CN110086677A - The configuration method that time-varying is intended in a kind of intention network - Google Patents
The configuration method that time-varying is intended in a kind of intention network Download PDFInfo
- Publication number
- CN110086677A CN110086677A CN201910390467.4A CN201910390467A CN110086677A CN 110086677 A CN110086677 A CN 110086677A CN 201910390467 A CN201910390467 A CN 201910390467A CN 110086677 A CN110086677 A CN 110086677A
- Authority
- CN
- China
- Prior art keywords
- intended
- timeslice
- intention
- time
- configuration
- 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
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L41/00—Arrangements for maintenance, administration or management of data switching networks, e.g. of packet switching networks
- H04L41/08—Configuration management of networks or network elements
- H04L41/0893—Assignment of logical groups to network elements
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L41/00—Arrangements for maintenance, administration or management of data switching networks, e.g. of packet switching networks
- H04L41/08—Configuration management of networks or network elements
- H04L41/0896—Bandwidth or capacity management, i.e. automatically increasing or decreasing capacities
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L41/00—Arrangements for maintenance, administration or management of data switching networks, e.g. of packet switching networks
- H04L41/50—Network service management, e.g. ensuring proper service fulfilment according to agreements
- H04L41/5041—Network service management, e.g. ensuring proper service fulfilment according to agreements characterised by the time relationship between creation and deployment of a service
- H04L41/5051—Service on demand, e.g. definition and deployment of services in real time
-
- 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
- H04L41/50—Network service management, e.g. ensuring proper service fulfilment according to agreements
- H04L41/5041—Network service management, e.g. ensuring proper service fulfilment according to agreements characterised by the time relationship between creation and deployment of a service
- H04L41/5054—Automatic deployment of services triggered by the service manager, e.g. service implementation by automatic configuration of network components
Landscapes
- Engineering & Computer Science (AREA)
- Computer Networks & Wireless Communication (AREA)
- Signal Processing (AREA)
- Data Exchanges In Wide-Area Networks (AREA)
Abstract
The invention discloses time-varying in a kind of intention network to be intended to configuration method, propose a kind of heuritic approach, the algorithm is based on a kind of greedy algorithm (MSP) for maximizing single timeslice configuration, for solving the allocation problem for supporting more tactful time-varying to be intended to, MSP algorithm is mainly made of timeslice Algorithms of Selecting and single timeslice placement algorithm.Using timeslice Algorithms of Selecting according to the sequencing for the total weight size being intended in timeslice is not configured determining the configuration of each timeslice, then single timeslice placement algorithm is used, consider to be intended to while the resource usage amount of entire effective time, the intention in each timeslice is configured.The time-varying that different priorities can be suitable for configuring is intended to, and has the characteristics that allocation plan solution is time-consuming short and it is high to be intended to configuration success rate.
Description
Technical field
The invention belongs to be intended to network technique field, and in particular to a kind of to be intended to the configuration method that time-varying is intended in network.
Background technique
As the research of software defined network (SDN) and network function virtualization (NFV) is gradually goed deep into, network user pair
The demand for simplifying Web vector graphic and management is increasingly urgent to, it is intended that network (IBN) is this to allow user only to need required for informing network
Network objectives, the new network structure that network is automatically realized and safeguarded is increasingly becoming the heat of software defined network area research
Point.
It is intended to a kind of novel network architecture when network, intention can be provided in this network structure for network user and connect
Mouthful, network user wants the network demand realized by being intended to interface expression.The demand of user's informing network itself is also
It is intended to, network is realized automatically and maintains this intention.What user only needs with informing network, and does not have to worry how to go
It realizes.Unified network interface is provided in user side due to being intended to network, further network is abstracted, so that network makes
User, which is able to use, is intended to the more convenient formulation network strategy of language, and by the network side of intention network to various resources and
The constraint of other conditions automatically realizes these network strategies after accounting for.
However it is intended to still suffer from challenge in the design of network: how rationally designs meaning in bottom physical resource limited situation
Figure allocation plan one configures the intention configuration being more intended to.Because be intended to horizontalization be not simply will be expressed by network user
Intention translate into corresponding bottom-layer network language, the flow table injected under SDN environment just completes.Under practical circumstances, bottom
Physical resource be usually all it is limited, how under conditions of resource-constrained, it is as much as possible realize network user intention, and
It is highly important for streamlining the network resource.More, many intentions are all based on group granularity at present, i.e., one is intended to
The not just network node (EP) that is related to but the network node group of multiple such as subscriber's main station network nodes composition
(EPG).Be intended to especially for time-varying, i.e., it is different to meeting in the operation of EPG in different times in intention.For this kind of intention
Configuration, existing allocation plan is simultaneously not perfect, the space that also can be further promoted to the use of bottom physical resource.
It is existing configuration dynamic intention method in, one is Practical Limited state machine to dynamic be intended to, i.e., time-varying be intended to
Intention inside is configured.When being intended to using finite state machine configuration time-varying, it is intended to save state change for each time-varying
Finite state machine make net further according to switching when being intended to a state from its finite state machine and reaching another state
The state of network reconfigures intention.Configuring intention using the mode of finite state machine can be when being intended to state change, in real time
Change option and installment according to network state；Since the number of intention state switching increases when in this way, intention is matched
Setting the number recalculated can also increase, and can consume computing resource；The quantity of intention increases, i.e., is subtracted by the way of being intended to merge
Lack the finite state machine quantity for needing to safeguard, but needs the finite state machine safeguarded that may also can increase.
In another method, in chronological sequence sequence carries out matching for time-varying intention in graph form again in the research having
It sets, configuration is exactly that the mode of all intention pictures is shown together in graph form, to be well understood between intention
Relationship.All time-varying is intended to first to carry out time cost, due to each intention time dimension transform node when it is known
, the mode for dividing timeslice is to take intersection by the state pause judgments time of each intention to divide timeslice.It will draw again later
It is intended to indicate in each timeslice divided in each timeslice figure.Then the timeslice divided since first, when being current
Between intention configured before piece required resource is reserved in future time.Rough process is first time from division
Piece starts, and maximizes the configuration of current time piece, and configuration is effectively anticipated in previous timeslice as far as possible in future time
Figure, until configuring the last one timeslice.In the method, it fixes since first timeslice, so that in the previous time
The intention of configuration configures in current time piece and prevents the intention of the more bigger weights of intention of current time piece from configuring, shadow
The situation for ringing the intention configuration number in entire timeslice occurs；And do not account for time-varying be intended in entire time-domain resource make
With situation, so that going out to be intended to resource occupation influences the case where being intended to configured number in different time piece gap great disparity.
Summary of the invention
It is provided by the invention to be intended to the configuration method solution that time-varying is intended in network for above-mentioned deficiency in the prior art
The problems in above-mentioned background technique.
