CN105634961A - Intelligent routing implementation method and system - Google Patents
Intelligent routing implementation method and system Download PDFInfo
- Publication number
- CN105634961A CN105634961A CN201511035147.5A CN201511035147A CN105634961A CN 105634961 A CN105634961 A CN 105634961A CN 201511035147 A CN201511035147 A CN 201511035147A CN 105634961 A CN105634961 A CN 105634961A
- Authority
- CN
- China
- Prior art keywords
- routing
- network path
- intelligent routing
- detection
- client application
- 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
- H04L45/00—Routing or path finding of packets in data switching networks
- H04L45/02—Topology update or discovery
Abstract
The invention relates to an intelligent routing implementation method and system. The intelligent routing implementation method comprises the following steps: acquiring configuration information of a client application and an intelligent routing system according to a request of acquiring a configuration information command in a preset intelligent routing protocol; detecting the full time of an appointed network path responding to a request and acquiring an average value of the full time according to a control parameter and a command request for detecting a network speed in a preset intelligent routing protocol; selecting a network path with a minimum average value and establishing a routing table based on the configuration information; when the client application needs to access a dynamic content of a target service, analyzing an URL (Uniform Resource Locator) of the target service based on information of the routing table; and accessing the target service by a network path acquired by the client application based on an analysis result. According to the intelligent routing implementation method and system provided by the invention, the client application, the routing node and the target service directly interact and detect the network speed of each network path, so that the client application can directly obtain a fastest network path which accesses services.
Description
Technical field
The present invention relates to computer network, data mining and field of artificial intelligence, particularly relate to a kind of Intelligent routing and realize method and system.
Background technology
The same with all of Internet service, it is submit request to by the Internet and receive service reaction that user accesses electricity business's service, and mutual speed is to determine the key factor that user uses quality; The network speed of the Internet is the important component part affecting user and service interactive speed, and quickening network speed can improve user and use quality.
Traditional Internet service, for accelerating network access speed, adopts CDN (ContentDeliveryNetwork). CDN builds the distributed content delivery network of one over data networks. The effect of CDN is to adopt streaming media server Clustering, overcomes the shortcoming that one-of-a-kind system output bandwidth and concurrent capability are not enough, can the concurrent flow amount supported of significant increase system, the harmful effect reducing or avoiding single point failure to bring. CDN utilizes GSLB technology to point on the nearest streaming media server working properly of user by accessing of user, by the streaming media server request directly in response to user. The content accessed without user in server, automatically can capture corresponding content from former server according to configuration and be supplied to user.
In realizing process, inventor have found that in conventional art, at least there are the following problems:
The access of the static content of Internet service can be improved network speed by CDN preferably, but the raising accessing network speed of the dynamic content of Internet service is only small, and network access connected ratio is undesirable. Cause reason: the result of the access of dynamic content must be provided by the machine of the terminal serviced, CDN technology rely improve network access speed caching technology can not play a role completely; CDN equally needs to be DNS (DomainNameSystem, domain name system) also like common internet access and resolves, and the unstable properties of its dns resolution is sometimes very slow; And CDN can not intermediate network node use caching technology as a result, it is desirable to be routed to the terminal machine of service, but the route of CDN is not provided that stable performance.
Summary of the invention
Based on this, it is necessary to, poor stability slow for the network access speed of Internet service in conventional art and the not enough problem of service access success rate, it is provided that a kind of Intelligent routing realizes method and system.
To achieve these goals, the embodiment of technical solution of the present invention is:
On the one hand, it is provided that a kind of Intelligent routing realizes method, including step:
According to the acquisition configuration information command request in default Intelligent routing agreement, obtain the configuration information of client application and Intelligent routing system; Configuration information includes the parameter controlling the control parameter of network speed in detection network path, the parameter of each routing node and all territories of client application;
According to the detection network speed command request controlled in parameter and default Intelligent routing agreement, all times that network path response is asked is specified in detection, obtains the meansigma methods of all times; Specify network path to include client application and accessed private line network path and the client application network path by CDN access destination service of destination service by routing node;
Based on configuration information, the network path choosing meansigma methods minimum sets up routing table;
When client application needs the dynamic content accessing destination service, based on the information of routing table, the URL (UniformResoureLocator: uniform resource locator) of destination service is resolved; The network path obtained based on the result resolved by client application accesses destination service.
On the other hand, it is provided that a kind of Intelligent routing system, including Intelligent routing SDK (SDK: SoftwareDevelopmentKit) unit, Intelligent routing SDK unit includes:
Intelligent routing service module, for according to the acquisition configuration information command request in the Intelligent routing agreement preset, obtaining the configuration information of client application and Intelligent routing system; Configuration information includes the parameter controlling the control parameter of network speed in detection network path, the parameter of each routing node and all territories of client application; And according to the detection network speed command request controlled in parameter and default Intelligent routing agreement, all times that network path response is asked is specified in detection, obtains the meansigma methods of all times; Specify network path to include client application and accessed private line network path and the client application network path by CDN access destination service of destination service by routing node;
Client SDK module, for based on configuration information, the network path choosing meansigma methods minimum sets up routing table; And when client application needs the dynamic content accessing destination service, based on the information of routing table, the URL of destination service is resolved; The network path obtained based on the result resolved by client application accesses destination service.
Technique scheme has the advantages that
Intelligent routing of the present invention realizes method and system, because by arranging routing node and setting up the special line of routing node and service endpoint, directly mutual by client application, routing node and destination service and detect the network speed of each network path, so that client application directly obtains the fastest network path of the service of access; And the invention provides a kind of simple Intelligent routing agreement preset, including two protocol commands of network speed obtaining Intelligent routing configuration information and detection network path, what Intelligent Optimal route realizes method, improve the efficiency of detection network speed, decrease network cost and execution time span that default Intelligent routing agreement performs, make the URL of Intelligent routing SDK resolve under network cost little as far as possible, obtain URL parsing effect big as far as possible. And then make user the fastest routing node can be selected to access service by client application, improve network access speed and stability.
Accompanying drawing explanation
Fig. 1 is the schematic flow sheet that Intelligent routing of the present invention realizes embodiment of the method 1;
Fig. 2 is the structural representation of Intelligent routing system embodiment 1 of the present invention;
Fig. 3 is that Intelligent routing of the present invention realizes Intelligent routing principle schematic in method and system one specific embodiment;
Fig. 4 is that Intelligent routing of the present invention realizes the routing strategy schematic diagram of Intelligent routing in method and system one specific embodiment;
Fig. 5 is that Intelligent routing of the present invention realizes Intelligent routing system structure schematic diagram in method and system one specific embodiment;
Fig. 6 is that applications client uses Intelligent routing of the present invention to realize method and system access destination service one specific embodiment schematic diagram;
Fig. 7 is that Intelligent routing of the present invention realizes connecting the obstructed parameter curve schematic diagram occurring to regulate all times with labelling in method and system one specific embodiment.
Detailed description of the invention
For the ease of understanding the present invention, below with reference to relevant drawings, the present invention is described more fully. Accompanying drawing gives the first-selected embodiment of the present invention. But, the present invention can realize in many different forms, however it is not limited to embodiment described herein. On the contrary, the purpose providing these embodiments is to make to the disclosure more thoroughly comprehensively.
Unless otherwise defined, all of technology used herein is identical with the implication that the those skilled in the art belonging to the present invention are generally understood that with scientific terminology. The term used in the description of the invention herein is intended merely to the purpose describing specific embodiment, it is not intended that in the restriction present invention. Term as used herein " and/or " include the arbitrary and all of combination of one or more relevant Listed Items.
Intelligent routing of the present invention realizes embodiment of the method 1:
Network access speed in order to solve Internet service in conventional art is slow, poor stability and the not enough problem of service access success rate, the invention provides a kind of Intelligent routing, to realize embodiment of the method 1, Fig. 1 be the schematic flow sheet that Intelligent routing of the present invention realizes embodiment of the method 1; As shown in Figure 1, it is possible to include step:
Step S110: according to the acquisition configuration information command request in default Intelligent routing agreement, obtain the configuration information of client application and Intelligent routing system; Configuration information includes the parameter controlling the control parameter of network speed in detection network path, the parameter of each routing node and all territories of client application;
Step S120: according to the detection network speed command request controlled in parameter and default Intelligent routing agreement, all times that network path response is asked is specified in detection, obtains the meansigma methods of all times; Specify network path to include client application and accessed private line network path and the client application network path by CDN access destination service of destination service by routing node;
Step S130: based on configuration information, the network path choosing meansigma methods minimum sets up routing table;
Step S140: when client application needs the dynamic content accessing destination service, based on the information of routing table, the URL of destination service is resolved; The network path obtained based on the result resolved by client application accesses destination service.
In a specific embodiment, step S140 specifically may comprise steps of:
Based on the information of routing table, the URL of destination service is resolved to the IP of routing node; The private line network path access destination service at routing node place is passed through by client application;
Or
Based on the information of routing table, resolve the URL returning destination service; Destination service is accessed by CDN by client application.
In a specific embodiment, the present embodiment 1 can also include step:
When meeting the execution trigger condition preset, perform the Intelligent routing agreement preset, obtain configuration information and meansigma methods;
The execution trigger condition preset includes any one in following condition or combination in any: detect when client application starts; Detect that client application enters foreground, and when the time carried on an agreement from last time exceedes Preset Time; Handover network path detected, and when the time carried on an agreement from last time exceedes Preset Time; Detect that all routing nodes of labelling connect the obstructed preset times that exceedes, and when the time carried on an agreement from last time exceedes Preset Time.
In a specific embodiment, step S120 specifically may include that
According to default detection execution sequence, based on controlling parameter, all times of network path is specified to obtain meansigma methods by repeatedly detecting same; Control parameter and include the number of times of detection network speed and the interval of detection; The detection execution sequence preset is preferential detection CDN path, then detects the private line network path of each routing node, wherein, the private line network path of the service endpoint being arrived the multiple territory of client application by same routing node is detected simultaneously.
In a specific embodiment, according to the detection network speed command request controlled in parameter and default Intelligent routing agreement in step S120, detection specifies all times that network path response is asked can obtain all times based on below equation:
RTT=tr-ts+extra_rtt-process_time
Wherein, RTT is all times specifying network path of detection; tsFor sending the clock time of detection network speed command request; trFor receiving the clock time of detection network speed command request response; Extra_rtt is the extra time that routing node arrives the terminal of destination service; Process_time is that routing node determines the process time returning request response.
Intelligent routing system embodiment 1 of the present invention:
Realize the technical thought of embodiment of the method 1 based on above Intelligent routing, present invention also offers the structural representation that a kind of Intelligent routing system embodiment 1, Fig. 2 is Intelligent routing system embodiment 1 of the present invention; As in figure 2 it is shown, Intelligent routing system embodiment 1 can include Intelligent routing SDK unit, and Intelligent routing SDK unit may include that
Intelligent routing service module 200, for according to the acquisition configuration information command request in the Intelligent routing agreement preset, obtaining the configuration information of client application and Intelligent routing system; Configuration information includes the parameter controlling the control parameter of network speed in detection network path, the parameter of each routing node and all territories of client application; And according to the detection network speed command request controlled in parameter and default Intelligent routing agreement, all times that network path response is asked is specified in detection, obtains the meansigma methods of all times; Specify network path to include client application and accessed private line network path and the client application network path by CDN access destination service of destination service by routing node;
Client SDK module 300, for based on configuration information, the network path choosing meansigma methods minimum sets up routing table; And when client application needs the dynamic content accessing destination service, based on the information of routing table, the URL of destination service is resolved; The network path obtained based on the result resolved by client application accesses destination service.
In a specific embodiment, client SDK module 300 includes parsing module 310,
Parsing module 310, for the information based on routing table, resolves to the URL of destination service the IP of routing node or resolves the URL returning destination service; Destination service is accessed by the private line network path at routing node place or CDN accordingly by client application.
In a specific embodiment, intelligent routing service module 200 includes detecting module 210,
Detecting module 210, for according to the detection execution sequence preset, based on controlling parameter, specifies all times of network path to obtain meansigma methods by repeatedly detecting same;
Control parameter and include the number of times of detection network speed and the interval of detection; The detection execution sequence preset is preferential detection CDN path, then detects the private line network path of each routing node, wherein, the private line network path of the service endpoint being arrived the multiple territory of client application by same routing node is detected simultaneously.
In a specific embodiment, detecting module 210 obtains all times based on below equation:
RTT=tr-ts+extra_rtt-process_time
Wherein, RTT is all times specifying network path of detection; tsFor sending the clock time of detection network speed command request; trFor receiving the clock time of detection network speed command request response; Extra_rtt is the extra time that routing node arrives the terminal of destination service; Process_time is that routing node determines the process time returning request response.
Realize the technical scheme of method and system in order to elaborate Intelligent routing of the present invention, special use Intelligent routing system of the present invention to access destination service for client application APP, illustrate that what Intelligent routing of the present invention realized method and system realizes process:
A, Intelligent routing realize principle;
Realize method:
Set up the routing node near user, set up the fast Web access between routing node and service; Specifically, on the one hand, for an application (client application APP), service arrangement is in limited data center, and user distribution is in different geographical position, and the network distance of user and service is relatively far away from. In a specific embodiment, can pass through in network topology structure, dispose a lot of routing node, make the network distance network distance of some user He some routing node being compared to the data center of service little, thus reaching the technique effect near user; Additionally, routing node and traditional the Internet routing node in the embodiment of the present invention are different. Traditional the Internet routing node is independently of concrete application, by the information of switch network architecture, safeguards respective routing table, to through the arbitrary data transmission selecting paths of this routing node. And the routing node of the present invention depends on concrete application, each routing node is attached directly to the service of concrete application by special line, the client of application selects routing node, service request is sent to routing node, and routing node forwards the request to service endpoint either directly through special line, and the reaction that service endpoint is returned is forwarded back to client. Routing node in the embodiment of the present invention is exclusively used in the access of the dynamic content of Internet service.
APP directly accesses routing node IP, and routing node forwards service access and the reaction of user, makes to access the speed of destination service faster than the speed being accessed destination service by CDN networking by routing node networking, it may be assumed that
RTTCON> RTTp_ip+RTTip_s=RTTip,
Wherein, RTTCONIt is all times (RoundTripTime:RTT) being accessed service by CDN, RTTp_ipIt is user's all times of accessing routing node, RTTip_sIt is that routing node accesses all times serviced, RTTipIt it is all times being accessed service by routing node.
The principle that realizes of Intelligent routing is shown in that Fig. 3, Fig. 3 are that Intelligent routing of the present invention realizes Intelligent routing principle schematic in method and system one specific embodiment, and wherein, Fwd represents forwarding; Request/Response: request/reaction; Striped blocks represents the routing node private line network path to App service endpoint (destination service); WLAN represents WirelessLAN (WLAN).
Intelligent routing routing strategy;
For a user, there is multiple routing node can use, and the strategy of APP selection network path is:
If there being n routing node, the condition that network path route is the optimal path of Intelligent routing is
RTTroute=Min (RTTCON, RTTip1..., RTTipn)��
Wherein, RTTCONIt is all times being accessed service by CDN, RTTip1It is all times being accessed service by routing node 1, RTTipnIt it is all times being accessed service by routing node n; The routing strategy figure of Intelligent routing is shown in Fig. 4; Fig. 4 is that Intelligent routing of the present invention realizes the routing strategy schematic diagram of Intelligent routing in method and system one specific embodiment, and as shown in Figure 4, route can be through the network path at a certain routing node place, it is also possible to is pass through CDN.
Intelligent routing Performance Calculation
If there being n routing node, the network speed of use Intelligent routing is:
RTT=RTTCON*pC+RTTip1+p1+��+RTTipn*pn,
And pC+p1+��+pn=1,
Wherein, RTT represents that use Intelligent routing network accesses average all times, RTTCONRepresent and access average all times, RTT by CDNip1Represent by routing node ipiNetwork accesses average all times; pCRepresent the probability accessed by CDN, piRepresent by routing node ipiThe probability that network accesses, wherein i=1,2 ... n.
Usual network stabilization, when Intelligent routing is not ready for routing table, adopts CDN to access, once routing table is set up, Intelligent routing selects the routing node that relative user is the fastest, then the computing formula of RTT can be reduced to:
RTT=RTTCON/pC+RTTipk*(1-pC),
Wherein, ipkIt is the IP of the fastest routing node, pCIt is that Intelligent routing resolves the ratio returning back to URL.
This formula describes, and the interval of the value of RTT is [RTTipk, RTTCON], namely the network speed of Intelligent routing will not be slower than the most slowly by the CDN speed accessed, the fastest network speed not over the fastest routing node relative to user. Wanting the speed ability that Intelligent Optimal route, it is possible to carry out from two aspects, one is the setting optimizing routing node, it is desirable to minimize the RTT of Intelligent routing optimal pathipk, namely for all users, minimum by the meansigma methods of all times of optimal path; Two is to reduce to resolve the ratio returning back to URL, namely to maximize the Intelligent routing actual time played a role.
The performance optimization of Intelligent routing can be weighed by following two formula:
(RTTCON-RTT)/RTTCON(RTTCON-RTT)/PTCON,
Wherein, PTCONIt it is average service time
The former has weighed the ratio that Intelligent routing network speed improves, and the latter has weighed the ratio that service access speed improves.
B. Intelligent routing agreement;
In order to realize the method for definition in above-mentioned Intelligent routing principle, the design packet of Intelligent routing agreement is containing two orders: obtain the network speed in Intelligent routing configuration information and detection network path. The concrete mode of agreement and data form can be different, but substantially, these two command functions are:
1. obtain Intelligent routing configuration information;
This protocol command transmit a request to the configuration service of Intelligent routing, request includes APP essential information (such as APPKey, APP name and version) and multidate information (the OS information such as user of user, machine ID, Network Provider, network type and geographical location information etc.), configuration service returns the configuration information of APP Intelligent routing, detects the parameter of routing node network speed including all of territory of APP, all routing nodes and control.
2. the network speed in detection network path;
This protocol command sends request and specifies network path (URL or routing node IP), the RTT specifying network path is determined in the reaction returned by accepting detection service endpoint (terminal in path, CDN networking is service endpoint, and the terminal of routing node is routing node itself). The computing formula of RTT is as follows:
RTT=tr-ts+extra_rtt-process_time
Wherein, RTT is all times in detected path, networking, tsIt is the clock time sending probe command request, trBeing the clock time receiving probe command reaction, extra_rtt is the detection terminal extra time to service endpoint, and process_time is that detection terminal determines the process time returning reaction. For path, CDN networking, extra_rtt and process_time is 0. For routing node path, routing node can detect the itself RTT to service endpoint in advance as extra_rtt, and returns immediately, and such process_time is 0; Otherwise, it is necessary to real-time detection routing node can detect the itself RTT to service endpoint in advance as extra_rtt.
The performance at networking usually instability, in order to reduce the impact on detection RTT of the unstability of network, agreement allows configuration service to control the number of times of detection network speed and the interval of detection, thus can obtain network speed more accurately by the average RTT repeatedly detected.
Generally, a routing node can connect the service endpoint in multiple territory, even the service endpoint in all territories of an APP. In order to improve the efficiency of detection network speed, probe command allows multiple territories that a routing node is connected to detect simultaneously, and so, the reaction of probe command comprises the sum of the extra_rtt in these territories.
C. the system of Intelligent routing realizes
The realization of Intelligent routing includes intelligent routing service, Intelligent routing client SDK.
Intelligent routing service includes configuration service and detection service, detection service is for the path, networking of CDN and routing node, (service endpoint in each territory of each routing node and APP is all disposed one detection and is serviced to be deployed in the service endpoint in each territory of APP and each routing node respectively. The service that APP accesses divides according to territory, and detection service needs the service endpoint being deployed in all routing nodes and all territories), the network speed order in detection service processing detection network path, including batch detection. Configuration service processes the acquisition Intelligent routing configuration information order of Intelligent routing agreement.
Intelligent routing client SDK performs Intelligent routing agreement, mutual with intelligent routing service, maintaining routing table, when APP needs the dynamic content accessing service, the URL that APP accesses service resolves to the IP (may return original URL) of routing node.
APP resolves the path, networking obtaining the service of access by Intelligent routing client SDK, and uses this networking path access service. If path, networking is by routing node, this routing node, as the relaying of APP and service, forwards the service of APP to ask service endpoint, forwards the reaction that service returns to APP.
Fig. 5 is that Intelligent routing of the present invention realizes Intelligent routing system structure schematic diagram in method and system one specific embodiment, and Fig. 5 describes Intelligent routing phylogenetic relationship, wherein, and CDN: content distributing network; WLAN: WLAN; AppStatusChange: application state changes; NetworkChange: network change; Init: initialize; Refresh: refresh; Resolve: resolve; MarkDown: token network is obstructed; API:ApplicationProgrammingInterface application programming interface; LocalCache: local cache; Node: routing node; RTT:RoundTripTime, round-trip delay; Down: network is obstructed; Score: scoring; Probe: detection.
Fig. 6 is that applications client uses Intelligent routing of the present invention to realize method and system access destination service one specific embodiment schematic diagram; Fig. 6 explains APP and the relation of Intelligent routing client SDK, wherein, and CDN: content distributing network; MobileApp: Mobile solution; AccessAppServices: access application service; ResolveURL: resolve URL; SmartRoutingSDK: Intelligent routing SDK; CallbyIP (fallbacktoURL): call (returning back to URL) by IP; ServiceRoutingNode: routing node; ApplicationService: application service.
D. Intelligent routing SDK
The function of Intelligent routing SDK performs Intelligent routing agreement automatically, safeguards routing table; When APP accesses service, find out best path, networking by routing table, then use the path access service of best networking.
The once execution of Intelligent routing agreement needs repeatedly to perform above-mentioned two protocol command, and the once execution of each protocol command needs to use networking to transmit a request to intelligent routing service and to accept the reaction that intelligent routing service returns. And, once complete agreement performs to require time for span, the execution that routing table is as these orders is perfect gradually, service URL based on routing table resolves the quality stepping up analysis result also with this perfect process, it is to say, the path, networking parsed is progressively close to path, the fastest networking.
So, Intelligent routing SDK to control the opportunity that Intelligent routing agreement performs, frequency and executive mode, pursues and obtains Optimization route table with minimum Executing Cost, and allows Optimization route table play the networking analytical capabilities of maximum duration. In order to reach this target, SDK adopts following method:
Utilize the APP operation characteristic of platform, control to trigger Intelligent routing protocol execution time. Control method (principle of the control of other platforms is also consistent) for mobile platform. At mobile platform, the Intelligent routing agreement implementation can be triggered under the following conditions:
1) APP starts;
2) APP enters foreground, and the time performed from last time exceedes the regular hour;
3) networking switching, the time performed from last time exceedes certain time;
4) all routing nodes are exceeded specific times by markdown (labelling connects obstructed), and exceed certain time from the time performed last time.
The execution sequence of control protocol. In order to allow the selecting paths strategy of Intelligent routing play effect at most in agreement implementation process as early as possible, the best execution sequence of the execution of the probe command of Intelligent routing is first detection path, CDN networking, then detects the fastest routing node path, by that analogy. The repeatedly detection of same paths to be distributed to after all path detection terminate.
Realize the seamless parsing of Intelligent routing. The purpose of the seamless parsing of Intelligent routing is to ensure that Intelligent routing SDK can resolve the URL of the service of access, the path, networking of backward reference service at any time, and is path, networking optimum as far as possible. In order to realize the seamless parsing of Intelligent routing, Intelligent routing SDK can adopt following technology:
1) no matter which kind of state the Intelligent routing agreement implementation is in, and the parsing of URL is that the information according to current routing table selects best path, networking;
2) the once execution of the agreement of Intelligent routing is to perform according to above-mentioned optimization order;
3) agreement of Intelligent routing performs the routing table of foundation, is saved in local cache, in order to when next time, APP was restarted, reads routing table, continues to instruct the parsing of URL;
4) each Intelligent routing agreement implementation, take the progressive mode of the result of each protocol command of progressive updating, guarantee the routing iinformation that Intelligent routing agreement implementation last time is set up, before the agreement execution of a new round does not carry out, continue to instruct the parsing of URL.
QoS (QualityofService, the service quality) mechanism using local markdown processes the unstability at networking. The performance of network is unstable, and a network path was probably optimum network path in the past, currently blocked and obstructed. If the data using agreement to perform are adhered in the parsing of URL always, will return over the soonest but now obstructed network path always, have a strong impact on APP and access service. If APP accesses service time-out and triggers the Intelligent routing agreement implementation, then Intelligent routing agreement is likely to perform frequently, and cost is very high every time; And the Intelligent routing agreement implementation is to have relatively long time span, APP can not wait for an Intelligent routing agreement implementation, then again accesses the access of time-out just now. For the invalid URL brought close to the unstability of the network problem resolved, Intelligent routing SDK can adopt a kind of local QoS mechanism without network cost:
1) APP uses the IP that Intelligent routing resolves to access network but timeout (time-out), markdown this routing node IP every time;
2) each markdown, recalculates the score (scoring) of routing node; The calculating of score can draw based on below equation,
Score=RTT+c0*max+c1*max+��+ck*max
Wherein, max is the once time-out maximum on RTT impact, the c that arrange according to network type0, c1..., and ckIt it is the parameter of time of origin according to relatively current time markdown, concrete parameter setting method in time is shown in that Fig. 7, Fig. 7 are that Intelligent routing of the present invention realizes connecting the obstructed parameter curve schematic diagram occurring to regulate all times with labelling in method and system one specific embodiment.
3) the URL parsing of Intelligent routing is according to the score regulated, rather than the RTT of detection. Substantially, RTT does not have any markdown score regulated.
4) routing node is more than 5 markdown, and this routing node is considered Invalid path, until next time detects successfully.
Intelligent routing SDK combines any of the above method, it is achieved with under the premise of minimum Intelligent routing agreement Executing Cost, the Intelligent routing optimized as far as possible resolves effect.
E. Intelligent routing Intelligentized method;
In order to make Intelligent routing play maximum efficiency, the embodiment of the present invention gathers the data that the agreement of Intelligent routing performs and uses, and excavates these data, and system many aspects are optimized. These data can include the result of the network speed detection that the geographical location information of user, the network information and agreement perform. According to a large amount of above-mentioned user data, it is possible to:
1) setting of network node is analyzed, Reasonable adjustment network node is in the layout in the geographical position of whole users, make the meansigma methods maximization of the network speed that all users are obtained by the fastest respective routing node, namely minimized by the average RTT of the fastest routing node;
2) analyze each geographical position RTT of each network path under heterogeneous networks, at user's acquisition request Intelligent routing configuration information, return the information of each network path, including analyze close to real RTT. Order by single acquisition Intelligent routing configuration information, it is possible to set up close to optimum routing table, more effectively supports that URL resolves earlier;
3) each geographical position RTT of each network path under heterogeneous networks is analyzed, each network path is ranked up, and detect with carrying out networking path velocity with this order, full out path, networking so can be made to effectively enter routing table as early as possible, URL is instructed to resolve, so that APP can use path, the fastest networking to access service as soon as possible.
Intelligent routing of the present invention realizes method and system, is different from conventional art and is based on network self interaction data to determine route, and the present invention is serviced by routing node linking objective by user APP, directly finds out the fastest network path. Namely improved speed and the quality of the network service access of application by structure ADN (ApplicationDeliveryNetwork), the configuration pin of routing node and ADN network structure is to concrete application. Routing node itself is not required to form virtual network, during applications client access network services, finds out the fastest path by the speed detection of application, applies by selecting the direct selected network path of routing node. Principles of the invention is simple, and mathematical model is complete, can optimize direction with instructing by analytical performance, it is easy to realize and application.
Additionally, the present invention is based on Intelligent routing principle, devise simple Intelligent routing agreement, including two protocol commands of network speed obtaining Intelligent routing configuration information and detection network path, and define the formula of computing network speed. Further, the protocol command of the network speed in the detection network path of Intelligent routing agreement runs batch detection mode simultaneously detecting the service endpoint through the multiple territories of same routing node arrival, greatly increase detection network speed ground efficiency, decrease network cost and execution time span that Intelligent routing agreement performs.
URL in order to play the best under the premise at less Intelligent routing agreement Executing Cost resolves effect, thus improving APP to access the overall network efficiency of service, Intelligent routing SDK takes multiple method to be carried out Intelligent Optimal Routing Protocol and performs, the seamless connection technology that the control in triggered time performed including Intelligent routing agreement and Intelligent routing perform. Meanwhile, in order to solve the invalid parsing problem that between twice Intelligent routing agreement execution, the instability of network is brought, the present invention introduces the local QoS mechanism of markdown at Intelligent routing SDK, it is possible to dynamically solve the network blockage problem between twice agreement execution.
The technology of big data analysis is applied to Intelligent routing by the present invention, according to the concrete data that geographical distribution and user's Intelligent routing agreement in each geographical position of user perform, analyzes the Intelligent routing Node distribution of optimum; And, analyze unique user networking speed in each path, networking under a certain geographical position and network, and be sent to user by Intelligent routing configuration, thus, the Intelligent routing SDK of user is but without when carrying out networking detection, it is possible to is effectively taking place URL and resolves; Further, it is possible to analyze each ground, path, networking speed, consumer-oriented Intelligent routing SDK carries out networking detection according to path, networking speed order, uses so that path, networking, fast ground can earlier be resolved by URL, optimizes APP and access networking speed.
Each technical characteristic of embodiment described above can combine arbitrarily, for making description succinct, the all possible combination of each technical characteristic in above-described embodiment is not all described, but, as long as the combination of these technical characteristics is absent from contradiction, all it is considered to be the scope that this specification is recorded.
Embodiment described above only have expressed the several embodiments of the present invention, and it describes comparatively concrete and detailed, but can not therefore be construed as limiting the scope of the patent. It should be pointed out that, for the person of ordinary skill of the art, under the premise conceived without departing from type of the present invention, it is also possible to making some deformation and improvement, these broadly fall into protection scope of the present invention. Therefore, the protection domain of patent of the present invention should be as the criterion with claims.
Claims (10)
1. an Intelligent routing realizes method, it is characterised in that include step:
According to the acquisition configuration information command request in default Intelligent routing agreement, obtain the configuration information of client application and Intelligent routing system; Described configuration information includes the parameter controlling the control parameter of network speed in detection network path, the parameter of each routing node and all territories of described client application;
Asking according to the detection network speed command in described control parameter and described default Intelligent routing agreement, all times of network path response request is specified in detection, obtains the meansigma methods of described all times; Described appointment network path includes described client application and accesses the network path of described destination service by private line network path and the described client application of described routing node access destination service by CDN;
Based on described configuration information, the network path choosing described meansigma methods minimum sets up routing table;
When described client application needs the dynamic content accessing described destination service, based on the information of described routing table, the URL of described destination service is resolved; The network path obtained based on the result of described parsing by described client application accesses described destination service.
2. Intelligent routing according to claim 1 realizes method, it is characterised in that based on the information of described routing table, the URL of described destination service is resolved; The step being accessed described destination service based on the network path that the result of described parsing obtains by described client application includes:
Based on the information of described routing table, the URL of described destination service is resolved to the IP of described routing node; By described client application by destination service described in the private line network path access at described routing node place;
Or
Based on the information of described routing table, resolve the URL returning described destination service; Described destination service is accessed by described CDN by described client application.
3. Intelligent routing according to claim 1 realizes method, it is characterised in that further comprise the steps of:
When meeting the execution trigger condition preset, perform described default Intelligent routing agreement, obtain described configuration information and described meansigma methods;
Described default execution trigger condition includes any one in following condition or combination in any: detect when client application starts; Detect that client application enters foreground, and when the time carried on an agreement from last time exceedes Preset Time; Handover network path detected, and when the time carried on an agreement from last time exceedes Preset Time; Detect that all routing nodes of labelling connect the obstructed preset times that exceedes, and when the time carried on an agreement from last time exceedes Preset Time.
4. Intelligent routing according to claim 1 realizes method, it is characterized in that, ask according to the detection network speed command in described control parameter and described default Intelligent routing agreement, the all times of network path response request is specified in detection, and the step of the meansigma methods obtaining described all times specifically includes:
According to default detection execution sequence, based on described control parameter, all times of network path is specified to obtain described meansigma methods by repeatedly detecting same;
Described control parameter includes the number of times of detection network speed and the interval of detection; Described default detection execution sequence is CDN path described in preferential detection, then the private line network path of each described routing node is detected, wherein, the private line network path of the service endpoint being arrived the multiple territory of described client application by same routing node is detected simultaneously.
5. Intelligent routing according to claim 4 realizes method, it is characterized in that, asking according to the detection network speed command in described control parameter and described default Intelligent routing agreement, detection is specified the step of all times of network path response request to be based on below equation and is obtained described all times:
RTT=tr-ts+extra_rtt-process_time
Wherein, RTT is all times specifying network path of detection; tsFor sending the clock time of described detection network speed command request; trFor receiving the clock time of described detection network speed command request response; Extra_rtt is the described routing node extra time to the terminal of described destination service; Process_time is that described routing node determines the process time returning described request response.
6. the Intelligent routing according to Claims 1-4 any one realizes method, it is characterised in that further comprise the steps of:
The described routing table set up is saved in local cache.
7. an Intelligent routing system, it is characterised in that including Intelligent routing SDK unit, described Intelligent routing SDK unit includes:
Intelligent routing service module, for according to the acquisition configuration information command request in the Intelligent routing agreement preset, obtaining the configuration information of client application and Intelligent routing system; Described configuration information includes the parameter controlling the control parameter of network speed in detection network path, the parameter of each routing node and all territories of described client application; And ask according to the detection network speed command in described control parameter and described default Intelligent routing agreement, all times of network path response request is specified in detection, obtains the meansigma methods of described all times; Described appointment network path includes described client application and accesses the network path of described destination service by private line network path and the described client application of described routing node access destination service by CDN;
Client SDK module, for based on described configuration information, the network path choosing described meansigma methods minimum sets up routing table; And when described client application needs the dynamic content accessing described destination service, based on the information of described routing table, the URL of described destination service is resolved; The network path obtained based on the result of described parsing by described client application accesses described destination service.
8. Intelligent routing system according to claim 7, it is characterised in that described client SDK module includes parsing module,
Described parsing module, for the information based on described routing table, resolves to the URL of described destination service the IP of described routing node or resolves the URL returning described destination service; Described destination service is accessed by the private line network path at described routing node place or described CDN accordingly by described client application.
9. Intelligent routing system according to claim 7, it is characterised in that described intelligent routing service module includes detecting module,
Described detecting module, for according to the detection execution sequence preset, based on described control parameter, specifies all times of network path to obtain described meansigma methods by repeatedly detecting same;
Described control parameter includes the number of times of detection network speed and the interval of detection; Described default detection execution sequence is CDN path described in preferential detection, then the private line network path of each described routing node is detected, wherein, the private line network path of the service endpoint being arrived the multiple territory of described client application by same routing node is detected simultaneously.
10. Intelligent routing system according to claim 9, it is characterised in that described detecting module obtains described all times based on below equation:
RTT=tr-ts+extrt_rtt-process_time
Wherein, RTT is all times specifying network path of detection; tsFor sending the clock time of described detection network speed command request; trFor receiving the clock time of described detection network speed command request response; Extra_rtt is the described routing node extra time to the terminal of described destination service; Process_time is that described routing node determines the process time returning described request response.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201511035147.5A CN105634961B (en) | 2015-12-31 | 2015-12-31 | Intelligent routing realization method and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201511035147.5A CN105634961B (en) | 2015-12-31 | 2015-12-31 | Intelligent routing realization method and system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN105634961A true CN105634961A (en) | 2016-06-01 |
| CN105634961B CN105634961B (en) | 2018-10-16 |
Family
ID=56049479
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201511035147.5A Active CN105634961B (en) | 2015-12-31 | 2015-12-31 | Intelligent routing realization method and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN105634961B (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109327392A (en) * | 2017-07-24 | 2019-02-12 | 网宿科技股份有限公司 | Routing resource and device in a kind of multi-path transmission |
| CN113542135A (en) * | 2021-08-04 | 2021-10-22 | 湖南快乐阳光互动娱乐传媒有限公司 | CDN communication method, system, client and server |
| CN114500367A (en) * | 2021-12-21 | 2022-05-13 | 联想(北京)有限公司 | Path selection method and electronic equipment |
| CN115348195A (en) * | 2022-06-30 | 2022-11-15 | 清华大学 | Method and system for transmitting speed measurement of earth system mode data |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101267449A (en) * | 2008-04-30 | 2008-09-17 | 中山大学 | A tree P2P system resource transmission method based on mobile agent mechanism |
| CN102594679A (en) * | 2012-02-15 | 2012-07-18 | 上海聚力传媒技术有限公司 | Method and equipment for determining operation information of network nodes in video application |
| CN103475654A (en) * | 2013-09-06 | 2013-12-25 | 北京奇虎科技有限公司 | Method for optimizing network paths, equipment and network system |
| CN104320335A (en) * | 2014-11-20 | 2015-01-28 | 北京美琦华悦通讯科技有限公司 | System and method for achieving optimal outlet route selection based on service quality dial testing |
| US20150207708A1 (en) * | 2009-01-28 | 2015-07-23 | Headwater Partners I Llc | Data Usage Monitoring for Media Data Services used by Applications |
-
2015
- 2015-12-31 CN CN201511035147.5A patent/CN105634961B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101267449A (en) * | 2008-04-30 | 2008-09-17 | 中山大学 | A tree P2P system resource transmission method based on mobile agent mechanism |
| US20150207708A1 (en) * | 2009-01-28 | 2015-07-23 | Headwater Partners I Llc | Data Usage Monitoring for Media Data Services used by Applications |
| CN102594679A (en) * | 2012-02-15 | 2012-07-18 | 上海聚力传媒技术有限公司 | Method and equipment for determining operation information of network nodes in video application |
| CN103475654A (en) * | 2013-09-06 | 2013-12-25 | 北京奇虎科技有限公司 | Method for optimizing network paths, equipment and network system |
| CN104320335A (en) * | 2014-11-20 | 2015-01-28 | 北京美琦华悦通讯科技有限公司 | System and method for achieving optimal outlet route selection based on service quality dial testing |
Cited By (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109327392A (en) * | 2017-07-24 | 2019-02-12 | 网宿科技股份有限公司 | Routing resource and device in a kind of multi-path transmission |
| CN109327392B (en) * | 2017-07-24 | 2022-04-22 | 网宿科技股份有限公司 | Path selection method and device in multi-path transmission |
| CN113542135A (en) * | 2021-08-04 | 2021-10-22 | 湖南快乐阳光互动娱乐传媒有限公司 | CDN communication method, system, client and server |
| CN113542135B (en) * | 2021-08-04 | 2023-04-07 | 湖南快乐阳光互动娱乐传媒有限公司 | CDN communication method, system, client and server |
| CN114500367A (en) * | 2021-12-21 | 2022-05-13 | 联想(北京)有限公司 | Path selection method and electronic equipment |
| CN114500367B (en) * | 2021-12-21 | 2023-10-24 | 联想(北京)有限公司 | Path selection method and electronic equipment |
| CN115348195A (en) * | 2022-06-30 | 2022-11-15 | 清华大学 | Method and system for transmitting speed measurement of earth system mode data |
| CN115348195B (en) * | 2022-06-30 | 2024-02-23 | 清华大学 | Method and system for transmitting and measuring speed of earth system mode data |
Also Published As
| Publication number | Publication date |
|---|---|
| CN105634961B (en) | 2018-10-16 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| KR102514250B1 (en) | Method, Apparatus and System for Selecting a Mobile Edge Computing Node | |
| KR101987784B1 (en) | Software-defined network-based method and system for implementing content distribution network | |
| TWI425789B (en) | Multiple node applications cooperatively managing a plurality of packet switched network pathways | |
| US20170180217A1 (en) | Use of virtual endpoints to improve data tranmission rates | |
| US8327008B2 (en) | Methods and apparatus for maintaining network addresses | |
| JP2022548152A (en) | Routing control method, device, program and computer device | |
| CN105634961A (en) | Intelligent routing implementation method and system | |
| CN109218455A (en) | A kind of conversion method and device of application example address | |
| US20130107794A1 (en) | Network communication device and method of selecting active network interface | |
| US10666769B2 (en) | Network system and method for establishing data link by using relay node | |
| CN101834799A (en) | Selection method of server of mobile IPTV (Internet Protocol Television), client, server and system | |
| Al Azad et al. | Cledge: A hybrid cloud-edge computing framework over information centric networking | |
| WO2012065531A1 (en) | Method, device, and system for implementing relay selection | |
| WO2023155617A1 (en) | Live streaming origin-pull method and apparatus | |
| CN100466832C (en) | Group broadcasting member management method for radio local network IP group broadcasting frame transmission | |
| Taleb et al. | Follow-me cloud: An OpenFlow-based implementation | |
| US11212329B2 (en) | Method, apparatus, device and storage medium for pushing video stream | |
| CN110768911B (en) | Method, device, equipment, system and storage medium for efficient flow drainage | |
| Choumas et al. | Testbed innovations for experimenting with wired and wireless software defined networks | |
| KR20110068700A (en) | Multi-mode terminal for managing multi-network interfaces and method for controll therof | |
| Baggan et al. | A comprehensive analysis and experimental evaluation of routing information protocol: An elucidation | |
| CN103188132B (en) | The instant communication method of content-based distributing network and system | |
| CN104660550A (en) | Method for performing session migration among plurality of servers | |
| Wu et al. | Cdn convergence based on multi-access edge computing | |
| CN114143372A (en) | End network control method and system based on converged network splitter |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| TA01 | Transfer of patent application right |
Effective date of registration: 20171215 Address after: 510000 Guangzhou City, Guangzhou, Guangdong, Fangcun Avenue, one of the 314 self compiled Applicant after: Guangzhou Pinwei Software Co., Ltd. Address before: 510370 Guangdong, Guangzhou, Fangcun, No., No. 20 Huahai street, building 1-5 Applicant before: Guangzhou VIPSHOP Information and Technology Co., Ltd. |
|
| TA01 | Transfer of patent application right | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| TR01 | Transfer of patent right |
Effective date of registration: 20220104 Address after: 510000 Room 601, 128 Dingxin Road, Haizhu District, Guangzhou City, Guangdong Province (office only) Patentee after: Vipshop (Guangzhou) Software Co.,Ltd. Address before: 510000 No. 314, Fangcun Avenue, Liwan District, Guangzhou City, Guangdong Province Patentee before: GUANGZHOU PINWEI SOFTWARE Co.,Ltd. |
|
| TR01 | Transfer of patent right |