node carries out safe capacity assessment, obtains safe capacity matrix;
Based on described safe capacity matrix, determine the physical node that each dummy node in described virtual net is corresponding;
The virtual link of dummy node each in described virtual net is mapped to many acyclic physical links of each self-corresponding physical node;
Random selecting acyclic physical link from many acyclic physical links of described physical node, selected acyclic physical link is used for the request of transfer of virtual net.
2. method according to claim 1, is characterized in that, describedly carries out safe capacity assessment to each physical node in described bottom-layer network, obtains safe capacity matrix, comprising:
To obtain in described bottom-layer network any physical node to the maximum flow valuve c between u and v
u,v;
Change the rebuilding of the link weight of physical node each in described bottom-layer network into each self-corresponding message influenced proportionality coefficient
inverse, the influenced proportionality coefficient of message and the respective Prevention-Security capacity factor of wherein said each physical node are inversely proportional to, and described Prevention-Security capacity factor is known coefficient;
To obtain in amended bottom-layer network any physical node to the maximum flow valuve f between u and v
u,v;
Based on described maximum flow valuve f
u,v, obtain the first calculating parameter e of each element in described safe capacity matrix
u,v;
Based on described maximum flow valuve c
u,vwith described first calculating parameter e
u,v, obtain the element m in described safe capacity matrix
u,v, wherein 1≤u≤N, 1≤v≤N, N represents total number of physical node in described bottom-layer network, and N be greater than 1 integer.
3. method according to claim 2, is characterized in that, described based on described safe capacity matrix, determines to comprise the physical node that each dummy node in described virtual net is corresponding:
To the dummy node that network resource requirement is maximum from described virtual net, based on breadth first traversal principle, determine the mapping order of each dummy node in described virtual net;
Based on described mapping order, each described dummy node is mapped, wherein the mapping process of an xth dummy node is comprised: based on the network resource requirement of an xth dummy node, determine the first both candidate nodes set of i-th dummy node in bottom-layer network; Based on the first average security capacity between the physical node that each physical node and xth-1 dummy node in the first both candidate nodes set described in described safe capacity matrix computations is corresponding; An xth dummy node is mapped to the physical node that in described first both candidate nodes set, the first average security capacity is maximum, and 2≤x≤M, M represents total number of dummy node in described virtual net, and M be greater than 2 integer;
The mapping process of the 1st dummy node is comprised: based on the network resource requirement of the 1st dummy node, determine the second both candidate nodes set of the 1st dummy node in bottom-layer network; Based on the second average security capacity of each physical node in the second both candidate nodes set described in described safe capacity matrix computations; 1st dummy node is mapped to the physical node that in described second both candidate nodes set, the second average security capacity is maximum.
4. method according to claim 3, is characterized in that, the described virtual link by dummy node each in described virtual net is mapped to many acyclic physical links of each self-corresponding physical node, comprising:
The virtual link mapping problems of dummy node each in described virtual net is converted to linear programming problem, and described linear programming problem is the linear function problem of the stochastic flow that each virtual link is corresponding;
Described linear programming problem is solved, obtains the optimal stochastic Flow Policy P of each virtual link
i, P
k=(p
1, k, p
2, k..., p
t,k), t is the number of the physical link that i-th virtual link is mapped to, p
j,krepresent that virtual net request from a kth virtual link is through the stream probability of physical link j, 1≤k≤M;
Remove the loop in the optimal stochastic Flow Policy of each virtual link, obtain the acyclic stochastic flow strategy of each virtual link;
Described acyclic stochastic flow strategy is converted to acyclic random routing strategy R
k, R
kthe virtual link being used to indicate each dummy node is mapped to many acyclic physical links of each self-corresponding physical node, R
k=(r
1, k..., r
j,k..., r
z,k)
t, r
j,kfor virtual net request is from physical link
a physical node u be forwarded to the probability of another physical node v, z is total number of the physical node that each dummy node is mapped to.