In order to achieve the above object of the invention, the technical solution adopted by the present invention are as follows: a kind of to be intended to what time-varying in network was intended to
Configuration method, comprising the following steps:
S1, reading and divide Lothrus apterus time-varying be intended to set S1 in timeslice object Tn；
Wherein, time-varying is intended to set S1={ int1, int2, int3 ..., intn }, and inti ∈ S1, i are to be intended to, i=1,
2,3,...,I；
Timeslice object Tn ∈ timeset, and timeset={ T1, T2, T3 ..., TN }, n are under timeslice object
Mark, n=1,2,3 ..., N, timeset are timeslice object set；
The meaning for being intended to set acIntentlist, success configuration to be configured is preserved in each timeslice object Tn
The set of graphs acedIntentlist and intention set notacIntentlist that cannot be configured；
S2, the son intention object subint being intended in each timeslice object is initialized；
Wherein, every height is intended to preserve the EP of intention inti in object subint to object stpair and is intended to id；
S3, each EP initializes object stpair, and passage path selection algorithm is each EP to object
Stpair selects corresponding path, and is stored in corresponding EP in object stpair；
S4, the current select time slices object choose Time and timeslice object queue queue to be configured of setting；And just
Beginningization set Aintents and set Nintents；
Wherein, intentional id of the element in set Aintents for success configuration in processed timeslice object；Collection
Closing the element in Nintents is the intentional id that configures not successfully in processed timeslice object；
S5, judge whether the timeslice object in timeset is fully completed processing；
If so, entering step is S13；
If it is not, then entering step S6；
S6, judge current select time slices object choose Time for sky, and timeslice object queue queue to be configured
Whether set up simultaneously for sky；
If so, entering step S7；
If it is not, then entering step S8；
S7, according to the path saved in stpair, timeslice object is selected in timeset by timeslice Algorithms of Selecting
Tn, and save into choose Time, enter step S9；
S8, the take-off time piece Ti from queue queue, and save into choose Time, enter step S9；
S9, judge whether choose Time configured；
If so, choose Time is emptied, and return step S5；
If it is not, then entering step S10；
S10, according to the path saved in stpair, by single timeslice placement algorithm to the meaning in choose Time
Figure is configured, and updates Aintents and Nintents and acedIntentlist therein and notacIntentlist, and
Mark choose Time be it is configured, enter step S11；
S11, judge that the intention in choose Time timeslice object adjacent thereto was not configured, and choose Time
In acedIntentlist and notacIntentlist timeslice object adjacent thereto in acIntentlist have intersection
Whether set up simultaneously；
If so, entering step S12；
If it is not, choose Time is emptied, and return step S5；
S12, the timeslice object adjacent with choose Time is stored in queue queue, and choose Time is set
Return step S5 after sky；
All processed timeslice objects in S13, traversal timeset, according to each timeslice object
AcedIntentlist and notacdIntentlist, the intention set S2 for configuration of succeeding complete to be intended to time-varying in network
The configuration of intention.
Further, passage path selection algorithm is that each EP selects corresponding road to object stpair in the step S3
The method of diameter specifically:
Link bandwidth in network topology is less than demand by A1, the greatest requirements bandwidth Z according to the intention in timeslice object
The link of bandwidth Z removes, and obtains being all larger than the topology of demand bandwidth Z as the network topology structure of path composition；
A2, diameter d in current network topology structure is determined；
A3, for each intention in timeslice object, determine its corresponding bandwidth demand and of length no more than d of meeting
Path；
A4, according to determining number of paths, after handling it, be stored in corresponding EP in object stpair.
Further, the step A4 specifically:
When determining number of paths is less than threshold value k and is greater than 0, determining path is stored directly in corresponding EP to right
As in stpair, as the input data in timeslice Algorithms of Selecting and single timeslice placement algorithm；
When determining number of paths is 0, the sub- intention is set without path tag to object stapir in corresponding EP；
When determining number of paths is greater than threshold value k, 01 mask that determining all coordinates measurements are random takes out wherein
Mask is 1 corresponding path, is stored in corresponding EP in object stpair, as timeslice Algorithms of Selecting and single timeslice
Input data in placement algorithm.
Further, select timeslice object Tn's in timeset by timeslice Algorithms of Selecting in the step S7
Method specifically:
B1, the timeslice object for selecting all configurations not yet in timeset；
Whether the quantity for the timeslice object being not configured that B2, judgement are selected is greater than threshold value j；
If so, entering step B3；
If it is not, then entering step B4；
B3, when each timeslice object of taking-up being sorted from large to small according to the total weight being wherein intended to, and taking out j first
Between piece object, enter step S4；
B4, it is determined in each timeslice object using timeslice Object Selection solving model and configures the quantity of intention, and selected
Configuration is intended to the most timeslice object Tn of quantity out, and timeslice object Tn is selected in completion in timeset.
Further, the timeslice Object Selection solving model in the step B4 are as follows:
max∑i∈ints(t)Wi×Ii
Wherein,
In formula, WiFor the weight for being intended to i；
IiIt is intended to the 0-1 variable of i for characterization, is intended to i for 1 and successfully configures, is 0, being then intended to i not can configure；
Ints (t) is effectively to be all intended in time period t；
Eps (i) is all EP pairs be intended in i；
Ki,jIt is intended to the EP that needs to configure in i to the 0-1 variable of j for characterization, is the EP to successfully configuring for 1；
PATi,jTo be intended to the EP that can be configured in i to all paths of j；
pi,j,kMeet the 0-1 variable for being intended to the kth paths of EPj in i for characterization, be to select the path for 1, is 0
The path is not selected；
Path (l) is all paths by link l；
BWi,tTo be intended to the bandwidth demand that i is configured in time period t；
CAPl,tFor bandwidth capacity of the link l in time period t；
Links is all links in current network；
TP is set of all periods.
Further, side choose Time configured by single timeslice placement algorithm in the step S10
Method specifically:
C1, judge in the corresponding intention id of intention in the acIntentlist of choose Time with it is adjacent processed when
Between piece object notacIntentlist in intention corresponding be intended to whether have identical intention id in id；
If so, entering step C2；
If it is not, then entering step C3；
C2, it is removed and the notacIntentlist of adjacent time piece object in the acIntentlist of choose Time
In the corresponding intention of identical intention id, enter step C3；
C3, judge in the corresponding intention id of intention in the acIntentlist of choose Time with it is adjacent processed when
Between intention in acedIntentlist in piece object is corresponding is intended to whether have identical intention id in id；
If so, entering step C4；
If it is not, then entering step C5；
C4, it is marked in the acIntentlist of choose Time and adjacent processed timeslice object
The corresponding intention of identical intention id, enters step C5 in acedIntentlist；
C5, intention is not configured in the current acIntentlist of choose Time by being intended to configuration reasoning model
It is configured, completes to configure the intention of choose Time.
Further, the intention configuration reasoning model in the step C5 are as follows:
Wherein,
In formula, WiFor the weight for being intended to i；
WjFor the weight for being intended to j；
α is resource coefficient of utilization；
RiTo be intended to the bandwidth resources that i is used；
RqTo be intended to the bandwidth resources that q is used；
IiIt is intended to the 0-1 variable of i for characterization, is intended to i for 1 and successfully configures, is 0, being then intended to i not can configure；
To be intended to preferential configuration variables εiCoefficient；
εiTo characterize the increased 0-1 variable for being intended to preferential configuration variables, does not need to set 1 when increasing intention i configuration, need
Increase and is intended to set 0 when i configuration；
Ints (t) is effectively to be all intended in time period t；
Conf (t) is that the son that part configuration is intended in time period t is intended to；
Eps (i) is all EP pairs be intended in i；
Ki,jIt is intended to the EP that needs to configure in i to the 0-1 variable of j for characterization, is the EP to successfully configuring for 1；
PATi,jTo be intended to the EP that can be configured in i to all paths of j；
pi,j,kMeet the 0-1 variable for being intended to the kth paths of EPj in i for characterization, be to select the path for 1, is 0
The path is not selected；
Path (l) is all paths by link l；
BWi,tTo be intended to the bandwidth demand that i is configured in time period t；
CAPl,tFor bandwidth capacity of the link l in time period t；
Links is all links in current network；
TP is set of all periods.
Further, the step S13 specifically:
All processed timeslice objects in timeset are traversed, it will be in all timeslice objects
AcedIntentlist takes its union to form set C1, and the notacdIntentlist in all timeslice objects takes union to form
Set C2, taking the intersection of set C1 and set C2 is C3, and set C1 subtracts the intention after set C3 and forms the intention successfully configured
Set S2.
The invention has the benefit that
(1) time-varying for being suitable for configuring different priorities is intended to.The MSP algorithm proposed in the present invention, which may make, successfully to be configured
Intention weight and maximization, make to be intended to priority corresponding with weight is intended to, therefore the applicable different priorities of this method are weighed
The intention configuration of weight, and the intention for making priority high configures as far as possible.
(2) allocation plan solves time-consuming short.In the method for the present invention, filtered before being configured to intention in each timeslice
Operation, so that the intention for participating in configuration in each timeslice is reduced, the time is reduced compared with existing configuration method.
(3) it is high to be intended to configuration success rate.Since algorithm proposed by the present invention considers the weight and time-varying meaning of time-varying intention
Resource consumption of the figure on its effective time can rationally obtain configuration method under the constraint for meeting physical resource, mutually sympathize with
Configuration success rate increases compared with existing configuration method under condition.
Detailed description of the invention
Fig. 1 is to be intended to time-varying in network in the present invention to be intended to configuration method flow chart.
Fig. 2 is the routing algorithm flow chart in the present invention.
Fig. 3 is the timeslice Algorithms of Selecting flow chart in the present invention.
Fig. 4 is the single timeslice placement algorithm flow chart in the present invention.
Specific embodiment
A specific embodiment of the invention is described below, in order to facilitate understanding by those skilled in the art this hair
It is bright, it should be apparent that the present invention is not limited to the ranges of specific embodiment, for those skilled in the art,
As long as various change is in the spirit and scope of the present invention that the attached claims limit and determine, these variations are aobvious and easy
See, all are using the innovation and creation of present inventive concept in the column of protection.
The technical problem to be solved in the present invention are as follows: to expression be intended to connection, by network service function it is related to QoS
Time-varying be intended to and known bottom-layer network is resource-constrained, is intended to granularity be group granularity under the premise of, configuration time-varying is intended to, and to the greatest extent may be used
The configuration quantity that time-varying can be made to be intended to is more.For expression connectivity and by the intention of network service function chain type come
It says, the allocation problem that time-varying is intended to substantially is that whether can find out path within each network physical resource-constrained period,
It will be intended to the problem of realizing.This method comprehensively considers each intention occupation condition in entire effective time, is entirely having
It imitates in the time and is intended to using maximizing intention configuration quantity and being configured as target, as far as possible the configuration quantity of configuration time-varying intention.
For the configuration quantity that the maximization in entire time-domain is intended to, therefore following 0-1 integer law optimization is set
Model:
max∑n∈TP∑i∈ints(n)Wi×Ii,n (1)
Wherein,
In formula, n is any one timeslice in the TP period, the corresponding ∑ W as i ∈ ints (n)i×IiFor in the time
The product of the 0-1 variable of the weight and intention i of intention i in piece object n；
WiFor the weight for being intended to i；
Ii,nBe intended to the 0-1 variable that whether configures in the live part of timeslice object n of i for characterization, be configured to 1, otherwise for
0；
Ints (n) is effectively to be all intended in timeslice n；
Eps (i) is all EP pairs be intended in i；
Ki,jIt is intended to the EP that needs to configure in i to the 0-1 variable of j for characterization, is the EP to successfully configuring for 1；
PATi,jTo be intended to the EP that can be configured in i to all paths of j；
pi,j,kMeet the 0-1 variable for being intended to the kth paths of EPj in i for characterization, be to select the path for 1, is 0
The path is not selected；
Path (l) is all paths by link l；
BWi,nTo be intended to the bandwidth demand that i is configured in time period t；
CAPl,nFor bandwidth capacity of the link l in time period t；
Links is all links in current network；
TP is set of all periods.
In constraint condition (2), in order to increase the association that time-varying is intended to each period, needed to increase a upper period pair
The influence of the period, i.e., the intention of the successful configuration in a upper timeslice, and still had in current time piece
Intention will also guarantee successfully to configure；The live part for the intention configured that cannot succeed in a upper timeslice, in current time piece
In it is also effective when, do not need to be configured with again for it.Formula (3) simultaneously for each period intention, successfully configuration be exactly
It is required that each of intention EP is to can successfully configure down.Exist certainly for each EP to the root that can be configured
In whether the path met the requirements can be found.Therefore the constraint such as (6) to each EP pairs.It is limited in bottom physical resource
In the case of, it is also necessary to consider the constraint of link bandwidth, i.e., for each link, the usage amount of bandwidth cannot be more than every
The bandwidth capacity of a link, as shown in formula (8), the sum of all path bandwidths for having used link l are less than chain in the period
The bandwidth capacity CAP of road liIt should be noted that since there are the relevant intention of SFC, different service function of the EP to process
Bandwidth can be different after energy, therefore the bandwidth demand of each intention is a band fat vector, the p in formulai,j,kCorresponding bandwidth
It is intended to i bandwidth corresponding on link l.
The allocation problem of intention, the problem of resource allocation class is belonged to from the point of view of essence, the resource allocation class of most of complexity
The problem of be all np difficulty problem.In addition, by objective optimization model (1) it is found that being intended to same in different time periods effective
In the case where being intended to difference, the constraint of bandwidth resources is existed simultaneously, it is desirable that solve time-varying in entire continuous time and be intended to configured number
Max problem also belong to np difficulty problem, be difficult to acquire optimal solution in polynomial time.Meanwhile time-varying is intended to placement algorithm
Main purpose be the intention number for maximizing configuration and being intended in each timeslice of number algorithm, but the time the piece number divided
Amount can have an immense impact on to the solution time of more timeslice combined optimization problems.Therefore proposed by the present invention is a kind of heuristic
Algorithm, the algorithm are based on a kind of greedy algorithm (the Maximize Single for maximizing single period configuration
TimePeriod, MSP), for solving the allocation problem for supporting more policy intents.
MSP algorithm mainly by timeslice Algorithms of Selecting (TimePeriod Select Algorithm, TPS) and individually
Timeslice placement algorithm (Complie in single TimePeriod, CST) algorithm composition.It is pressed using timeslice Algorithms of Selecting
According to the sequencing for the total weight size being intended in timeslice is not configured determining the configuration of each timeslice, single timeslice is then used
Placement algorithm considers to be intended to while the resource usage amount of entire effective time, match to the intention in each timeslice
It sets.In single timeslice placement algorithm, in order to embody the relationship of same intention live part in different time piece, the present invention
It is introduced into and is intended to preferred number for controlling in the intention live part of other times piece configuration to the intention in current time piece
The influence of live part；Resource is introduced into configure intention using resource usage amount of the coefficient of discharge reflection intention in its effective time
Influence.
In order to realize above-mentioned Optimized model, it is intended to the present invention provides time-varying in a kind of intention network as shown in Figure 1
Configuration method, comprising the following steps:
S1, reading and divide Lothrus apterus time-varying be intended to set S1 in timeslice object Tn；
Wherein, time-varying is intended to set S1={ int1, int2, int3 ..., intn }, and inti ∈ S1, i are to be intended to, i=1,
2,3,...,I；
Timeslice object Tn ∈ timeset, and timeset={ T1, T2, T3 ..., TN }, n are under timeslice object
Mark, n=1,2,3 ..., N, timeset are timeslice object set；
The live part being intended in the time is saved in a timeslice object, i.e., son is intended to.Such as: it is intended to have in set
It is intended to a, it is intended that b, it is intended that c；It is effective in timeslice 1,2 to be intended to a, it is intended that b is effective in timeslice 2,3, it is intended that c has in timeslice 3
Effect then saves in timeslice object 1 and is intended to a in the live part (son for being intended to a is intended to) of timeslice 1, timeslice pair
It is intended to b as saving in 2, it is intended that (son for being intended to b is intended to a, it is intended that a son meaning of a in the live part of timeslice 2
Figure), the live part for being intended to c is saved in timeslice object 3 (son for being intended to c is intended to).
The meaning for being intended to set acIntentlist, success configuration to be configured is preserved in each timeslice object Tn
The set of graphs acedIntentlist and intention set notacIntentlist that cannot be configured；
The duration of the period is preserved in each timeslice object；
S2, the son intention object subint being intended in each timeslice object is initialized；
Wherein, every height is intended to preserve the EP of intention inti in object subint to object stpair, intention id, son
It is intended to corresponding bandwidth demand list bdlist and specific behavior list actionlist；
S3, each EP initializes object stpair, and passage path selection algorithm is each EP to object
Stpair selects corresponding path, and is stored in corresponding EP in object stpair；
S4, the current select time slices object choose Time and timeslice object queue queue to be configured of setting；And just
Beginningization set Aintents and set Nintents；
Wherein, intentional id of the element in set Aintents for success configuration in processed timeslice object；Collection
Closing the element in Nintents is the intentional id that configures not successfully in processed timeslice object；
S5, judge whether the timeslice object in timeset is fully completed processing；
If so, entering step is S13；
If it is not, then entering step S6；
S6, judge current select time slices object choose Time for sky, and timeslice object queue queue to be configured
Whether set up simultaneously for sky；
If so, entering step S7；
If it is not, then entering step S8；
S7, according to the path saved in stpair, timeslice object is selected in timeset by timeslice Algorithms of Selecting
Tn, and save into choose Time, enter step S9；
S8, the take-off time piece Ti from queue queue, and save into choose Time, enter step S9；
S9, judge whether choose Time configured；
If so, choose Time is emptied, and return step S5；
If it is not, then entering step S10；
S10, according to the path saved in stpair, by single timeslice placement algorithm to the meaning in choose Time
Figure is configured, and updates Aintents and Nintents and acedIntentlist therein and notacIntentlist, and
Mark choose Time be it is configured, enter step S11；
S11, judge that the intention in choose Time timeslice object adjacent thereto was not configured, and choose Time
In acedIntentlist and notacIntentlist timeslice object adjacent thereto in acIntentlist have intersection
Whether simultaneously no establishment；
If so, entering step S12；
If it is not, choose Time is emptied, and return step S5；
S12, the timeslice object adjacent with choose Time is stored in queue queue, and choose Time is set
Return step S5 after sky；
All processed timeslice objects in S13, traversal timeset, according to each timeslice object
AcedIntentlist and notacdIntentlist, the intention set S2 for configuration of succeeding complete to be intended to time-varying in network
The configuration of intention.
In above-mentioned steps S3, due to using each EP pairs of path as the variable in model, when topological scale and topology
When the degree of communication of node is higher or when EP pairs of quantity increases, path variable can be sharply increased, and can thus seriously affect mould
The time that type solves.Therefore, path Algorithms of Selecting (Random-Path-Select Alogrithm, RPS) is used in the present invention,
Appropriate reduce participates in EP pairs of the number of paths that algorithm calculates.
Therefore, as shown in Fig. 2, passage path selection algorithm is the side that each EP selects corresponding path to object stpair
Method specifically:
Link bandwidth in network topology is less than demand by A1, the greatest requirements bandwidth Z according to the intention in timeslice object
The link of bandwidth Z removes, and obtains being all larger than the topology of demand bandwidth Z as the network topology structure of path composition；
A2, diameter d in current network topology structure is determined；
A3, for each intention in timeslice object, determine its corresponding bandwidth demand and the road of of length no more than d
Diameter；
In addition to meeting above-mentioned bandwidth demand, specific particular service functionality is also needed to meet；
A4, according to determining number of paths, after handling it, be stored in corresponding EP in object stpair.
Step A4 specifically:
When determining number of paths is less than threshold value k and is greater than 0, determining path is stored directly in corresponding EP to right
As in stpair, as the input data in timeslice Algorithms of Selecting and single timeslice placement algorithm；
When determining number of paths is 0, the sub- intention is set without path tag to object stapir in corresponding EP；
When determining number of paths is greater than threshold value k, 01 mask that determining all coordinates measurements are random takes out wherein
Mask is 1 corresponding path, is stored in corresponding EP in object stpair, as timeslice Algorithms of Selecting and single timeslice
Input data in placement algorithm.
Each EP pairs of path is the shortest path algorithm using k long in routing algorithm RPS, and length ensure that path
Of length no more than d, in this way can be too long to avoid the path found out and the case where there are loops.After selecting path, to each
EP is to random 01 mask is generated, and wherein the number in mask for 1 is k, and the algorithm for next step that k value is namely chosen calculates
Path.The mode randomly selected can preferably make multiple EP pairs of path not select identical link, cause single chain
Road becomes bottleneck link, thus the successful configured number being intended in influence time piece object.
Timeslice Algorithms of Selecting in above-mentioned steps S7 is the embodiment for the Greedy strategy that time-varying is intended to placement algorithm.It needs to infuse
Meaning is intended to the beginning of placement algorithm in time-varying, has carried out to intention live part to be configured in each timeslice pre-
Processing, i.e., the intention in each period, the intention that the number of path obtained in the routing algorithm is 0 is included in advance can not
In the set of configuration.When the period being not configured is according to weight sequencing is intended to, the intention for being 0 without actual number of path
It participates in.It is namely not in j most intention of the intention weight selected in algorithm, the intention number configured that can succeed is 0
The case where, it can guarantee that algorithm can be carried out correctly in this way
Therefore, as shown in figure 3, selecting the method tool of timeslice object Tn in timeset by timeslice Algorithms of Selecting
Body are as follows:
B1, the timeslice object for selecting all configurations not yet in timeset；
Whether the quantity for the timeslice object being not configured that B2, judgement are selected is greater than threshold value j；
If so, entering step B3；
If it is not, then entering step B4；
B3, when each timeslice object of taking-up being sorted from large to small according to the total weight being wherein intended to, and taking out j first
Between piece object, enter step S4；
B4, it is determined in each timeslice object using timeslice Object Selection solving model and configures the quantity of intention, and selected
Configuration is intended to the most timeslice object Tn of quantity out, and timeslice object Tn is selected in completion in timeset.
Intention in TPS algorithm is intended to the configuration unity of thinking of configuration main flow with time-varying when configuring, that is, realizes maximum
The intention of change configures, therefore the timeslice Object Selection solving model in step B4 are as follows:
max∑i∈ints(t)Wi×Ii (9)
Wherein,
In formula, WiFor the weight for being intended to i；
IiIt is intended to the 0-1 variable of i for sign, is intended to i for 1 and successfully configures, is 0, being then intended to i not can configure；
Ints (t) is effectively to be all intended in time period t；
Eps (i) is all EP pairs be intended in i；
Ki,jIt is intended to the EP that needs to configure in i to the 0-1 variable of j for characterization, is the EP to successfully configuring for 1；
PATi,jTo be intended to the EP that can be configured in i to all paths of j；
pi,j,kMeet the 0-1 variable for being intended to the kth paths of EPj in i for characterization, be to select the path for 1, is 0
The path is not selected；
Path (l) is all paths by link l；
BWi,tTo be intended to the bandwidth demand that i is configured in time period t；
CAPl,tFor bandwidth capacity of the link l in time period t；
Links is all links in current network；
TP is set of all periods.
The solver model can range Zero-one integer programming problem, have very for Zero-one integer programming problem at present
More mature method for solving, such as branch and bound method and cutting plane algorithm and the implicit enumeration of one-zero programming etc..Therefore it is adopted in the present invention
With the solver of modern optimization problem, solution can be found out faster.
Placement algorithm CST is when the resource usage amount of intention is included in one in single timeslice in above-mentioned steps S10
Between in piece to the coverage for being intended to configuration.When configuring to the intention live part in single timeslice, there can be this
The problem of sample: first is that after option and installment timeslice starting point, how to guarantee the intention being configured in its effective other times
Still configuration is gone down in piece；Second is that due to time-varying be intended to time variation, the bandwidth demand in timeslice 1 readily satisfy but when
Between bandwidth demand in piece 2 be not able to satisfy, how to be configured in single timeslice consider when being intended to the intention be intended to it is effective
The case where resource requirement in entire timeslice, it is determined which is configured to be intended to.
For above-mentioned two problems, the invention proposes the placement algorithm being intended in single timeslice, to list in algorithm
The allocation problem of a timeslice re-establishes model.The main purpose for being intended to placement algorithm in single timeslice is improved at it
The configured intention of live part is intended to the configuration preference level of live part in timeslice to be processed in his timeslice, simultaneously
When configuring intention in each timeslice, the resource service condition being intended in all effective times is accounted for.Below
It is the flow chart of CST algorithm, wherein the timeslice of a upper configuration referred in the time that current time piece is put into queue queue
Piece, while current time piece is put into queue queue, accordingly at that time between succeeded in piece of configuration be intended to collection
The son intention set closed and cannot configured also saves together, can not only mark when configuring future time piece so excellent
The intention first configured can also directly remove the intention that cannot be configured.
Therefore, as shown in figure 4, the method configured to choose Time by single timeslice placement algorithm is specific
Are as follows:
C1, judge in the corresponding intention id of intention in the acIntentlist of choose Time with it is adjacent processed when
Between piece object notacIntentlist in intention corresponding be intended to whether have identical intention id in id；
If so, entering step C2；
If it is not, then entering step C3；
C2, it is removed and the notacIntentlist of adjacent time piece object in the acIntentlist of choose Time
In the corresponding intention of identical intention id, enter step C3；
C3, judge in the corresponding intention id of intention in the acIntentlist of choose Time with it is adjacent processed when
Between intention in acedIntentlist in piece object is corresponding is intended to whether have identical intention id in id；
If so, entering step C4；
If it is not, then entering step C5；
C4, it is marked in the acIntentlist of choose Time and adjacent processed timeslice object
The corresponding intention of identical intention id, enters step C5 in acedIntentlist；
C5, intention is not configured in the current acIntentlist of choose Time by being intended to configuration reasoning model
It is configured, completes to configure the intention of choose Time.
Intention configuration reasoning model in above-mentioned steps C5 are as follows:
Wherein,
In formula, WiFor the weight for being intended to i；
WjFor the weight for being intended to j；
α is resource coefficient of utilization；
RiTo be intended to the bandwidth resources that i is used；
RqTo be intended to the bandwidth resources that q is used；
IiIt is intended to the 0-1 variable of i for characterization, is intended to i for 1 and successfully configures, is 0, being then intended to i not can configure；
To be intended to preferential configuration variables εiCoefficient；
εiTo characterize the increased 0-1 variable for being intended to preferential configuration variables, does not need to set 1 when increasing intention i configuration, need
Increase and is intended to set 0 when i configuration；
Ints (t) is effectively to be all intended in time period t；
Conf (t) is that the son that part configuration is intended in time period t is intended to；
Eps (i) is all EP pairs be intended in i；
Ki,jIt is intended to the EP that needs to configure in i to the 0-1 variable of j for characterization, is the EP to successfully configuring for 1；
PATi,jTo be intended to the EP that can be configured in i to all paths of j；
pi,j,kMeeting EP in intention i for characterization, to the 0-1 variable of the kth paths of j, is to select the path for 1, is 0
The path is not selected then；
Path (l) is all paths by link l；
BWi,tTo be intended to the bandwidth demand that i is configured in time period t；
CAPl,tFor bandwidth capacity of the link l in time period t；
Links is all links in current network；
TP is set of all periods.
Target formula (1), RiIt is intended to the set of the resource consumption where it in timeslice object of i, is set in the present invention
Determine resource consumption and refer to the product for being intended to the time of size and occupied bandwidth in each timeslice object, it may be assumed that
RiFor the sum of products for being intended to i all EP pairs of bandwidth demand and time span within each period, communicating
In field, people using the product of bandwidth and time delay as one of index of network performance, expression be data link ability.With
The i.e. product of Occupation time length and the amount of bandwidth used can then be expressed as using in the resource of the period in intention
Situation.The resource service condition of multiple periods is added as resource usage amount of the intention in the entire period again.Cause
This present invention indicates that the intention to the resource occupation of entire period, avoids the occurrence of intention and exists using the resource usage amount being intended to
The situation of the bandwidth resources gap great disparity used in its effective period.Such as, it is intended that A is all EP pairs in timeslice 1
Bandwidth demand is 10Mbps but bandwidth demand is 100Mbps in its effective timeslice 3,4,5, in this way in setup time piece 1
In intention when, successfully configuration be intended to A but be intended to A other effective time pieces in be but difficult to complete to configure.Such
In the case of, the resource consumption situation for increasing intention can increase the validity for being intended to configuration.
ε in target formula (16)iIt is newly-increased artificial variables, that is, is intended to preferential configuration variables, for makes area to intention
Point.For the son in the timeslice needed to configure is intended to, it is understood that there may be two seeds are intended to, and one is the intentions configured for the first time
The effective part in current time piece, is referred to as the intention first appeared；One is be configured in other times piece
Part of the intention in current time piece, the referred to as configured intention in part.It is only all effective where it when being intended to
It is all successfully configured in timeslice and is just intended to complete configuration.Therefore, in order to enable part it is configured intention as far as possible
It can also be configured down in current time piece, need to increase " priority " that these parts configured intention.
In this case, if force configured the intentions variable I in these parts for making force to be equal to 1 as by force
Constraint exist, and can prevent the period configuration intention number from reaching most.Based on this, the present invention uses formula (17)
Soft-constraint allows the configured intention in part to complete to configure as far as possible: for for the constraint of the intention first appeared in formula,
It is 0 on the right of equation, Ii and ε when not being able to satisfy the intention, in levoformiIt is all 0.When being 1 on the right of equation, Ii and εiIn
Can only have a variate-value is that the variable weight in 1. pairs of optimization aims is reasonably set, willValue setting is much smaller than IiWeight,
Under the premise of ensuring that maximization is intended to configuration, it is intended that variable 1, it is intended that preferential configuration variables are 0.Part is configured
The intention crossed, constraint remain as formula (18), i.e. the intention for part configuration, if its variable IiIt is 1, ε at this timeiWith IiBecome
It is not constrained between amount, then the value in optimization aim the big just needs εiWith IiAll it is being intended to the identical situation of weight in this way for 1.
Under, if the configured intention in part and the intention that first appears can only configure one, can preferential selected section it is configured
It is intended to, that is, it is different " priority " realizes the intention that part configuration is intended to and first appears.
In order to enable being intended to configuration to maximize to be primary optimization aim in optimization aim formula (16), it is intended that weight is main
Factor, it is necessary to which the coefficient setting of resource usage amount is much smaller than the coefficient for being intended to weight.It is used due to intention weight and resource
The order of magnitude is different, then needs that weight and resource usage amount is normalized in optimization aim, to be currently intended to weight
Weight of the ratio as the intention that timeslice is intended to total weight is accounted for, is accounted in current time piece with the resource usage amount being currently intended to
Resource usage amount of the usage amount weight proportion of intention as the intention.
Other constraint conditions are consistent with the constraint condition of solver in timeslice selection algorithm.
In above-mentioned steps S13 specifically:
All processed timeslice objects in timeset are traversed, it will be in all timeslice objects
AcedIntentlist takes its union to form set C1, and the notacdIntentlist in all timeslice objects takes union to form
Set C2, taking the intersection of set C1 and set C2 is C3, and set C1 subtracts the intention after set C3 and forms the intention successfully configured
Set S2.
The invention has the benefit that
(1) time-varying for being suitable for configuring different priorities is intended to.The MSP algorithm proposed in the present invention, which may make, successfully to be configured
Intention weight and maximization, make to be intended to priority corresponding with weight is intended to, therefore the applicable different priorities of this method are weighed
The intention configuration of weight, and the intention for making priority high configures as far as possible.
(2) allocation plan solves time-consuming short.In the method for the present invention, filtered before being configured to intention in each timeslice
Operation, so that the intention for participating in configuration in each timeslice is reduced, the time is reduced compared with existing configuration method.
(3) it is high to be intended to configuration success rate.Since algorithm proposed by the present invention considers the weight and time-varying meaning of time-varying intention
Resource consumption of the figure on its effective time can rationally obtain configuration method under the constraint for meeting physical resource, mutually sympathize with
Configuration success rate increases compared with existing configuration method under condition.
Claims (8)
1. a kind of be intended to the configuration method that time-varying is intended in network, which comprises the following steps:
S1, reading and divide Lothrus apterus time-varying be intended to set S1 in timeslice object Tn；
Wherein, time-varying is intended to set S1={ int1, int2, int3 ..., intn }, and inti ∈ S1, i are to be intended to, i=1, and 2,
3,...,I；
Timeslice object Tn ∈ timeset, and timeset={ T1, T2, T3 ..., TN }, n are the subscript of timeslice object, n
=1,2,3 ..., N, timeset be timeslice object set；
The intention collection for being intended to set acIntentlist, success configuration to be configured is preserved in each timeslice object Tn
The intention set notacIntentlist that closes acedIntentlist and cannot configure；
S2, the son intention object subint being intended in each timeslice object is initialized；
Wherein, every height is intended to preserve the EP of intention inti in object subint to object stpair and is intended to id；
S3, each EP initializes object stpair, and passage path selection algorithm is each EP to object stpair
Corresponding path is selected, and is stored in corresponding EP in object stpair；
S4, the current select time slices object choose Time and timeslice object queue queue to be configured of setting；And it initializes
Set Aintents and set Nintents；
Wherein, intentional id of the element in set Aintents for success configuration in processed timeslice object；Set
Element in Nintents is the intentional id that configures not successfully in processed timeslice object；
S5, judge whether the timeslice object in timeset is fully completed processing；
If so, entering step is S13；
If it is not, then entering step S6；
S6, judge current select time slices object choose Time for sky, and timeslice object queue queue to be configured be sky
Whether set up simultaneously；
If so, entering step S7；
If it is not, then entering step S8；
S7, according to the path saved in stpair, timeslice object Tn is selected in timeset by timeslice Algorithms of Selecting,
And save into choose Time, enter step S9；
S8, the take-off time piece Ti from queue queue, and save into choose Time, enter step S9；
S9, judge whether choose Time configured；
If so, choose Time is emptied, and return step S5；
If it is not, then entering step S10；
S10, according to the path saved in stpair, by single timeslice placement algorithm to the intention in choose Time into
Row configuration, and Aintents and Nintents and acedIntentlist therein and notacIntentlist are updated, and mark
Choose Time be it is configured, enter step S11；
S11, judge that the intention in choose Time timeslice object adjacent thereto was not configured, and in choose Time
Whether acIntentlist in timeslice object acedIntentlist and notacIntentlist adjacent thereto has intersection
It sets up simultaneously；
If so, entering step S12；
If it is not, choose Time is emptied, and return step S5；
S12, by adjacent with choose Time timeslice object deposit queue queue, and after choose Time is emptied
Return step S5；
All processed timeslice objects in S13, traversal timeset, according to each timeslice object
AcedIntentlist and notacdIntentlist, the intention set S2 for configuration of succeeding complete to be intended to time-varying in network
The configuration of intention.
2. according to claim 1 be intended to the configuration method that time-varying is intended in network, which is characterized in that in the step S3
Passage path selection algorithm is the method that each EP selects corresponding path to object stpair specifically:
Link bandwidth in network topology is less than demand bandwidth by A1, the greatest requirements bandwidth Z according to the intention in timeslice object
The link of Z removes, and obtains being all larger than the topology of demand bandwidth Z as the network topology structure of path composition；
A2, diameter d in current network topology structure is determined；
A3, for each intention in timeslice object, determine its corresponding road for meeting bandwidth demand and of length no more than d
Diameter；
A4, according to determining number of paths, after handling it, be stored in corresponding EP in object stpair.
3. according to claim 2 be intended to the configuration method that time-varying is intended in network, which is characterized in that the step A4 tool
Body are as follows:
When determining number of paths is less than threshold value k and is greater than 0, determining path is stored directly in corresponding EP to object
In stpair, as the input data in timeslice Algorithms of Selecting and single timeslice placement algorithm；
When determining number of paths is 0, the sub- intention is set without path tag to object stapir in corresponding EP；
When determining number of paths is greater than threshold value k, by 01 mask that determining all coordinates measurements are random, wherein mask is taken out
For 1 corresponding path, corresponding EP is stored in in object stpair, is configured as timeslice Algorithms of Selecting and single timeslice
Input data in algorithm.
4. according to claim 1 be intended to the configuration method that time-varying is intended in network, which is characterized in that in the step S7
The method that timeslice object Tn is selected in timeset by timeslice Algorithms of Selecting specifically:
B1, the timeslice object for selecting all configurations not yet in timeset；
Whether the quantity for the timeslice object being not configured that B2, judgement are selected is greater than threshold value j；
If so, entering step B3；
If it is not, then entering step B4；
B3, each timeslice object of taking-up is sorted from large to small according to the total weight being wherein intended to, and takes out preceding j timeslice
Object enters step S4；
B4, it is determined in each timeslice object using timeslice Object Selection solving model and configures the quantity of intention, and selected and match
It sets and is intended to the most timeslice object Tn of quantity, timeslice object Tn is selected in completion in timeset.
5. according to claim 4 be intended to the configuration method that time-varying is intended in network, which is characterized in that in the step B4
Timeslice Object Selection solving model are as follows:
max∑i∈ints(t)Wi×Ii
Wherein, Ii∈ { 0,1 },
Ii=Ki,1∩Ki,2...Ki,m,
Ki,j∈ { 0,1 },
pi,j,k∈ { 0,1 },
In formula, WiFor the weight for being intended to i；
IiIt is intended to the 0-1 variable of i for characterization, is intended to i for 1 and successfully configures, is 0, being then intended to i not can configure；
Ints (t) is effectively to be all intended in time period t；
Eps (i) is all EP pairs be intended in i；
Ki,jIt is intended to the EP that needs to configure in i to the 0-1 variable of j for characterization, is the EP to successfully configuring for 1；
PATi,jTo be intended to the EP that can be configured in i to all paths of j；
pi,j,kMeet the 0-1 variable for being intended to the kth paths of EPj in i for characterization, is to select the path for 1, is not selected then for 0
Select the path；
Path (l) is all paths by link l；
BWi,tTo be intended to the bandwidth demand that i is configured in time period t；
CAPl,tFor bandwidth capacity of the link l in time period t；
Links is all links in current network；
TP is set of all periods.
6. according to claim 1 be intended to the configuration method that time-varying is intended in network, which is characterized in that the step S10
In choose Time is configured by single timeslice placement algorithm method specifically:
C1, judge in the corresponding intention id of intention in the acIntentlist of choose Time with adjacent processed timeslice
Whether there is identical intention id in the corresponding intention id of intention in the notacIntentlist of object；
If so, entering step C2；
If it is not, then entering step C3；
C2, removal and phase in the notacIntentlist of adjacent time piece object in the acIntentlist of choose Time
With the corresponding intention of id is intended to, C3 is entered step；
C3, judge in the corresponding intention id of intention in the acIntentlist of choose Time with adjacent processed timeslice
Whether there is identical intention id in the corresponding intention id of intention in acedIntentlist in object；
If so, entering step C4；
If it is not, then entering step C5；
C4, it is marked in the acIntentlist of choose Time and adjacent processed timeslice object
The corresponding intention of identical intention id, enters step C5 in acedIntentlist；
C5, the intention that is not configured in the current acIntentlist of choose Time is carried out by being intended to configuration reasoning model
Configuration is completed to configure the intention of choose Time.
7. according to claim 6 be intended to the configuration method that time-varying is intended in network, which is characterized in that in the step C5
Intention configuration reasoning model are as follows:
Wherein, Ii+εi=Ki,1∩Ki,2...Ki,m,
Ii=Ki,1∩Ki,2...Ki,m,
Ii∈ { 0,1 },
Ki,j∈ { 0,1 },
pi,j,k∈ { 0,1 },
εi∈ { 0,1 },
In formula, WiFor the weight for being intended to i；
WjFor the weight for being intended to j；
α is resource coefficient of utilization；
RiTo be intended to the bandwidth resources that i is used；
RqTo be intended to the bandwidth resources that q is used；
IiIt is intended to the 0-1 variable of i for characterization, is intended to i for 1 and successfully configures, is 0, being then intended to i not can configure；
To be intended to preferential configuration variables εiCoefficient；
εiTo characterize the increased 0-1 variable for being intended to preferential configuration variables, does not need to set 1 when increasing intention i configuration, need to increase
It is intended to set 0 when i configuration；
Ints (t) is effectively to be all intended in time period t；
Conf (t) is that the son that part configuration is intended in time period t is intended to；
Eps (i) is all EP pairs be intended in i；
Ki,jIt is intended to the EP that needs to configure in i to the 0-1 variable of j for characterization, is the EP to successfully configuring for 1；
PATi,jTo be intended to the EP that can be configured in i to all paths of j；
pi,j,kMeet the 0-1 variable for being intended to the kth paths of EPj in i for characterization, is to select the path for 1, is not selected then for 0
Select the path；
Path (l) is all paths by link l；
BWi,tTo be intended to the bandwidth demand that i is configured in time period t；
CAPl,tFor bandwidth capacity of the link l in time period t；
Links is all links in current network；
TP is set of all periods.
8. according to claim 1 be intended to the configuration method that time-varying is intended in network, which is characterized in that the step S13
Specifically:
All processed timeslice objects in timeset are traversed, the acedIntentlist in all timeslice objects is taken
Its union forms set C1, and the notacdIntentlist in all timeslice objects takes union to form set C2, takes set C1
Intersection with set C2 is C3, and set C1 subtracts the intention after set C3 and forms the intention set S2 successfully configured.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910390467.4A CN110086677B (en) | 2019-05-10 | 2019-05-10 | Method for configuring time-varying intention in intention network |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910390467.4A CN110086677B (en) | 2019-05-10 | 2019-05-10 | Method for configuring time-varying intention in intention network |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN110086677A true CN110086677A (en) | 2019-08-02 |
| CN110086677B CN110086677B (en) | 2020-07-17 |
Family
ID=67419663
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910390467.4A Active CN110086677B (en) | 2019-05-10 | 2019-05-10 | Method for configuring time-varying intention in intention network |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110086677B (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114095330A (en) * | 2020-07-29 | 2022-02-25 | 华为技术有限公司 | Intention negotiation method and device |
| WO2022042544A1 (en) * | 2020-08-31 | 2022-03-03 | 华为技术有限公司 | Network performance data subscription method and apparatus |
Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20160241436A1 (en) * | 2015-02-17 | 2016-08-18 | Futurewei Technologies, Inc. | Intent Based Network Configuration |
-
2019
- 2019-05-10 CN CN201910390467.4A patent/CN110086677B/en active Active
Patent Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20160241436A1 (en) * | 2015-02-17 | 2016-08-18 | Futurewei Technologies, Inc. | Intent Based Network Configuration |
Non-Patent Citations (1)
| Title |
|---|
| A.ABHASHKUMAR: "Supporting diverse dynamic intent-based policies using Janus", 《PROCEEDING OF THE 13TH INTERNATIONAL CONFERENCE ON EMERGING NETWORKING EXPERIMENTS》 * |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114095330A (en) * | 2020-07-29 | 2022-02-25 | 华为技术有限公司 | Intention negotiation method and device |
| US11909600B2 (en) | 2020-07-29 | 2024-02-20 | Huawei Technologies Co., Ltd. | Intent negotiation method and apparatus |
| WO2022042544A1 (en) * | 2020-08-31 | 2022-03-03 | 华为技术有限公司 | Network performance data subscription method and apparatus |
Also Published As
| Publication number | Publication date |
|---|---|
| CN110086677B (en) | 2020-07-17 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107819840A (en) | Distributed mobile edge calculations discharging method in the super-intensive network architecture | |
| US8724508B2 (en) | Automated policy generation for mobile communication networks | |
| CN103746852B (en) | Service routing configuration method and network management equipment | |
| CN103825838B (en) | A kind of data center removes bandwidth fragmentation stream scheduling method | |
| CN107846714A (en) | The switching method and equipment of a kind of visible light communication and WiFi heterogeneous systems | |
| CN109951869A (en) | A kind of car networking resource allocation methods calculated based on cloud and mist mixing | |
| CN110086677A (en) | The configuration method that time-varying is intended in a kind of intention network | |
| CN107801194B (en) | It is a kind of in heterogeneous network based on the user-association method of return link capacity | |
| CN110708259A (en) | Information-agnostic Coflow scheduling system capable of automatically adjusting queue threshold and scheduling method thereof | |
| CN109254726A (en) | QoS guarantee method, control node and system in distributed memory system | |
| WO2020134133A1 (en) | Resource allocation method, substation, and computer-readable storage medium | |
| CN110191155A (en) | Parallel job scheduling method, system and storage medium for fat tree interconnection network | |
| Libman et al. | The designer's perspective to atomic noncooperative networks | |
| CN105744627B (en) | Wireless dummy net resource allocation methods in OFDMA system | |
| CN108737268A (en) | Software definition industry Internet of Things resource regulating method | |
| CN104935524B (en) | The soft load-balancing method that a kind of multipath network is controlled based on chain-circuit time delay | |
| Wu et al. | Network congestion avoidance through packet-chaining reservation | |
| CN110460546B (en) | Data acquisition method based on network on chip | |
| Suzuki et al. | Multi-agent deep reinforcement learning for cooperative offloading in cloud-edge computing | |
| CN103023819B (en) | A kind of coarseness distribution method of dynamic bandwidth | |
| Sen | Intelligent admission and placement of O-RAN slices using deep reinforcement learning | |
| CN107820295A (en) | A kind of distributed relay resource allocation methods based on user's request | |
| Sun et al. | Hop-by-hop bandwidth allocation and deployment for SFC with end-to-end delay QoS guarantees | |
| CN102612092B (en) | Radio resource management function game-based method for optimizing radio resources | |
| Yang et al. | A game-theoretic approach to stable routing in max-min fair networks |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |