CN102388585B - Network optimization flow control method, device and system - Google Patents
Network optimization flow control method, device and system Download PDFInfo
- Publication number
- CN102388585B CN102388585B CN201180001972.7A CN201180001972A CN102388585B CN 102388585 B CN102388585 B CN 102388585B CN 201180001972 A CN201180001972 A CN 201180001972A CN 102388585 B CN102388585 B CN 102388585B
- Authority
- CN
- China
- Prior art keywords
- data flow
- link
- transmission path
- price
- node
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Expired - Fee Related
Links
Images
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L45/00—Routing or path finding of packets in data switching networks
- H04L45/12—Shortest path evaluation
Abstract
Embodiments of the invention provide a network optimization flow control method, device and system, wherein the method includes the steps that a data source node obtains a sub-data flow transmission path from the data source node to a sever node, and loads a first data flow amount onto the sub-data flow transmission path; a router node aggregates link prices of a plurality of links on the sub-data flow transmission path to obtain link aggregated price and feeds the link aggregated price back to the data source node; and the data source node obtains a second data flow amount according to the link aggregated price and server weight of the server node and loads the second data flow amount instead of the first data flow amount onto the sub-data flow transmission path. Embodiments of the invention greatly reduce calculation amount, and improves instantaneity of network optimization.
Description
Technical field
The present invention relates to the communication technology, particularly a kind of network optimization flow control method, device and system.
Background technology
(the Internet Service Provider of conventional the Internet service provider, be called for short: ISP) mainly be to provide the Internet and connect, mainly solve traffic engineering (TE) problem on network optimization function, specify optimum path, minimize the Congestion Level SPCC of network for flow, the performance of network of take is optimization aim; And content service provider (Content Provider, be called for short: CP) being mainly the user provides required content, and for the user provides rational resource, settlement server is selected the problem of (SS), is that to take user's experience be optimization aim.If the optimization of ISP and CP is joined together, carry out the combined optimization (JO) of TE and SS, the optimum that can obtain whole network solves, and minimizes the congested simultaneous minimization end-to-end time delay of link reaching.Obtain the balance of network performance global optimum, can significantly promote network performance.
The network optimization flow control method of prior art, one optimization process server normally is set separately, this server need to be collected the required the whole network state information of optimization problem (as link capacity, link flow, user request information, server state information etc.), adopt centralized fashion to obtain the network optimal solution, and optimal solution is applied to network.This centralized fashion refers to, there are a plurality of data source nodes in this network and as a plurality of server nodes of transfer of data destination, also there is accordingly many strips data flow transmission route, the data traffic of all sub data flow transmission paths is all to be obtained by unified processing of this optimization process server, again the optimal data flow obtained is loaded on the sub data flow transmission path in network, the network optimal solution normally in network data source nodes to the data traffic of the sub data flow transmission path between server node.
But, there is following technological deficiency in above-mentioned prior art: because this scheme adopts centralized system, solved, in real network, the network state information that solves these optimal problem needs is more, and especially, when network size enlarges, information requirement is with the number of nodes exponent increase, collection information and spent time of calculating will sharply increase, and amount of calculation is excessive, network node is difficult to real time reaction, brings great difficulty to the enforcement of prioritization scheme.
Summary of the invention
The embodiment of the present invention provides a kind of network optimization flow control method, device and system, and to realize the combined optimization of ISP and CP, and amount of calculation is little, improves speed and the real-time of optimization process.
The embodiment of the present invention provides a kind of network optimization flow control method on the one hand, described network comprises data source nodes, router node and server node, between two adjacent router nodes, be a link, and described two adjacent router nodes form link control module; Described method comprises:
Described data source nodes, according to network topological information, obtains described data source nodes to the sub data flow transmission path between described server node, and loads the first data flow traffic on described sub data flow transmission path; Article one, described sub data flow transmission path comprises some described links;
Described router node aggregates into link flow by the first data flow traffic on described link, and described link control module obtains the link price of described link according to described link flow; Described router node obtains the link aggregation price by the link price polymerization of the some links in described sub data flow transmission path, and described link aggregation price is fed back to described data source nodes;
Described data source nodes is according to described link aggregation price, and the server weight of described server node, obtains the second data flow traffic, and described the second data flow traffic is replaced to described the first data flow traffic is loaded on described sub data flow transmission path.
The embodiment of the present invention provides a kind of network optimization volume control device on the other hand, comprising:
The original upload module, for according to network topological information, obtain data source nodes to the sub data flow transmission path between server node, and load the first data flow traffic on described sub data flow transmission path; Article one, described sub data flow transmission path comprises some described links;
The feedback receiver module, for the link aggregation price of the described sub data flow transmission path that receives described router node feedback, described link aggregation price is obtained the some link price polymerizations in described sub data flow transmission path by described router node;
The Flow-rate adjustment module, be used for according to described link aggregation price, and the server weight of described server node, obtain the second data flow traffic, and described the second data flow traffic is replaced to described the first data flow traffic be loaded on described sub data flow transmission path.
The embodiment of the present invention provides a kind of network optimization flow control system more on the other hand, comprise: data source nodes, server node and routing node, between two adjacent router nodes, be a link, and described two adjacent router nodes form link control module;
Described data source nodes, for according to network topological information, obtain described data source nodes to the sub data flow transmission path between described server node, and load the first data flow traffic on described sub data flow transmission path; Article one, described sub data flow transmission path comprises some described links;
Also for the link aggregation price according to described router node feedback, and the server weight of described server node, obtain the second data flow traffic, and described the second data flow traffic is replaced to described the first data flow traffic be loaded on described sub data flow transmission path;
Described router node, aggregate into link flow for the first data flow traffic by described link, and described link control module obtains the link price of described link according to described link flow; The link price polymerization of the some links in described sub data flow transmission path is obtained to the link aggregation price, and described link aggregation price is fed back to described data source nodes.
The network optimization flow control method of the embodiment of the present invention, device and system, decompose on each node in network and link and realize respectively by the flow rate calculation by the network optimization, solved the large problem of the network optimization flow rate calculation quantity of information requirement, greatly reduce amount of calculation, improved the real-time of the network optimization.
The accompanying drawing explanation
In order to be illustrated more clearly in the embodiment of the present invention or technical scheme of the prior art, in below describing embodiment, the accompanying drawing of required use is briefly described, apparently, accompanying drawing in the following describes is some embodiments of the present invention, for those of ordinary skills, under the prerequisite of not paying creative work, can also obtain according to these accompanying drawings other accompanying drawing.
The schematic flow sheet that Fig. 1 is network optimization flow control method embodiment mono-of the present invention;
The schematic flow sheet that Fig. 2 is network optimization flow control method embodiment bis-of the present invention;
Fig. 3 is that the sub data flow transmission path in network optimization flow control method embodiment bis-of the present invention is divided schematic diagram;
Fig. 4 is the simulation example topological diagram in network optimization flow control method embodiment bis-of the present invention;
Fig. 5 is that the exponential function in network optimization flow control method embodiment bis-of the present invention solves the flow status variation diagram obtained;
Fig. 6 is that the exponential function in network optimization flow control method embodiment bis-of the present invention solves the link utilization state change map obtained;
Fig. 7 is that the power function in network optimization flow control method embodiment bis-of the present invention solves the flow status variation diagram obtained;
Fig. 8 is that the power function in network optimization flow control method embodiment bis-of the present invention solves the link utilization state change map obtained;
The structural representation that Fig. 9 is network optimization volume control device embodiment mono-of the present invention;
The structural representation that Figure 10 is network optimization volume control device embodiment bis-of the present invention;
The structural representation that Figure 11 is network optimization flow control system embodiment of the present invention.
Embodiment
For making the purpose, technical solutions and advantages of the present invention clearer, below in conjunction with the accompanying drawing in the embodiment of the present invention, the technical scheme in the embodiment of the present invention is clearly and completely described, obviously, described embodiment is the present invention's part embodiment, rather than whole embodiment.Embodiment based in the present invention, the every other embodiment that those of ordinary skills obtain under the prerequisite of not making creative work, belong to the scope of protection of the invention.
The main technical schemes of the embodiment of the present invention is, flow rate calculation in the network optimization is decomposed on each node in network and link and realizes respectively, for example, each strip data flow transmission route in network is calculated respectively to its sub data flow flow, and the network node that needed parameter also is distributed in the sub data flow transmission path in flow rate calculation is realized, thereby with respect to optimization process server set Chinese style processing mode of the prior art, greatly reduce amount of calculation, improved the real-time of the network optimization.
Below by drawings and Examples, technical scheme of the present invention is described in further detail.
Embodiment mono-
The schematic flow sheet that Fig. 1 is network optimization flow control method embodiment mono-of the present invention, wherein, can comprise data source nodes, router node and server node in network; And, between two adjacent router nodes, be that a link and these two adjacent router nodes form link control module.As shown in Figure 1, the flow control methods of the present embodiment can comprise the following steps:
Step 101, data source nodes, according to network topological information, obtain described data source nodes to the sub data flow transmission path between described server node, and load the first data flow traffic on the sub data flow transmission path;
For example, compared with prior art, be that the optimization process server arranged separately outside Adoption Network carrys out the computing network flow in prior art, and above-mentioned optimization process server no longer is set in the present embodiment, but directly by the data source nodes in network, carry out flow rate calculation.And network comprises a plurality of data source nodes, the plurality of data source nodes is to calculate respectively flow on its sub data flow transmission path connected.
The source node of data stream transmitting is data source nodes, and the destination node of data stream transmitting is server node, and the path between data source nodes and server node is data flow transmission route.In the present embodiment, each data source nodes can be according to network topological information, and the possible transmission path according to data flow marks off some strip data flow transmission routes of server node.And data flow may be passed through a plurality of router nodes between from the data source nodes to the server node, corresponding, on a strip data flow transmission route, may pass through some links.Wherein, above-mentioned network topological information can be that data source nodes obtains from router node.
Load respectively the first data flow traffic on each sub data flow transmission path that data source nodes obtains in above-mentioned division, as initial flow, this initial flow can be a default initial value, is a less constant.For example, tentation data source node A is to three strip data flow transmission routes are arranged between server node B, what at this three strips data flow transmission route, load so is the first data flow traffic, but the first data flow traffic on each strip data flow transmission route can be identical, also can be not identical.
Step 102, router node obtain the link price of the some links in the sub data flow transmission path, and its polymerization is obtained to the link aggregation price, and described link aggregation price is fed back to described data source nodes;
For example, router node can, according to route matrix, be loaded into the first data flow traffic obtained in step 101 on each link in the sub data flow transmission path.Wherein, likely occur that many strips data flow transmission route, through the same link, just may be loaded with a plurality of data flow on this link accordingly; For example, link between router node A and router node B, simultaneously by the first sub data flow transmission path and the second sub data flow transmission path, just be loaded with so flow a on the first sub data flow transmission path and the flow b on the second sub data flow transmission path on this link.Router node A can aggregate into link flow by flow a and flow b.
Link control module can calculate link price according to link flow obtained above.Wherein, the calculating of link price can have various ways, according to network optimization target, the performance requirement of link is determined; For example, in the network optimization, lay particular emphasis on the utilance of link, wish to make the utilance of link higher, link price just can be set as reflecting the index of link utilization, as " link price=link flow/link heap(ed) capacity ".
May have multilink on one strip data flow transmission route, each link can obtain link price according to aforesaid way.Router node can obtain the link aggregation price by the link price polymerization of the some links in this sub data flow transmission path, and described link aggregation price is fed back to data source nodes.Wherein, the account form of link aggregation price also can have multiple, for example, can select maximum in a plurality of link prices as the link aggregation price, perhaps a plurality of link price additions are obtained to the link aggregation price, or by a plurality of link prices compose respectively power addition obtain link aggregation price etc.
Step 103, data source nodes are according to the link aggregation price, and the server weight of server node, obtain the second data flow traffic, and described the second data flow traffic is replaced to described the first data flow traffic are loaded on described sub data flow transmission path;
For example, the index that the link price of take is the reflection link utilization is example, data source nodes can be according to the initial flow loaded in link price set-up procedure 101, if link price is lower, mean that link utilization is also very low, can increase link flow, data source nodes can continue according to certain growth factor to increase flow on the basis of the initial flow of step 101.
Data source nodes can also be according to the initial flow loaded in the server weight set-up procedure 101 of server node, this server weight can be the disposal ability of reflection server, such as composing with weighted value server with performance parameters such as serving average delay according to the load of server, if the server weight is higher, the disposal ability that shows this server higher (as also less as load, time delay is less etc.), data source nodes can continue according to certain growth factor to increase flow on the basis of initial flow.
Through the above-mentioned adjusting for flow, data source nodes can obtain the second data flow traffic, this second data flow traffic is the flow after adjusting, and can, by the first data flow traffic in this second data flow traffic replacement step 101, be loaded on the sub data flow transmission path.Wherein, considered link performance requirement (link price) and server performance requirement (server weight) in the method for the present embodiment, be the combined optimization of ISP and CP simultaneously, can obtain the balance of network performance global optimum.
Compared with prior art, the network optimization flow control method of the present embodiment, be not only a distributed computing, be about to that flow rate calculation is distributed to each data source nodes and network node is calculated, with respect to the centralized processing mode of prior art, greatly reduced amount of calculation and network signal is mutual; And, the network optimization flow control method of the present embodiment is a process of flow being carried out to dynamic adjustments, can carry out according to network link status and server state etc. the real-time adjusting of flow, with respect to static optimization account form of the prior art, can the dynamically adapting network change, tackle fast the Network Abnormal state, the robustness of system is good.
The network optimization flow control method of the present embodiment, decompose on each node in network and link and realize respectively by the flow rate calculation by the network optimization, solve the large problem of the network optimization flow rate calculation quantity of information requirement, greatly reduced amount of calculation, improved the real-time of the network optimization.
Embodiment bis-
The schematic flow sheet that Fig. 2 is network optimization flow control method embodiment bis-of the present invention, the present embodiment and embodiment mono-are basic identical, just each step in embodiment mono-are described in more details.As shown in Figure 2, the flow control methods of the present embodiment can comprise the following steps:
For example, shown in Figure 3, Fig. 3 is that the sub data flow transmission path in network optimization flow control method embodiment bis-of the present invention is divided schematic diagram.Data source nodes R1 may have many to the data flow transmission route between server node S1, shown in Fig. 3, is two, is respectively x11 and x12.
Wherein, data source nodes is to carry out the division of sub data flow transmission path according to network topological information, and this network topological information can be to be received by router node.Specifically, router node is broadcast to all mid-side nodes by the network topological information of collecting, and data source nodes, can be in local generating network topology table after receiving this information; Data source nodes is again according to the content of network topology table, and according to the data flow to server node, possible transmission path is divided and obtained some independently sub data flow transmission paths.
Data source nodes can generate the first data flow traffic, as initial flow, is loaded on sub data flow transmission path obtained above.Wherein, when the loading of data flow, intermediate router need to be done corresponding forwarding strategy, can comprise based on MPLS and the loading based on two kinds of modes of IP.For example, during based on MPLS, because router node is supported MPLS data pipe function, can directly the first data flow traffic be loaded on the MPLS pipeline, the path while setting up according to pipeline forwards; During based on IP, the mark that adds paths in can the packet corresponding in the first data flow traffic, router node can carry out package forward according to this path tag.
Wherein, router node can, according to route matrix, be loaded into the first data flow traffic obtained in step 201 on each link in the sub data flow transmission path.Wherein, likely occur that many strips data flow transmission route, through the same link, just may be loaded with a plurality of data flow on this link accordingly.
For example, the link between router node A and router node B, simultaneously by x11 and x12 process, just be loaded with flow a on x11 and the flow b on x12 so on this link.Router node A can aggregate into link flow by flow a and flow b.
Wherein, link price is the parameter of reflection link performance, and link control module can directly be set according to the performance requirement of link.For example, in the network optimization, lay particular emphasis on the utilance of link, wish to make the utilance of link higher, link price just can be set as reflecting the index of link utilization, as " link price=link flow/link heap(ed) capacity ".
Link price is according to predetermined utility function f (δ
i) set, with the utility function f (δ of link
i) linear correlation, this utility function is nonlinear function, can be for example dull convex function; Link control module can be according to the utility function f (δ of link
i) the renewal link price, this f (δ
i) just can be used as the price obtained on a certain link.Wherein, utility function f (δ
i) calculated such as the link flow obtained in step 202, chain-circuit time delay, the desired control length of buffer queue of link etc. by link-state information.Wherein, it is nonlinear function that utility function is set, and can, so that the response speed of this optimization method is faster, optimizes flow and approach faster target balance point flow.
In addition, topology table and the link calculation parameter of all right synchronizing network of link control module, with the Distributed Calculation error between synchronization link.
For example, on a strip data flow transmission route, may have multilink, each link can obtain link price according to aforesaid way.Router node can obtain the link aggregation price by the link price polymerization of the some links in this sub data flow transmission path, and described link aggregation price is fed back to data source nodes.
Wherein, router node is according to selected fairness rule, and the link price of each link on the antithetical phrase data flow transmission route carries out polymerization; This fairness rule generally has minimax equity criterion and equitable proportion criterion etc., respectively corresponding different link aggregation methods; For example, if select the minimax equity criterion, can select maximum in a plurality of link prices as the link aggregation price, perhaps selection percentage equity criterion, a plurality of link price additions are obtained to the link aggregation price, perhaps select weighting ratio equity criterion, a plurality of link prices are composed respectively to the power addition and obtained link aggregation price etc.
For example, the server weight can be the index of the disposal ability of reflection server node, wherein, link price represents the optimization requirement of ISP, this server weight represents the optimization requirement of CP, be about to the selection parametrization of server node, according to the performance information of server node to server node dynamic assignment weight.In the present embodiment, following formula (1) the calculation server weight of foundation:
w＝H(y，D)...........................................(1)
Above-mentioned formula (1) can be for example: w=H (y, D)=γ (y/sc
0)+(1-γ) (D/D
0);
Wherein: γ is constant, and 0≤γ≤1;
Sc
0for the treatable data traffic of server in the unit interval;
D
0for the reference delay constant;
In above-mentioned formula (1), the total load that y is server node, the service average delay that D is server node.Server performance information (as load, time delay etc.) can be that server node is sent to data source nodes, then calculates the server weight of this server node according to server performance information by this data source nodes.In the process sent to data source nodes at server node, the router node in network can be forwarded.
Because each sub data flow is independent calculating, when many data flow through a link, these many data flow will be at war with, and the server weight of the destination server node that the ability of competition can be corresponding by each sub data flow is embodied; For example, the server weight is higher, shows that the disposal ability of this server is higher (as also less as load, time delay is less etc.), and the sub data flow that this server node is corresponding is suitable augmented flow just.Regulate by the self adaptation between above-mentioned each sub data flow, can realize the target of TE and SS simultaneously.
The restriction of the not free step of the calculating of above-mentioned server weight, as long as can be before flow rate calculation.After obtaining the server weight, data source nodes can be according to the link aggregation price, and the server weight of server node, calculates the flow of sub data flow transmission path.Concrete, data source nodes can be carried out flow rate calculation according to following formula (2):
Wherein,
K is constant, and 0<k;
The weight that w is connected server;
The result that above-mentioned formula (2) obtains is the derivative of flow, according to derivative value, can calculate flow.The mode of this formula (2) calculated flow rate is a kind of exploration account form of greediness, and formula comprises two parts, i.e. rate increase part and rate limit part.Wherein, rate increase is partly a kind of expansion pattern of greediness, and spreading rate is relevant to price and server weight; Time per unit increases kwf (δ
i); The meaning of rate limit part is the greedy expansion of restriction, and the data flow that speed is higher is expanded limited highlyer, and rate-constrained is in-kx
i.
Wherein, there is the exploration factor delta in the utility function in formula (2)
i, this δ
ivalue according to following formula (3):
Wherein,
A is normal number;
Q is the polymerization price that every sub data flow obtains;
By above-mentioned formula, δ
ican realize oscillation on small scale, and then can realize the dynamic approximation of flow so that the flow calculated is finally done oscillation on small scale near balance point, thereby need the less network information can realize higher control precision, improve the degree of approaching optimum point.Wherein, balance point is a kind of dbjective state of network global optimization, is equivalent to solve in the static optimization mode the desirable optimal solution obtained in prior art.In the present embodiment, balance point corresponding to
the balance point solved with NBS (receive assorted agreed-upon price solution) static state is consistent, adopts the flow rate calculation mode of the present embodiment can the default optimum point of auto-feeding, realizes the optimization of whole network in dynamic competition.
The flow calculated according to above-mentioned formula in this step 205, for the flow according to after network state information and server state information adjustment, can be called the second data flow traffic.Data source nodes can be by the first data flow traffic in this second data flow traffic replacement step 201, and is loaded on the sub data flow transmission path.Then, after loading the second data flow traffic, router node can carry out the flow polymerization again as step 202, and follow-up step; Whole network can automatically obtain the flow control result of the network optimization in the dynamic feedback adjustment shown in Fig. 2.
Below the flow control result that method finally reached with a simulation example explanation the present embodiment is consistent with static optimal solution, and the method for the present embodiment is effective:
Referring to Fig. 4, Fig. 4 is the simulation example topological diagram in network optimization flow control method embodiment bis-of the present invention.This network comprises four data source nodes, is respectively Source1, Source2, Source3 and Source4, and each data source nodes connects a strip data flow transmission route; Wherein, comprise three sections links on the sub data flow transmission path that Source1 connects, these three sections links are Link1, Link2 and Link3.The capacity of supposing above-mentioned three sections links is c
0=10000; This capacity is the peak transfer rate between router node, is normalized capacity, and unit can be for example kb.
If solve according to static mode, the static optimal solution of this network is as follows:
Wherein, x
*the assignment of traffic situation that means four strip data flow transmission routes in network, for example, mean flow on the four strip data flow transmission routes link capacity C that should respectively do for oneself in the present embodiment
01/3, y
*mean the link utilization situation of three sections links, for example, the utilance of the Link1 in the present embodiment, Link2 and Link3 is respectively 2/3*C
0, C
0and C
0.
Adopt the dynamic solution mode calculation optimization flow in the present embodiment, for example, suppose the utility function f (δ of link
i) be exponential function, corresponding formula (2) is
sound out factor delta
ivalue with reference to formula (3), the link price formula is
the link aggregation price that data source nodes obtains is
the result of calculation obtained can be referring to Fig. 5 and Fig. 6.Wherein, in the present embodiment, for convenience of explanation, simplify the design of utility function, adopted utility function
the direct computing formula as price;
being a kind of comparatively simple way of realization of utility function, can also be other the functional forms such as power function.
Fig. 5 is that the exponential function in network optimization flow control method embodiment bis-of the present invention solves the flow status variation diagram obtained, and Fig. 6 is that the exponential function in network optimization flow control method embodiment bis-of the present invention solves the link utilization state change map obtained.By Fig. 5, can be obtained, the Packet Generation rate of four data source nodes in the network shown in final Fig. 4 is all on average 1/3*C
0, the flow that is equivalent to the sub data flow transmission path is 1/3*C
0, with the x in static optimal solution
*consistent; By Fig. 6, can be obtained, two of the utilances of three sections links in the network shown in final Fig. 4 are C
0, another is 2/3*C
0, with the y in static optimal solution
*unanimously.And, by Fig. 5 and Fig. 6, also can be found out, flow and utilance are being a process that dynamic self-adapting approaches by initial value to the final optimization pass value, and do oscillation on small scale until reach poised state near balance point.
Again for example, suppose the utility function f (δ of link
i) be power function, corresponding formula (2) is
sound out factor delta
ivalue with reference to formula (3), the link price formula is
the link aggregation price that data source nodes obtains is
the result of calculation obtained can be referring to Fig. 7 and Fig. 8.Fig. 7 is that the power function in network optimization flow control method embodiment bis-of the present invention solves the flow status variation diagram obtained, and Fig. 8 is that the power function in network optimization flow control method embodiment bis-of the present invention solves the link utilization state change map obtained.The result obtained is consistent with Fig. 5 and Fig. 6, also consistent with static optimal solution, just the process difference of dynamic approximation.
The network optimization flow control method of the present embodiment, decompose on each node in network and link and realize respectively by the flow rate calculation by the network optimization, solve the large problem of the network optimization flow rate calculation quantity of information requirement, greatly reduced amount of calculation, improved the real-time of the network optimization.
Embodiment tri-
The structural representation that Fig. 9 is network optimization volume control device embodiment mono-of the present invention, the network optimization volume control device of the present embodiment can, for the data source nodes described in any embodiment of the present invention, can be carried out the network optimization flow control method described in any embodiment of the present invention.
The present embodiment is simply introduced the structure of this device, and each concrete functions of modules can be described referring to embodiment of the method with the execution principle.As shown in Figure 9, this device can comprise original upload module 91, feedback receiver module 92 and Flow-rate adjustment module 93.
Wherein, original upload module 91 can obtain data source nodes to the sub data flow transmission path between server node according to network topological information, and loads the first data flow traffic on described sub data flow transmission path; Article one, described sub data flow transmission path comprises some described links;
Flow-rate adjustment module 93 can be according to described link aggregation price, and the server weight of described server node, obtain the second data flow traffic, and described the second data flow traffic is replaced to described the first data flow traffic be loaded on described sub data flow transmission path.
The network optimization volume control device of the present embodiment, by original upload module, feedback receiver module and Flow-rate adjustment module are set, flow rate calculation in the network optimization is decomposed on each node in network and link and realizes respectively, solved the large problem of the network optimization flow rate calculation quantity of information requirement, greatly reduce amount of calculation, improved the real-time of the network optimization.
Embodiment tetra-
The structural representation that Figure 10 is network optimization volume control device embodiment bis-of the present invention, the present embodiment has carried out refinement to the structure in device embodiment mono-, and each concrete module and the function of unit can be described referring to embodiment of the method with the execution principle.As shown in figure 10, in this device,
Original upload module 91 can comprise the first loading unit 911 and/or the second loading unit 912.Wherein, the first loading unit 911 can be loaded on described MPLS pipeline by described the first data flow traffic, and described sub data flow transmission path is the MPLS pipeline; The second loading unit 912 mark that adds paths in can the packet corresponding in described the first data flow traffic, so that described router node, according to described path tag, is loaded on described the first data flow traffic on described sub data flow transmission path.
Wherein, in above-mentioned " the first loading unit 911 and/or the second loading unit 912 " " and/or " refer to, this network optimization volume control device can have the first loading unit 911 and the second loading unit 912 simultaneously, has two kinds of corresponding functions simultaneously; Perhaps, also can only there is one of them in the first loading unit 911 and the second loading unit 912, only there is wherein a kind of function.
Further, the device of the present embodiment can also comprise weight computation module 94, this weight computation module 94 can receive the server performance information that described server node sends, and obtains the server weight of described server node according to described server performance information.
The network optimization volume control device of the present embodiment, by original upload module, feedback receiver module and Flow-rate adjustment module are set, flow rate calculation in the network optimization is decomposed on each node in network and link and realizes respectively, solved the large problem of the network optimization flow rate calculation quantity of information requirement, greatly reduce amount of calculation, improved the real-time of the network optimization.
Embodiment five
The structural representation that Figure 11 is network optimization flow control system embodiment of the present invention, the system of the present embodiment can be carried out the network optimization flow control method of any embodiment of the present invention, and concrete wherein each module and the function of unit can be described referring to embodiment of the method with the execution principle.
As shown in figure 11, the system of the present embodiment can comprise that between data source nodes 1101, server node 1102 and 1103, two adjacent router nodes of routing node be a link, and described two adjacent router nodes form link control module; Wherein, data source nodes is equivalent to the network optimization volume control device described in the embodiment of the present invention.
Wherein, described data source nodes 1101, for according to network topological information, obtain described data source nodes to the sub data flow transmission path between described server node, and load the first data flow traffic on described sub data flow transmission path; Article one, described sub data flow transmission path comprises some described links;
Described router node 1103, aggregate into link flow for the first data flow traffic by described link, and described link control module obtains the link price of described link according to described link flow; The link price polymerization of the some links in described sub data flow transmission path is obtained to the link aggregation price, and described link aggregation price is fed back to described data source nodes;
Described data source nodes 1101, also for the link aggregation price according to described router node feedback, and the server weight of described server node, obtain the second data flow traffic, and described the second data flow traffic is replaced to described the first data flow traffic be loaded on described sub data flow transmission path.
The network optimization flow control system of the present embodiment, decompose on each node in network and link and realize respectively by the flow rate calculation by the network optimization, solve the large problem of the network optimization flow rate calculation quantity of information requirement, greatly reduced amount of calculation, improved the real-time of the network optimization.
One of ordinary skill in the art will appreciate that: realize that the hardware that all or part of step of said method embodiment can be relevant by program command completes, aforesaid program can be stored in a computer read/write memory medium, this program, when carrying out, is carried out the step that comprises said method embodiment; And aforementioned storage medium comprises: various media that can be program code stored such as ROM, RAM, magnetic disc or CDs.
Finally it should be noted that: above embodiment only, in order to technical scheme of the present invention to be described, is not intended to limit; Although with reference to previous embodiment, the present invention is had been described in detail, those of ordinary skill in the art is to be understood that: its technical scheme that still can put down in writing aforementioned each embodiment is modified, or part technical characterictic wherein is equal to replacement; And these modifications or replacement do not make the essence of appropriate technical solution break away from the spirit and scope of various embodiments of the present invention technical scheme.
Claims (9)
1. a network optimization flow control method, it is characterized in that, described network comprises data source nodes, router node and server node, between two adjacent router nodes, is a link, and described two adjacent router nodes form link control module; Described method comprises:
Described data source nodes, according to network topological information, obtains described data source nodes to the sub data flow transmission path between described server node, and loads the first data flow traffic on described sub data flow transmission path; Article one, described sub data flow transmission path comprises some described links;
Described router node aggregates into link flow by the first data flow traffic on described link, and described link control module obtains the link price of described link according to described link flow; Described router node obtains the link aggregation price by the link price polymerization of the some links in described sub data flow transmission path, and described link aggregation price is fed back to described data source nodes;
Described data source nodes is according to described link aggregation price, and the server weight of described server node, obtains the second data flow traffic, and described the second data flow traffic is replaced to described the first data flow traffic is loaded on described sub data flow transmission path.
2. network optimization flow control method according to claim 1, is characterized in that, loads the first data flow traffic on described sub data flow transmission path, comprising:
Described sub data flow transmission path is the MPLS pipeline, and described the first data flow traffic is loaded on described MPLS pipeline; Perhaps,
The mark that adds paths in corresponding packet in described the first data flow traffic, so that described router node, according to described path tag, is loaded on described the first data flow traffic on described sub data flow transmission path.
3. network optimization flow control method according to claim 1, is characterized in that, also comprises:
Described data source nodes receives the server performance information that described server node sends, and obtains the server weight of described server node according to described server performance information.
4. network optimization flow control method according to claim 1, is characterized in that, described link control module obtains the link price of described link according to described link flow, comprising:
Described link control module obtains the utility function of described link according to described link flow, and obtains the link price of described link according to described utility function;
Described utility function is nonlinear function.
5. network optimization flow control method according to claim 4, is characterized in that,
Described utility function comprises the exploration factor, and it is normal number that the described exploration factor is less than or equal at 1 o'clock in described link aggregation price, and being greater than at 1 o'clock in described link aggregation price is negative constant.
6. a network optimization volume control device, is characterized in that, comprising:
The original upload module, for according to network topological information, obtain data source nodes to the sub data flow transmission path between server node, and load the first data flow traffic on described sub data flow transmission path; Article one, described sub data flow transmission path comprises some described links;
The feedback receiver module, for the link aggregation price of the described sub data flow transmission path that receives described router node feedback, described link aggregation price is obtained the some link price polymerizations in described sub data flow transmission path by described router node;
The Flow-rate adjustment module, be used for according to described link aggregation price, and the server weight of described server node, obtain the second data flow traffic, and described the second data flow traffic is replaced to described the first data flow traffic be loaded on described sub data flow transmission path.
7. network optimization volume control device according to claim 6, is characterized in that, described original upload module comprises:
The first loading unit, for described the first data flow traffic is loaded on to described MPLS pipeline, described sub data flow transmission path is the MPLS pipeline; Perhaps,
The second loading unit, for the mark that adds paths at packet corresponding to described the first data flow traffic, so that described router node, according to described path tag, is loaded on described the first data flow traffic on described sub data flow transmission path.
8. network optimization volume control device according to claim 6, is characterized in that, also comprises:
Weight computation module, for receiving the server performance information of described server node transmission, and obtain the server weight of described server node according to described server performance information.
9. a network optimization flow control system, is characterized in that, comprising: data source nodes, server node and routing node are a link between two adjacent router nodes, and described two adjacent router nodes form link control module;
Described data source nodes, for according to network topological information, obtain described data source nodes to the sub data flow transmission path between described server node, and load the first data flow traffic on described sub data flow transmission path; Article one, described sub data flow transmission path comprises some described links;
Also for the link aggregation price according to described router node feedback, and the server weight of described server node, obtain the second data flow traffic, and described the second data flow traffic is replaced to described the first data flow traffic be loaded on described sub data flow transmission path;
Described router node, aggregate into link flow for the first data flow traffic by described link, and described link control module obtains the link price of described link according to described link flow; The link price polymerization of the some links in described sub data flow transmission path is obtained to the link aggregation price, and described link aggregation price is fed back to described data source nodes.
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| PCT/CN2011/079827 WO2012149748A1 (en) | 2011-09-19 | 2011-09-19 | Network optimization traffic control method, device and system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN102388585A CN102388585A (en) | 2012-03-21 |
| CN102388585B true CN102388585B (en) | 2014-01-08 |
Family
ID=45826513
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201180001972.7A Expired - Fee Related CN102388585B (en) | 2011-09-19 | 2011-09-19 | Network optimization flow control method, device and system |
Country Status (2)
| Country | Link |
|---|---|
| CN (1) | CN102388585B (en) |
| WO (1) | WO2012149748A1 (en) |
Families Citing this family (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US9998531B2 (en) | 2013-09-18 | 2018-06-12 | International Business Machines Corporation | Computer-based, balanced provisioning and optimization of data transfer resources for products and services |
| CN105471997B (en) * | 2015-12-04 | 2019-02-22 | 燕山大学 | Method for controlling flow in P2P file sharing network based on price mechanism |
| CN105721573B (en) * | 2016-02-04 | 2019-06-25 | 燕山大学 | P2P file sharing network fair bandwidth sharing method based on optimization utility |
| CN107070803A (en) * | 2017-01-18 | 2017-08-18 | 中国人民解放军信息工程大学 | A kind of jamming control method communicated suitable for multi-path network |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101924680A (en) * | 2009-06-10 | 2010-12-22 | 谢海永 | Distributed network flow combined optimization system and method based on feedback |
| 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 |
Family Cites Families (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP3923863B2 (en) * | 2002-07-09 | 2007-06-06 | 株式会社日立製作所 | Request router device |
| CN102025628B (en) * | 2010-12-07 | 2012-11-14 | 北京邮电大学 | Distribution method of network flow |
-
2011
- 2011-09-19 CN CN201180001972.7A patent/CN102388585B/en not_active Expired - Fee Related
- 2011-09-19 WO PCT/CN2011/079827 patent/WO2012149748A1/en active Application Filing
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101924680A (en) * | 2009-06-10 | 2010-12-22 | 谢海永 | Distributed network flow combined optimization system and method based on feedback |
| 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 |
Non-Patent Citations (2)
| Title |
|---|
| 《DISTRIBUTED JOINT OPTIMIZATION OF TRAFFIC ENGINEERING AND SERVER SELECTION》;Pengye Xia等;《Proceedings of 2010 IEEE 18th International Packet Video Workshop》;20101214;全文 * |
| Pengye Xia等.《DISTRIBUTED JOINT OPTIMIZATION OF TRAFFIC ENGINEERING AND SERVER SELECTION》.《Proceedings of 2010 IEEE 18th International Packet Video Workshop》.2010,全文. |
Also Published As
| Publication number | Publication date |
|---|---|
| WO2012149748A1 (en) | 2012-11-08 |
| CN102388585A (en) | 2012-03-21 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109257287B (en) | Shortest path determining method and controller | |
| CN102710489B (en) | Dynamic shunt dispatching patcher and method | |
| CN102355670B (en) | Multichannel wireless mesh network channel distribution method | |
| CN112346854B (en) | In-network resource scheduling method and system for hierarchical collaborative decision and storage medium | |
| CN102158938B (en) | Power-adjustable zonal sensor network topology control method | |
| CN105282038B (en) | For the distributed group of stars group optimization method based on stability analysis in mobile satellite network | |
| CN102625404B (en) | Distributed routing protocol method for three-dimensional underwater acoustic sensor networks | |
| CN106209669A (en) | Towards SDN data center network maximum of probability path stream scheduling method and device | |
| CN112350949B (en) | Rerouting congestion control method and system based on flow scheduling in software defined network | |
| CN102388585B (en) | Network optimization flow control method, device and system | |
| CN103036787A (en) | Network route convergence processing method and network route convergence processing device | |
| CN105553845A (en) | Software defined network route selection method | |
| CN105978805A (en) | Anycast attribute service oriented energy-saving routing method | |
| CN103036792A (en) | Transmitting and scheduling method for maximizing minimal equity multiple data streams | |
| CN103124420A (en) | Wireless on-chip network structuring method | |
| CN101729365A (en) | Route establishing method, carrying network router and network system | |
| CN103117957A (en) | Cache management method based on numbers of message replications and comprehensive effectiveness in opportunistic network | |
| EP3002915B1 (en) | Routing in dependence of the application | |
| CN107204933A (en) | Distributed stream amount adjustment method and device | |
| CN105337899A (en) | Optical network energy-saving routing method capable of flexibly adjusting reserved-type service transmission bandwidth | |
| CN103929782A (en) | Resource leveling multi-path routing method applicable to industrial wireless sensor network | |
| CN102025628A (en) | Distribution method of network flow | |
| CN103001892A (en) | Network resource distribution method and system based on cloud computing | |
| CN102427596B (en) | Routing method and scheduling method of node mobile network assisted by positioning information | |
| CN105050149A (en) | Routing method for overcoming congestion of tree type Zigbee network |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20140108 Termination date: 20190919 |
|
| CF01 | Termination of patent right due to non-payment of annual fee |