5. method according to claim 4, is characterized in that, the loop in the optimal stochastic Flow Policy of each virtual link of described removal, obtains the acyclic stochastic flow strategy of each virtual link, comprising:
For optimal stochastic Flow Policy P
k, build by p
j,kthe network topology G of the physical link composition of >0
a;
If at described network topology G
ain find most short transmission path B between physical node u to physical node v, the minimum stream probability f on most short transmission path B described in acquisition; Wherein u and v is virtual link
be mapped to two end points after bottom-layer network, and u for described in the initial physical node in most short transmission path, v for described in the termination physical node in most short transmission path;
Upgrade network topology based on each described minimum stream probability obtained, wherein renewal process comprises: from network topology G
ain most short transmission path on remove stream probability and minimum stream probability difference be the physical link of zero, obtain the network topology G after upgrading
a, and to the network topology G after renewal
aperform at described network topology G
ain search most short transmission path B between physical node u to physical node v, and until there is not most short transmission path in minimum stream probability f on most short transmission path B described in obtaining, forms acyclic stochastic flow strategy for the stream probability that the physical link removed in each virtual link is corresponding between physical node u and physical node v;
If at described network topology G
ain find most short transmission path B between physical node u to physical node v, determine that the optimal stochastic Flow Policy of each virtual link is acyclic stochastic flow strategy.
6. a virtual net mapping device, is characterized in that, described device comprises:
Assessment unit, for virtual net share bottom-layer network in each physical node carry out safe capacity assessment, obtain safe capacity matrix;
Determining unit, for based on described safe capacity matrix, determines the physical node that each dummy node in described virtual net is corresponding;
Map unit, for being mapped to many acyclic physical links of each self-corresponding physical node by the virtual link of dummy node each in described virtual net;
Choose unit, for random selecting acyclic physical link from many acyclic physical links of described physical node, selected acyclic physical link is used for the request of transfer of virtual net.
7. device according to claim 6, is characterized in that, described assessment unit comprises:
First obtains subelement, for obtaining in described bottom-layer network any physical node to the maximum flow valuve c between u and v
u,v;
Amendment subelement, for changing the rebuilding of the link weight of physical node each in described bottom-layer network into each self-corresponding message influenced proportionality coefficient
inverse, the influenced proportionality coefficient of message and the respective Prevention-Security capacity factor of wherein said each physical node are inversely proportional to, and described Prevention-Security capacity factor is known coefficient;
Second obtains subelement, for obtaining in amended bottom-layer network any physical node to the maximum flow valuve f between u and v
u,v;
First computation subunit, for based on described maximum flow valuve f
u,v, obtain the first calculating parameter e of each element in described safe capacity matrix
u,v;
Second computation subunit, for based on described maximum flow valuve c
u,vwith described first calculating parameter e
u,v, obtain the element m in described safe capacity matrix
u,v, wherein 1≤u≤N, 1≤v≤N, N represents total number of physical node in described bottom-layer network, and N be greater than 1 integer.
8. device according to claim 7, is characterized in that, described determining unit comprises:
Determine subelement, for from described virtual net to the dummy node that network resource requirement is maximum, based on breadth first traversal principle, determine the mapping order of each dummy node in described virtual net;
Map subelement, for based on described mapping order, each described dummy node is mapped, wherein the mapping process of an xth dummy node is comprised: based on the network resource requirement of an xth dummy node, determine the first both candidate nodes set of i-th dummy node in bottom-layer network; Based on the first average security capacity between the physical node that each physical node and xth-1 dummy node in the first both candidate nodes set described in described safe capacity matrix computations is corresponding; An xth dummy node is mapped to the physical node that in described first both candidate nodes set, the first average security capacity is maximum, and 2≤x≤M, M represents total number of dummy node in described virtual net, and M be greater than 2 integer;
The mapping process of the 1st dummy node is comprised: based on the network resource requirement of the 1st dummy node, determine the second both candidate nodes set of the 1st dummy node in bottom-layer network; Based on the second average security capacity of each physical node in the second both candidate nodes set described in described safe capacity matrix computations; 1st dummy node is mapped to the physical node that in described second both candidate nodes set, the second average security capacity is maximum.
9. device according to claim 8, is characterized in that, described map unit comprises:
First conversion subelement, for the virtual link mapping problems of dummy node each in described virtual net is converted to linear programming problem, described linear programming problem is the linear function problem of the stochastic flow that each virtual link is corresponding;
Solving subelement, for solving described linear programming problem, obtaining the optimal stochastic Flow Policy P of each virtual link
i, P
k=(p
1, k, p
2, k..., p
t,k), t is the number of the physical link that i-th virtual link is mapped to, p
j,krepresent that virtual net request from a kth virtual link is through the stream probability of physical link j, 1≤k≤M;
Removing subelement, for removing the loop in the optimal stochastic Flow Policy of each virtual link, obtaining the acyclic stochastic flow strategy of each virtual link;
Second conversion subelement, for being converted to acyclic random routing strategy R by described acyclic stochastic flow strategy
k, R
kthe virtual link being used to indicate each dummy node is mapped to many acyclic physical links of each self-corresponding physical node, R
k=(r
1, k..., r
j,k..., r
z,k)
t, r
j,kfor virtual net request is from physical link
a physical node u be forwarded to the probability of another physical node v, z is total number of the physical node that each dummy node is mapped to.
10. device according to claim 9, is characterized in that, described removal subelement comprises:
Build subelement, for for optimal stochastic Flow Policy P
k, build by p
j,kthe network topology G of the physical link composition of >0
a;
Obtain subelement, if at described network topology G
ain find most short transmission path B between physical node u to physical node v, the minimum stream probability f on most short transmission path B described in acquisition; Wherein u and v is virtual link
be mapped to two end points after bottom-layer network, and u for described in the initial physical node in most short transmission path, v for described in the termination physical node in most short transmission path;
Upgrade subelement, for upgrading network topology based on each described minimum stream probability obtained, wherein renewal process comprises: from network topology G
ain most short transmission path on remove stream probability and minimum stream probability difference be the physical link of zero, obtain the network topology G after upgrading
a, and to the network topology G after renewal
aperform at described network topology G
ain search most short transmission path B between physical node u to physical node v, and until there is not most short transmission path in minimum stream probability f on most short transmission path B described in obtaining, forms acyclic stochastic flow strategy for the stream probability that the physical link removed in each virtual link is corresponding between physical node u and physical node v;
Strategy determines subelement, if at described network topology G
ain find most short transmission path B between physical node u to physical node v, determine that the optimal stochastic Flow Policy of each virtual link is acyclic stochastic flow strategy.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201510344602.3A CN104869044B (en) | 2015-06-19 | 2015-06-19 | A kind of virtual net mapping method and device |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201510344602.3A CN104869044B (en) | 2015-06-19 | 2015-06-19 | A kind of virtual net mapping method and device |
Publications (2)
Publication Number | Publication Date |
---|---|
CN104869044A true CN104869044A (en) | 2015-08-26 |
CN104869044B CN104869044B (en) | 2019-01-29 |
Family
ID=53914583
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201510344602.3A Active CN104869044B (en) | 2015-06-19 | 2015-06-19 | A kind of virtual net mapping method and device |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN104869044B (en) |
Cited By (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN107124287A (en) * | 2016-02-24 | 2017-09-01 | 中移(苏州)软件技术有限公司 | A kind of virtual network service method for quality control and device |
CN107231412A (en) * | 2017-05-22 | 2017-10-03 | 浙江工商大学 | A kind of service path building method mapped based on virtual net |
CN107733677A (en) * | 2016-08-12 | 2018-02-23 | 南京中兴软件有限责任公司 | The method for generating alarm and device of a kind of stacking network |
EP3929781A1 (en) * | 2020-06-25 | 2021-12-29 | Vocalink Limited | Reconstructing a dataset after detection of a network security threat in a network |
Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102136998A (en) * | 2010-08-30 | 2011-07-27 | 华为技术有限公司 | Traffic engineering and server selection joint optimization method, system and related equipment |
CN102904794A (en) * | 2012-09-27 | 2013-01-30 | 北京邮电大学 | Method and device for mapping virtual network |
CN102932479A (en) * | 2012-11-16 | 2013-02-13 | 北京邮电大学 | Virtual network mapping method for realizing topology awareness based on historical data |
JP2013157855A (en) * | 2012-01-31 | 2013-08-15 | Nec Corp | Method of connecting virtual network, virtual network connection device, and program |
CN103457752A (en) * | 2012-05-30 | 2013-12-18 | 中国科学院声学研究所 | Virtual network mapping method |
CN104320343A (en) * | 2014-10-28 | 2015-01-28 | 北京邮电大学 | Electric energy perception cross-domain virtual network mapping method and system |
-
2015
- 2015-06-19 CN CN201510344602.3A patent/CN104869044B/en active Active
Patent Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102136998A (en) * | 2010-08-30 | 2011-07-27 | 华为技术有限公司 | Traffic engineering and server selection joint optimization method, system and related equipment |
JP2013157855A (en) * | 2012-01-31 | 2013-08-15 | Nec Corp | Method of connecting virtual network, virtual network connection device, and program |
CN103457752A (en) * | 2012-05-30 | 2013-12-18 | 中国科学院声学研究所 | Virtual network mapping method |
CN102904794A (en) * | 2012-09-27 | 2013-01-30 | 北京邮电大学 | Method and device for mapping virtual network |
CN102932479A (en) * | 2012-11-16 | 2013-02-13 | 北京邮电大学 | Virtual network mapping method for realizing topology awareness based on historical data |
CN104320343A (en) * | 2014-10-28 | 2015-01-28 | 北京邮电大学 | Electric energy perception cross-domain virtual network mapping method and system |
Non-Patent Citations (2)
Title |
---|
SEN SU等: "《Energy-Aware Virtual Network Embedding》", 《IEEE/ACM TRANSACTIONS ON NETWORKING》 * |
余建军等: "《基于负载均衡的虚拟网映射随机算法》", 《计算机科学》 * |
Cited By (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN107124287A (en) * | 2016-02-24 | 2017-09-01 | 中移(苏州)软件技术有限公司 | A kind of virtual network service method for quality control and device |
CN107733677A (en) * | 2016-08-12 | 2018-02-23 | 南京中兴软件有限责任公司 | The method for generating alarm and device of a kind of stacking network |
CN107231412A (en) * | 2017-05-22 | 2017-10-03 | 浙江工商大学 | A kind of service path building method mapped based on virtual net |
EP3929781A1 (en) * | 2020-06-25 | 2021-12-29 | Vocalink Limited | Reconstructing a dataset after detection of a network security threat in a network |
US12015620B2 (en) | 2020-06-25 | 2024-06-18 | VocaLink Limited | Reconstructing a dataset after detection of a network security threat in a network |
Also Published As
Publication number | Publication date |
---|---|
CN104869044B (en) | 2019-01-29 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN104466959B (en) | Power system critical circuits discrimination method and system | |
CN104869044A (en) | Mapping method and mapping device for virtual network | |
Agarwal et al. | Parametric and kinetic minimum spanning trees | |
CN109146447A (en) | The transparent sharding method of block chain, apparatus and system | |
CN104298541A (en) | Data distribution algorithm and data distribution device for cloud storage system | |
CN101420355B (en) | Star-cluster double-loop on chip network topology construction | |
CN108053037A (en) | A kind of power distribution network based on two net fusions repairs policy development method and device | |
CN105743763B (en) | A kind of mapping method of virtual network based on study and reasoning | |
CN109728955A (en) | Based on the network node sort method for improving k-shell | |
Fernandes et al. | Minimal spanning trees with a constraint on the number of leaves | |
Camelo et al. | Efficient routing in data center with underlying cayley graph | |
CN101330417A (en) | Quotient space overlay model for calculating network shortest path and building method thereof | |
CN105119833A (en) | Hybrid interconnection structure for network-on-chip, network node encoding method and hybrid routing algorithm thereof | |
CN105634974A (en) | Route determining method and apparatus in software-defined networking | |
CN102780628A (en) | On-chip interconnection network routing method oriented to multi-core microprocessor | |
CN109739433A (en) | The method and terminal device of data processing | |
US9674083B2 (en) | Path calculation order deciding method, program and calculating apparatus | |
CN105987702B (en) | A kind of the section expansion method and its device of navigation routine | |
CN114003775A (en) | Graph data processing and querying method and system | |
CN108521344A (en) | Based on two end reliability estimation method of side expander graphs command and control network | |
CN103428087B (en) | The longest path preset protection P using depth of round first traversal encloses generation method | |
CN109033603A (en) | Secondary system of intelligent substation emulation mode based on source stream path chain | |
JP4408000B2 (en) | Route search method and apparatus | |
CN113572624B (en) | DRAN-based transmission access layer planning method and device | |
CN107026779B (en) | Obtain the method and device of transmission network security evaluation parameter |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
C06 | Publication | ||
PB01 | Publication | ||
EXSB | Decision made by sipo to initiate substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant |