CN108075909A - A kind of method for predicting and device - Google Patents

A kind of method for predicting and device Download PDF

Info

Publication number
CN108075909A
CN108075909A CN201610997850.2A CN201610997850A CN108075909A CN 108075909 A CN108075909 A CN 108075909A CN 201610997850 A CN201610997850 A CN 201610997850A CN 108075909 A CN108075909 A CN 108075909A
Authority
CN
China
Prior art keywords
flow
region
user
local dns
output port
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
Application number
CN201610997850.2A
Other languages
Chinese (zh)
Other versions
CN108075909B (en
Inventor
元艳素
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Alibaba Group Holding Ltd
Original Assignee
Alibaba Group Holding Ltd
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Alibaba Group Holding Ltd filed Critical Alibaba Group Holding Ltd
Priority to CN201610997850.2A priority Critical patent/CN108075909B/en
Publication of CN108075909A publication Critical patent/CN108075909A/en
Application granted granted Critical
Publication of CN108075909B publication Critical patent/CN108075909B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Classifications

    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L41/00—Arrangements for maintenance, administration or management of data switching networks, e.g. of packet switching networks
    • H04L41/14—Network analysis or design
    • H04L41/147—Network analysis or design for predicting network behaviour
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L43/00—Arrangements for monitoring or testing data switching networks
    • H04L43/08—Monitoring or testing based on specific metrics, e.g. QoS, energy consumption or environmental parameters
    • H04L43/0876—Network utilisation, e.g. volume of load or congestion level
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L61/00—Network arrangements, protocols or services for addressing or naming
    • H04L61/45—Network directories; Name-to-address mapping
    • H04L61/4505—Network directories; Name-to-address mapping using standardised directories; using standardised directory access protocols
    • H04L61/4511—Network directories; Name-to-address mapping using standardised directories; using standardised directory access protocols using domain name system [DNS]

Landscapes

  • Engineering & Computer Science (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • Signal Processing (AREA)
  • Environmental & Geological Engineering (AREA)
  • Data Exchanges In Wide-Area Networks (AREA)

Abstract

This programme embodiment provides a kind of method for predicting and device, the described method includes：Gather the flow from each user area；In the flow from each user area gathered, the accounting in the output port region of each local dns of flux and flow direction in the user region is determined；According to the accounting in the output port region of each local domain name system DNS of the flux and flow direction of the flow from each user area and each user area, the total flow in the output port region of each local dns is determined.Due to determining the total flow in the output port region of each local dns according to the accounting in the output port region of each local domain name system DNS of the flux and flow direction of each user area in this programme embodiment, the accuracy of the flow predicted is improved come predicted flow rate according to the total flow in the output port region of each local dns.

Description

A kind of method for predicting and device
【Technical field】
The present invention relates to field of communication technology more particularly to a kind of method for predicting and device.
【Background technology】
In CDN (Content Delivery Network, content distributing network) network, in order to avoid some CDN is saved The problem of network congestion caused by point server is risen suddenly and sharply due to flow and service quality decline is, it is necessary to CDN node server Flowing of access is predicted.
In the prior art, it is according to User IP (Internet when predicting the flowing of access of CDN node server Protocol, Internet protocol) flow that generates of attributed region predicts the access stream of the CDN node server of corresponding region Amount.But since user can be to LDNS (local Domain Name System, the home domain name system of user equipment ownership System) (hereinafter referred to as：Local dns) server is modified, such as：Beijing user is manually set local dns server Input port IP be arranged to Shanghai local dns server input port IP, if using the prior art to CDN Shanghai save Point server carries out volume forecasting, can be by the flowing of access of CDN Shanghai node server since the user IP belongs to Beijing Area It is predicted as the flowing of access of CDN Beijing node server.Therefore, the stream predicted in the prior art according to User IP attributed region The accuracy of amount is relatively low.
【The content of the invention】
In view of this, this programme embodiment provides a kind of method for predicting and device, to solve in the prior art The problem of accuracy of the flow predicted according to User IP attributed region is relatively low.
In a first aspect, this programme embodiment provides a kind of method for predicting, including：
Gather the flow from each user area；
In the flow from each user area gathered, each local dns of flux and flow direction in the user region are determined Output port region accounting；
According to the output terminal mouth region of each local dns of the flux and flow direction of the flow from each user area and each user area The accounting in domain determines the total flow in the output port region of each local dns.
Aspect as described above and any possible realization method, it is further provided a kind of realization method, acquisition is from each The flow of user area, including：
According to the corresponding flow of User IP, IP and the mapping relations in region, it is useful that acquisition belongs to institute in same user area The total flow of family IP.
Aspect as described above and any possible realization method, it is further provided a kind of realization method, what is gathered In flow from each user area, determine that the output port region of each local dns of flux and flow direction in the user region accounts for Than, including：
Obtain first flow ratio；The first flow ratio represents input of the single user region to single local dns The access times of port areas account for ratio of the user region to the access times in the input port region of all local dns；
According to the mapping relations and the first flow ratio in input port region and output port region, second is obtained Flow proportional, using the accounting in the output port region of each local dns of flux and flow direction as the user region；The second flow Ratio represents that single user region accounts for the user region to all to the access times in the output port region of single local dns The ratio of the access times in the output port region of ground DNS.
Aspect as described above and any possible realization method, it is further provided a kind of realization method obtains first-class Amount ratio, including：
Access times of each User IP to the input port IP of each local dns are counted, same user area will be belonged to User IP is clustered, and obtains access times of each user area for the input port IP of each local dns, to be visited as first Ask number；
Based on first access times, the input port in the input port region of same local dns will be belonged to IP is clustered, and obtains access times of each user area to the input port region of each local dns, secondary to be accessed as second Number；
According to the first access times and the second access times, first flow ratio is obtained.
Aspect as described above and any possible realization method, it is further provided a kind of realization method, the method is also Including：
According to the total flow in the output port region of each local dns and region and the mapping relations of scheduling domain name, obtain Dispatch the flow of domain name.
Aspect as described above and any possible realization method, it is further provided a kind of realization method, the method is also Including：
And, relevant with dispatching zone name historical traffic record obtains and institute according to the flow of the scheduling domain name State the relevant predicted flow rate of dispatching zone name.
A technical solution in above-mentioned technical proposal has the advantages that：In this programme embodiment, gathering After flow from each user area, the total flow for belonging to all users in the same area generation, Ran Hou are just obtained In the flow from each user area gathered, the output port of each local dns of flux and flow direction in the user region is determined The accounting in region is assured that out the flow export ratio of flow that user in each region generates in each local dns at this time, It is accounted for further according to the output port region of each local dns of flux and flow direction of the flow from each user area and each user area Than determining the total flow in the output port region of each local dns, that is, defining the traffic flow that the user in all areas generates To the specific flow in each local dns output port region, further, according to the output terminal mouth region for each local dns determined The total flow in domain understands that the output port region of each local dns is the flow that corresponding CDN node server introduces, due to CDN The flowing of access of node server is introduced according to the output port region of local dns, therefore according to the output of each local dns The total flow of port areas carrys out the accuracy that predicted flow rate improves the flow predicted.
Second aspect, this programme embodiment provide a kind of volume forecasting device, including：
Flow collection unit, for gathering the flow from each user area；
Flow accounting computing unit, in the flow from each user area gathered, determining the user area The accounting in the output port region of each local dns of flux and flow direction in domain；
Traffic statistics unit, for according to flow and each of the flux and flow direction of each user area from each user area The accounting in the output port region of ground DNS determines the total flow in the output port region of each local dns.
Aspect as described above and any possible realization method, it is further provided a kind of realization method, the flow are adopted Collect unit, be specifically used for：
According to the corresponding flow of User IP, IP and the mapping relations in region, it is useful that acquisition belongs to institute in same user area The total flow of family IP.
Aspect as described above and any possible realization method, it is further provided a kind of realization method, the flow account for Than computing unit, it is specifically used for：
Obtain first flow ratio；The first flow ratio represents input of the single user region to single local dns The access times of port areas account for ratio of the user region to the access times in the input port region of all local dns；
According to the mapping relations and the first flow ratio in input port region and output port region, second is obtained Flow proportional, using the accounting in the output port region of each local dns of flux and flow direction as the user region；The second flow Ratio represents that single user region accounts for the user region to all to the access times in the output port region of single local dns The ratio of the access times in the output port region of ground DNS.
Aspect as described above and any possible realization method, it is further provided a kind of realization method, when the flow When accounting computing unit is used to obtain first flow ratio, including：
Access times of each User IP to the input port IP of each local dns are counted, same user area will be belonged to User IP is clustered, and obtains access times of each user area for the input port IP of each local dns, to be visited as first Ask number；
Based on first access times, the input port in the input port region of same local dns will be belonged to IP is clustered, and obtains access times of each user area to the input port region of each local dns, secondary to be accessed as second Number；
According to the first access times and the second access times, first flow ratio is obtained.
Aspect as described above and any possible realization method, it is further provided a kind of realization method, described device is also Including：
Dispatching zone name flow rate calculation unit, total flow and region for the output port region according to each local dns Mapping relations with dispatching domain name obtain the flow of scheduling domain name.
Aspect as described above and any possible realization method, it is further provided a kind of realization method, described device is also Including：
Dispatching zone name flow predicting unit, for according to it is described scheduling domain name flow and, with the dispatching zone famous prime minister The historical traffic record of pass, obtains and the relevant predicted flow rate of dispatching zone name.
A technical solution in above-mentioned technical proposal has the advantages that：In this programme embodiment, in flow After collecting unit acquires the flow from each user area, just obtain and belong to the total of the generation of all users in the same area Flow, then flow accounting computing unit determine the user region in the flow from each user area gathered The accounting in the output port region of each local dns of flux and flow direction, the stream that the user being assured that out at this time in each region generates The flow export ratio in each local dns is measured, traffic statistics unit is further according to the flow from each user area and each user The accounting in the output port region of each local dns of flux and flow direction in region determines total stream in the output port region of each local dns Amount defines the specific flow in each local dns output port region of flux and flow direction that the user in all areas generates, into One step, the output port region of each local dns according to the total flow in the output port region for each local dns determined For corresponding CDN node server introduce flow, since the flowing of access of CDN node server is according to the defeated of local dns What exit port region introduced, therefore prediction is improved come predicted flow rate according to the total flow in the output port region of each local dns The accuracy of the flow arrived.
【Description of the drawings】
It, below will be to needed in the embodiment attached in order to illustrate more clearly of the technical solution of this programme embodiment Figure is briefly described, it should be apparent that, the accompanying drawings in the following description is only some embodiments of the present invention, for this field For those of ordinary skill, without having to pay creative labor, it can also be obtained according to these attached drawings other attached Figure.
Fig. 1 is a kind of flow example figure of the domain name in this programme embodiment in CND network accelerations；
Fig. 2 is a kind of method for predicting flow example figure in this programme embodiment one；
Fig. 3 is a kind of flow example figure of implementation method for step 102 in this programme embodiment three；
Fig. 4 is a kind of flow example figure of implementation method for step 201 in this programme example IV；
Fig. 5 is a kind of signal for clustering the User IP for belonging to same user area in this programme example IV Figure；
Fig. 6 a are a kind of input terminals in input port region that will belong to same local dns in this programme example IV The schematic diagram that mouth IP is clustered；
Fig. 6 b are the inputs in another input port region that will belong to same local dns in this programme example IV The schematic diagram that Port IP is clustered；
Fig. 7 is a kind of flow example figure of method for predicting in this programme embodiment five；
Fig. 8 is the flow example figure of another method for predicting in this programme embodiment five；
Fig. 9 is a kind of topology example figure of volume forecasting device in this programme embodiment six；
Figure 10 is a kind of simplified block diagram of CDN dispatch servers 100 in this programme embodiment seven.
【Specific embodiment】
For a better understanding of the technical solution of the present invention, this programme embodiment is retouched in detail below in conjunction with the accompanying drawings It states.
It will be appreciated that described embodiment is only part of the embodiment of the present invention, instead of all the embodiments.Base Embodiment in the present invention, those of ordinary skill in the art obtained without creative efforts it is all its Its embodiment, belongs to the scope of protection of the invention.
The term used in this programme embodiment is only merely for the purpose of description specific embodiment, and is not intended to be limiting The present invention.In this programme embodiment and " one kind " of singulative used in the attached claims, " described " and "the" It is also intended to including most forms, unless context clearly shows that other meanings.
It should be appreciated that term "and/or" used herein is only a kind of incidence relation for describing affiliated partner, represent There may be three kinds of relations, for example, A and/or B, can represent：Individualism A, exists simultaneously A and B, individualism B these three Situation.In addition, character "/" herein, it is a kind of relation of "or" to typically represent forward-backward correlation object.
Depending on linguistic context, word as used in this " if " can be construed to " ... when " or " when ... When " or " in response to determining " or " in response to detection ".Similarly, depending on linguistic context, phrase " if it is determined that " or " if detection (condition or event of statement) " can be construed to " when definite " or " in response to determining " or " when the detection (condition of statement Or event) when " or " in response to detecting (condition or event of statement) ".
As shown in Figure 1, CDN network includes user equipment, local dns server, Authoritative DNS server, DNS scheduling clothes Business device and CDN node server when needing to carry out CDN acceleration to domain name, comprise the following steps：
1st, domain name is sent to local dns by the input port IP of the local dns server set and taken by user equipment Business device.
2nd, the domain name is transmitted to Authoritative DNS server by the local dns server, and generates user access information, should Include at least the correspondence of User IP and local dns input port IP in access information, and the access information can be by DNS dispatch servers obtain.
3rd, Authoritative DNS server can be recorded according to the domain name lookup to the corresponding CNAME of the domain name (specification name), and will CNAME records return to the local dns server.
4th, the local dns server can carry out recursive query after CNAME records are received, when carrying out recursive query, The message for carrying the domain name and output port IP is sent to by DNS dispatch servers by output port, the wherein output Port is output port corresponding with receiving the input port of the domain name.
5th, then the DNS dispatch servers are saved according to the available CDN node server of the domain name lookup at available CDN Inquiry belongs to the CDN node server of the same area with output port IP in point server, and generates Query Information, then will The IP of the CDN node server is sent to the local dns server, and domain name, local dns service are included at least in the Query Information The correspondence of device output port IP and the CDN node server inquired.
6th, the IP of the CDN node server is sent to the user's equipment by the local dns server.
If the 7, the IP of the CDN node server is the IP of the CDN node server of a-quadrant, user equipment is according to A areas The domain name is sent to the CDN node server of a-quadrant by the IP of the CDN node server in domain.
8th, the data of acquisition are returned to the user's equipment by the CDN node server of the a-quadrant according to the domain name, and are generated User's flowing of access information includes at least User IP, domain name, CDN node server ip, flowing of access in the flowing of access information With the correspondence of timestamp, the user's flow information can also be obtained by DNS dispatch servers.
From above-mentioned flow, the flowing of access of CDN node server is introduced by local dns server, therefore, It, can be according to the region of local dns server ownership, for corresponding area when predicting the flowing of access of CDN node server The flow that the CDN node server in domain introduces is predicted, further, since CDN scheduling systems are exported according to local dns Corresponding CDN node server is distributed in the region of port ownership, and the region of local dns ownership and local dns output terminal Mouth region domain is identical, therefore the total flow that can be flowed out according to the output port region of local dns, carries out corresponding CDN node clothes The volume forecasting of business device.
The method predicted below the flowing of access of CDN node server mentioned above is specifically described.
Embodiment one
This programme embodiment provides a kind of method for predicting, specifically as shown in Fig. 2, this method comprises the following steps：
101st, the flow from each user area is gathered.
Specifically, user area is the User IP set for belonging to the same area, further, since IP and region are that have Correspondence, therefore the region of the user IP ownership can be determined according to User IP, determining that each User IP returns Behind the region of category, it is possible to determine that those User IPs belong to same user area, and can determine that the user IP specifically belongs to In which user area, such as：User IP 1 belongs to Beijing Area, and User IP 2 belongs to Beijing Area, the upper sea area of the ownership of User IP 3 Domain, User IP 4 belong to upper sea region, then User IP 1 and User IP 2 belong to same user area, that is, belong to Beijing user area Domain, User IP 3 and User IP 4 belong to same user area, that is, belong to Shanghai user area.
It is assured that out after the flow from each user area is acquired and belongs to the generation of the user in the same area Flow.
102nd, in the flow from each user area gathered, each local of flux and flow direction in the user region is determined The accounting in the output port region of DNS.
Specifically, the output port region of different local dns can be passed through in the flow generated due to the user region It is introduced into different CDN node servers, it is thus necessary to determine that each local dns of the flux and flow direction for going out the user region is defeated The accounting in exit port region further, stream is accessed since user equipment accesses to generate in corresponding CDN node server It is that the IP of the CDN node server obtained according to the user's equipment accesses generation during amount, and the CDN node server IP be to be determined according to the output port region of DNS, i.e., the user's equipment accesses generation in the CDN node server Flowing of access is to be introduced by the output port region of the DNS for the CDN node server, further, the user's equipment The flowing of access of generation is accessed in the CDN node server and the user's equipment flows to the stream in the DNS output ports region Amount is identical, also, since the User IP of each user equipment belongs to different user areas, works as and determine user area After the accounting in the output port region of the flux and flow direction in the domain local dns, it is possible to determine the user region in the local dns The corresponding CDN node server in output port region on the accounting of flowing of access that generates, therefore, determining the user area After the accounting in the output port region of each local dns of flux and flow direction in domain, it is possible to know the output port region of each local dns The flow in the user region being introduced into accounts for the ratio of the total flow that user generates in the user region, wherein, the user region The sum of accounting in output port region of each local dns of flux and flow direction is 1.
For example, the flow of Beijing user area has only flowed to the local dns for belonging to Beijing Area and has belonged to sea area A flow part for the local dns in domain, i.e. Beijing user area has flowed to the local dns for belonging to Beijing Area, and remainder The flow divided then can all flow to the local dns for belonging to sea region, also, since the flow for flowing to local dns can all lead to The output port region outflow of the local dns is crossed, therefore flow direction belongs to the output port region of the local dns of Beijing Area Accounting and flow direction belong to the sum of accounting in output port region of local dns of sea region as 1.
103rd, according to the output terminal of each local dns of flux and flow direction of flow and each user area from each user area The accounting in mouth region domain determines the total flow in the output port region of each local dns.
Specifically, when determine each user area each local dns of flux and flow direction output port region accounting and After flow from each user area, it may be determined that go out the output port region of each local dns of flux and flow direction in the user region Flow, then in the sum of flow in output port region of each local dns of flux and flow direction according to each user area, determine The total flow in the output port region of each local dns.
Such as：User area includes Beijing user area and Shanghai user area, and local dns include Pekinese's local dns The sheet in Pekinese's local dns and Shanghai is flowed to respectively with the flow of the local dns in Shanghai, also, the generation of Beijing user area Ground DNS, the flow that Shanghai user area generates have flowed to Pekinese's local dns and the local dns in Shanghai respectively, wherein, Beijing The flow that user area flows to Pekinese's local dns is 5, and the flow that Beijing user area flows to the local dns in Shanghai is 10, on The flow that extra large user area flows to Pekinese's local dns is 5, and the flow that Shanghai user area flows to the local dns in Shanghai is 10, So it is 10 to flow to the sum of flow in output port region of Pekinese's local dns, the output port of the local dns in Shanghai is flowed to The sum of the flow in region is 20, i.e., the total flow in the output port region of Pekinese's local dns is 10, the local dns in Shanghai The total flow in output port region is 20.
Also, since local dns are after a flow is received, which can pass through the local dns by the flow The flow is sent to DNS dispatch servers by output port, which can be according to the output port of the local dns Corresponding CDN node server is determined in region, and then user equipment is made to be generated on the CDN node server and once accesses stream Amount, therefore local dns get how many flow, then it will be in CDN node corresponding with the output port region of the local dns Generate how many flowing of access in server, therefore can be according to the total flow flowed out from the output port region of the local dns, really The fixed local dns are the flow that CDN node server introduces, further, according to the total of the output port region of each local dns Flow to corresponding CDN node server carry out volume forecasting, it is predicted that flow it is more accurate.
After the total flow in output port region of each local dns is determined, this method further includes：
According to the total flow in the output port region of each local dns and region and the mapping relations of scheduling domain name, obtain Dispatch the flow of domain name.
Specifically, the entitled domain name group of dispatching zone, that is, belong to the set of the domain name of the same area, due in CDN network The domain name of middle service is more, is managed for convenience, it is necessary to which the domain name of service is restrained, the domain name that will belong to the same area is received It holds back into a scheduling domain name, and establishes the scheduling domain name and the mapping relations in region.When the output port for knowing each local dns After the total flow in region, the total flow in the output port region of each local dns can be converged to the total flow of each scheduling domain name, Belong to all domain names in the same area and carry out the flow generated during CDN acceleration.
After the flow of scheduling domain name is obtained, this programme embodiment further includes：
And, relevant with dispatching zone name historical traffic record obtains and institute according to the flow of the scheduling domain name State the relevant predicted flow rate of dispatching zone name.
It interior data on flows ought be determined for the previous period specifically, the flow of the scheduling domain name obtained is basis, In order to improve the accuracy of predicted flow rate, it is also necessary to obtain dispatching zone name relevant historical traffic record, come further determine that with The relevant predicted flow rate of dispatching zone name.Such as：For dispatch domain name flow and with the dispatching zone name relevant one go through History discharge record sets coefficient respectively, the coefficient include but it is unlimited be expressed as a percentage, for representing the flow of the scheduling domain name Or the ratio of weight shared by this historical traffic, wherein, the ratio of weight shared by the flow of the scheduling domain name or this historical traffic The sum of example is 1, wherein, historical traffic record relevant with the dispatching zone name is more long, and the coefficient of setting is smaller, i.e. shared power Weight ratio is smaller, after coefficient is set, obtains the flow of scheduling domain name and the product of corresponding coefficient and acquisition and institute The product of dispatching zone name relevant historical traffic record and corresponding coefficient is stated, it is long-pending and determine according to two obtained above Historical traffic record relevant with the dispatching zone name.
In this programme embodiment, since the flowing of access of CDN node server is introduced by local dns, further , since local dns receive how many flow how many flow will be flowed out from the output port of the local dns, and from local Output port region outflow how many total flows of DNS will generate how many flowing of access in CDN node server, therefore After obtaining the total flow in output port region of local dns, it is possible to determine to be introduced for CDN node server by local dns Flow number, i.e. CDN node server can generate how many flowing of access, therefore according to the output port of each local dns The total flow in region carrys out the accuracy that predicted flow rate improves the flow predicted.
Embodiment two
In order to realize flow of the acquisition from each user area, this programme embodiment provides a kind of realization for gathering flow Mode, specific method are as follows：
According to the corresponding flow of User IP, IP and the mapping relations in region, it is useful that acquisition belongs to institute in same user area The total flow of family IP.
Specifically, after user equipment receives CDN node server ip, domain name can be sent to the CDN node service Device carries out CDN acceleration, the flow that the user IP is generated can be known after CDN acceleration is carried out, due to pre-setting IP and region Correspondence, according to IP and the correspondence, it may be determined that go out the region belonging to the IP, therefore according to the corresponding stream of User IP Amount and the correspondence, can know the total flow for belonging to all User IPs in the same area.
Embodiment three
In order to determine the accounting in the output port region of each local dns of the flux and flow direction in the user region, this programme is implemented Example provides a kind of specific implementation for step 102, as shown in figure 3, specifically including following steps：
201st, first flow ratio is obtained.
Wherein, the first flow ratio represents visit of the single user region to the input port region of single local dns Ask that number accounts for ratio of the user region to the access times in the input port region of all local dns.
Specifically, user is when carrying out CDN acceleration, it is necessary first to local dns are accessed by the input port of local dns, And different local dns can be accessed in a User IP, are accessed and recorded according to the user of local dns, it may be determined that go out use The access times and the user IP for each local dns that family IP is accessed are to total access times of all local dns, Ran Hougen According to default IP and the correspondence in region, it may be determined that go out the input port region and user area of local dns, and then can be with Determine single user region to the access times in the input port region of single local dns and the user region to all Total access times in the input port region of local dns, according to single user region to the input port region of single local dns Access times and the user region the ratio between total access times in input port region of all local dns can be determined with the One flow proportional.
202nd, according to the mapping relations and the first flow ratio in input port region and output port region, obtain Second flow ratio, using the accounting in the output port region of each local dns of flux and flow direction as the user region.
Wherein, the second flow ratio represents visit of the single user region to the output port region of single local dns Ask that number accounts for ratio of the user region to the access times in the output port region of all local dns.
Specifically, due to needing the flow generated according to the output port region of local dns when predicting flow It is predicted, therefore, it is necessary to the input port region of local dns is received the output port that flow is converted to the local dns The flow of region output, also, can pass through the DNS input ports after receiving data by input port due to a local dns Data are all exported, so after first flow ratio is known, according to pre-set input port region and output port The mapping relations in region, it may be determined that go out to represent visit of the single user region to the output port region of single local dns Ask that number accounts for ratio of the user region to the access times in the output port region of all local dns, i.e. second flow ratio.
Further, a local dns include multiple input port and multiple output ports, and an input port receives Data can be exported by a corresponding output port, but since all input ports of a local dns belong to same A region (all input port IP correspond to same region), all output ports of a local dns belong to same region (all output port IP correspond to same region), and the region of input port ownership is identical with the region that output port belongs to, All it is the region of local dns ownership, therefore, the first flow ratio and second of all input ports of local dns Amount ratio can represent the total flow that single user region is generated in the flow that the DNS attributed regions generate with the user region The ratio between, further, belong to the first flow ratio of same local dns and second flow ratio be it is identical, further , the mapping relations in input port region and output port region be belong to the same area input port and input port have Correspondence.
Example IV
This programme embodiment provides a kind of specific implementation for step 201, specifically as shown in figure 4, the realization Mode comprises the following steps：
301st, each User IP is counted to the access times of the input port IP of each local dns, will belong to same user area The User IP in domain is clustered, and obtains access times of each user area for the input port IP of each local dns, using as One access times.
Specifically, when domain name is sent to local dns by user, can be generated in the local dns<User IP, local dns Input port IP>It is right, according to what is generated in the local dns<User IP, local dns input port IP>To number can obtain Each User IP is obtained to the access times of the input port IP of the local dns, according to IP and the correspondence in region, can will be belonged to together User IP in same user area is clustered, i.e.,：It will be all<User IP, local dns input port IP>To being converted to< User area, local dns input port IP>It is right, according to<User area, local dns input port IP>To each use can be obtained Family region further, can know certain for the access times of the input port IP of each local dns according to the first access times Total access times of one user area.
It is if specific as shown in figure 5, being generated in local dns<User IP, local dns input port IP>To including：<With Family IP1, the input port IPA of local dns>、<User IP 2, the input port IPB of local dns>、<User IP 3, local dns Input port IPC>、<User IP 4, the input port IPD of local dns>、<User IP 5, the input port IPE of local dns>, such as Fruit IP and the correspondence in region are：<IP1, region A>、<IP2, region A>、<IP3, region A>、<IP4, region A>、<IP5, Region A>, 3 can be obtained after the User IP for belonging to same user area is clustered<User area A, local dns Input port IP>It is right, i.e.,<User area A, local dns input port IPA>、<User area A, local dns input port IPB> With<User area A, local dns input port IPC>And 2<User area B, local dns input port IP>It is right, i.e.,<With Family region B, local dns input port IPD>With<User area B, local dns input port IPE>, and then it is known that user Region A accesses the number of all local dns as 3 times, and the number that user area B accesses all local dns is 2 times.
302nd, based on first access times, the input in the input port region of same local dns will be belonged to Port IP is clustered, and obtains access times of each user area to the input port region of each local dns, to be visited as second Ask number.
Specifically, after total access times in a certain user area are known, it, will be same according to IP and the correspondence in region The input port IP for belonging to the input port region of same local dns is clustered, and can obtain each user area to each local The access times in the input port region of DNS, i.e., a certain user area access the number of the local dns of ownership different zones.
Specifically as shown in Figure 6 a, if the User IP for belonging to same user area clustered<User area Domain, local dns input port IP>To for：<User area A, local dns input port IPA>、<User area A, local dns are defeated Inbound port IPB>With<User area A, local dns input port IPC>If IP and the correspondence in region are：<IPA, region A>、<IPB, region A>With<IPC, region B>, the input port IP progress in the input port region of same local dns will be belonged to It is available after cluster<User area, local dns input port region>To for：<User area A, local dns input port Region A>、<User area A, local dns input port region A>With<User area A, local dns input port region B>, root According to above-mentioned<User area, local dns input port region>To understanding：User area A accesses local dns input port region A Number for 2 times, the number that user area A accesses local dns input port region B is 1 time.
Specifically as shown in Figure 6 b, if the User IP for belonging to same user area clustered<User area Domain, local dns input port IP>To for：<User area B, local dns input port IPD>With<User area B, local dns Input port IPE>If IP and the correspondence in region are：<IPD, region A>With<IPE, region B>, same will be belonged to The input port IP in the input port region of ground DNS is available after being clustered<User area, local dns input port Region>To for：<User area B, local dns input port region A>With<User area B, local dns input port region B>, According to above-mentioned<User area, local dns input port region>To understanding：User area B accesses local dns input port region The number of A is 1 time, and the number that user area B accesses local dns input port region B is 1 time.
303rd, according to the first access times and the second access times, first flow ratio is obtained.
Specifically, user area A accesses the total degree of the input port IP of all local dns as 3 times according to Fig. 6 a, And also know that user area A accesses the number of the input port region A of local dns as 2 times according to Fig. 6 a, user area A is visited The number of the input port region B of local dns is asked as 1 time, so as to according to Fig. 6 a：Inputs of the user area A to local dns The access times of port areas A account for user area A 67%, user area A account for user area B to the defeated of all local dns to the access times of the input port region A of local dns The ratio of the access times in inbound port region is about 33%.
Meanwhile user area B accesses the total degree of the input port IP of all local dns as 2 times according to Fig. 6 b, and User area B accesses the number of the input port region A of local dns as 1 time according to Fig. 6 b, and user area B accesses local The number of the input port region B of DNS is 1 time, so as to according to Fig. 6 b：User area B is to the input terminal mouth region of local dns It is 50% to the ratio of the access times in the input port region of all local dns that the access times of domain A, which account for user area B, is used Family region B accounts for input terminal mouth regions of the user area B to all local dns to the access times of the input port region B of local dns The ratio of the access times in domain is 50%.
Further, 4 first flow ratios can be obtained according to Fig. 6 a and Fig. 6 b, and according to input port region with it is defeated The mapping relations in exit port region can obtain corresponding 4 second flow ratios, wherein, 4 second flow ratio difference It represents：User area A accounts for user area A to the defeated of all local dns to the access times of the output port region A of local dns The ratio of the access times in exit port region is about 67%, access time of the user area A to the output port region B of local dns It is about 33% to the ratio of the access times in the output port region of all local dns that number, which accounts for user area A, B pairs of user area The access times of the output port region A of local dns account for access of the user area B to the output port region of all local dns The ratio of number is 50%, and user area B accounts for the access times of the output port region B of local dns user area B to all The ratio of the access times in the output port region of local dns is 50%.
Embodiment five
It is real to this programme in conjunction with specific application scenarios in order to which the technological thought of this programme embodiment is expanded on further The technical solution for applying example illustrates, and the program is performed in CDN dispatch servers, has known certain in the CDN dispatch servers A domain name carries out the correspondence of the related data of domain name acceleration by CDN network, such as：First correspondence<IP, region>, the Two correspondences<User IP, the input port IP of local dns>, the 3rd correspondence<Domain name, the output port IP of local dns, Node server>, the 4th correspondence<User IP, domain name, CDN node server, flowing of access, timestamp>And the 5th Correspondence<Domain name dispatches domain name>, wherein, the second correspondence and the 3rd correspondence are remembered by the daily record of local dns What record obtained, the 4th correspondence is obtained according to the log recording of CDN node server, wherein, it often carries out a CDN and adds Speed can all generate second correspondence, the 3rd correspondence and the 4th correspondence, know preset time After above-mentioned correspondence in section, when carrying out volume forecasting, specifically as shown in fig. 7, method is as follows：
1st, according to the first correspondence, the second correspondence is polymerize, obtains the first sub- correspondence：<User IP Affiliated area, local dns input port IP affiliated areas>.
Specifically, the input port IP of User IP and local dns is converted to by the first correspondence belonging to User IP First sub- correspondence of the input port IP affiliated areas of region and local dns, can obtain according to the first sub- correspondence Know each user area for the access times in the input port region of each local dns and each user area to all local dns Total access times in input port region, specific implementation is referring to Fig. 5, Fig. 6 a and the corresponding example IVs of Fig. 6 b, herein not It repeats one by one again.
2nd, according to the first correspondence, the 3rd correspondence is polymerize, obtains the second sub- correspondence<Local dns Output port IP affiliated areas>.
Specifically, the local dns output port IP in the 3rd correspondence is converted into local by the first correspondence Region belonging to DNS output ports IP.
3rd, according to local dns output port IP affiliated areas and the correspondence of local dns input port IP affiliated areas, First sub- correspondence and the second sub- correspondence are restrained, obtain the 3rd sub- correspondence<User IP affiliated area, this Ground DNS output port IP affiliated areas, flow proportional>.
It specifically, can be by User IP affiliated area to belonging to local dns input port IP according to the second sub- correspondence The access times in region are converted to access times of the User IP affiliated area to local dns output port IP affiliated areas, by this User IP affiliated area is converted to the affiliated areas of the user IP to total access times of all local dns input port IP affiliated areas It domain, and can be according to the access times and total access to total access times of all local dns output port IP affiliated areas Number can obtain the 3rd flow proportional from correspondence, which is used to represent single user region to single The access times in the output port region of ground DNS account for access time of the user region to the output port region of all local dns Several ratios.
It can certainly be obtained according to the first sub- correspondence and represent input terminal of the single user region to single local dns The access times in mouth region domain account for ratio of the user region to the access times in the input port region of all local dns, then The flow proportional is obtained according to the mapping relations in input port region and output port region, specific implementation is referring to implementation Example three, this is no longer going to repeat them.
4th, according to the first correspondence, the 4th correspondence is polymerize, obtains the 4th sub- correspondence<User IP institute Belong to region, domain name, flowing of access>.
Specifically, the User IP in the 4th correspondence is converted into the area belonging to User IP by the first correspondence Domain can know the total flow of User IP affiliated area outflow according to the 4th sub- correspondence, further, according to the described 4th Sub- correspondence can obtain the total flow of each area outflow.
5th, the 3rd sub- correspondence and the 4th sub- correspondence are polymerize, obtains the 5th sub- correspondence<Local dns Output port IP affiliated areas, domain name, flowing of access>.
Specifically, it can be known from the total of local dns output port IP affiliated areas outflow according to the 5th sub- correspondence Flow, it is the flow that CDN node server introduces just to have known a local dns output port IP affiliated areas at this time.
6th, according to the 5th correspondence and the 5th sub- correspondence, the 6th correspondence is obtained<Local dns output port IP Affiliated area dispatches domain name, flowing of access>.
Specifically, the scheduling domain name of the local dns output port IP affiliated areas can be known according to the 6th correspondence Flow.
7th, it is relevant according to the flow of the scheduling domain name of local dns output port IP affiliated areas and with the dispatching zone name The coefficient of historical traffic record setting, is weighted, obtains and the relevant predicted flow rate of dispatching zone name.
Concrete mode is referring to embodiment one, and this is no longer going to repeat them.
In a specific application scenarios, user area includes Beijing user area and Shanghai user area, local dns Local dns including Pekinese's local dns and Shanghai, CDN node server include Pekinese's CDN node server and Shanghai CDN node server, and the access domain name of Beijing user area and Shanghai user area is all a.test.com, wherein, Record has the log recording of the correspondence of User IP and local dns input port IP in local dns, in CDN node server The log recording of the correspondence of User IP, domain name, CDN node server, flowing of access and timestamp is recorded, it is specific at one Realization method in, specifically as shown in figure 8, comprising the following steps：
1st, the log recording of the log recording to the local dns in Shanghai and Pekinese's local dns is respectively according to IP and region Correspondence carry out IP- regions conversion process, obtain region described in User IP affiliated area and local dns input port IP Correspondence.
2nd, correspondence obtained above is analyzed and counted, according to IP affiliated areas and local dns input port IP The number of the correspondence in the region obtains Beijing user area and accesses in the access times and access of Pekinese's local dns The access times and visit Shanghai of the access times of the local dns in sea and Shanghai user area access Pekinese's local dns The access times of local dns.
3rd, determine that Beijing user uses Beijing in the access times of Beijing local dns according to access times obtained above Family region generate total access times the first ratio, Beijing user Shanghai local dns access times to Beijing user area Second ratio of total access times that domain generates, Shanghai user produce Shanghai user area in the access times of Beijing local dns 3rd ratio of raw total access times, Shanghai user generate Shanghai user area in the access times of Shanghai local dns 4th ratio of total access times.
Such as：It is 0.9 to obtain the first ratio, and the second ratio is 0.1, and the 3rd ratio is 0, and the 4th ratio is 1, and generates use Family represents the correspondence of above-mentioned data：<Beijing user area, Pekinese's local dns, 0.9>,<Beijing user area, Shanghai Local dns, 0.1>,<Shanghai user area, Pekinese's local dns, 0>With<Shanghai user area, the local dns in Shanghai, 1 >。
4th, according to IP and the correspondence in region, the region belonging to local dns output port IP is determined, wherein, Beijing Local dns output port IP belonging to region for Beijing, the region belonging to the local dns output port IP in Shanghai is Shanghai.
5th, by the first ratio obtained above, the second ratio, the 3rd ratio and the 4th ratio, the affiliated area of User IP is converted to The correspondence in domain, local dns output port IP affiliated areas and flow proportional, i.e.,<Beijing user area, Pekinese The output port of DNS, 0.9>,<Beijing user area, the output port of the local dns in Shanghai, 0.1>,<Shanghai user area, The output port of Pekinese's local dns, 0>With<Shanghai user area, the output port of the local dns in Shanghai, 1>.
6th, the log recording generated to the CDN node server of Pekinese's CDN node server and Shanghai is respectively according to IP The conversion process in IP- regions is carried out with the correspondence in region, obtained correspondence is：<User IP affiliated area, domain name, CDN node server, flowing of access and timestamp>.
7th, according to the correspondence obtained in step 6, the log recording for belonging to the same area is restrained, is polymerize, is obtained To the correspondence for including total flowing of access that user area and the user region generate on the CDN node server.
For example, if the user of Shanghai user area does not generate flow on Pekinese's CDN node server, then can To obtain following three correspondences：<Shanghai user area, a.test.com, the CDN node server in Shanghai, 3000, 1476093274>；<Beijing user area, a.test.com, the CDN node server in Shanghai, 1000,1476093274>And< Beijing user area, a.test.com, Pekinese's CDN node server, 6000,1476093274>, above three correspondence It is for the correspondence of expression：<User area, domain name, CDN node server, flowing of access, timestamp>, further, It is according to the total flow that above three correspondence can obtain generating when Shanghai user area accesses a.test.com 3000Bytes, it is 7000Bytes that Beijing user area, which accesses the total flow generated during a.test.com,.
The total flow generated when the 8th, accessing a.test.com according to the correspondence obtained in step 5 and Beijing user area It is that the total flow that Pekinese CDN node server introduces is 6300Byes that can determine Pekinese's local dns, Pekinese DNS is that the total flow that the CDN node server in Shanghai introduces is 700Byes, according to the correspondence and Shanghai obtained in step 5 User area accesses the total flow generated during a.test.com, it may be determined that goes out the local dns in Shanghai and is taken for Pekinese's CDN node The total flow that device introduces of being engaged in is 0Byes, and the local dns in Shanghai are that the total flow of the CDN node server introducing in Shanghai is 3000Byes.Further, it is total stream that Pekinese's CDN node server introduces by local dns when accessing a.test.com It measures as 6300Byes, the total flow introduced by the CDN node server that local dns are Shanghai is 3700Byes.
9th, when the user in the user area of Beijing has accessed multiple domain names, that is, the plurality of domain name constitutes a scheduling Domain name, can according to Beijing user access the plurality of domain name when, in Pekinese the total flow in the output port region of local dns and The total flow in the output port region of local dns in Shanghai determines the flow of the scheduling domain name.
10th, when the flow for dispatching domain name when one is 8000Byes, the historical traffic of the scheduling domain name is recorded as 10000Byes, if the weight coefficient of the flow of current scheduling domain name is set as 0.6, the power of the historical traffic of the scheduling domain name Weight coefficient is 0.4, and the predicted flow rate of the scheduling domain name of acquisition is：8000Byes*0.6+10000Byes*0.4= 8800Byes。
11, after the predicted flow rate of scheduling domain name is obtained, global traffic prediction is carried out according to predicted flow rate.
In a concrete implementation scheme, when carrying out global traffic prediction, according to the predicted flow rate in corresponding CDN Match somebody with somebody floating resources in node server for the dispatching zone status.
Such as：Pekinese's CDN node server provides service, and the predicted flow rate of the scheduling domain name for the scheduling domain name For 8800Byes, then the floating resources of 8800Byes are allocated in advance for the scheduling domain name in Pekinese's CDN node server, it is northern The CDN node server in capital provides service using the floating resources of the 8800Byes for the scheduling domain name, but if Pekinese Maximum stream flow resource in CDN node server is less than 8800Byes, then needs to increase a CDN node server newly so that former The sum of Pekinese's CDN node server and the newly-increased floating resources of CDN node server for coming are not less than 8800Byes, then It is allocated in advance in original Pekinese's CDN node server and newly-increased CDN node server for the scheduling domain name The floating resources of 8800Byes, Pekinese's CDN node server and the use of newly-increased CDN node server originally should The floating resources of 8800Byes provide service for the scheduling domain name；Either, if had in the CDN node server in Shanghai idle Floating resources, and the idle floating resources and the sum of the maximum stream flow resource in Pekinese's CDN node server be not small In 8800Byes, then the idle floating resources can also be distributed to scheduling domain name use, i.e. Pekinese's CDN node takes Business device and the CDN node server in Shanghai provide floating resources for the scheduling domain name jointly.
Embodiment six
This programme embodiment also proposes a kind of volume forecasting device, which can be respectively used to realize embodiment one, implement The method flow proposed in example two, embodiment three and example IV, specifically as shown in figure 9, the device includes：
Flow collection unit 71, for gathering the flow from each user area；
Flow accounting computing unit 72, in the flow from each user area gathered, determining the user The accounting in the output port region of each local dns of flux and flow direction in region；
Traffic statistics unit 73, for each according to the flow from each user area and the flux and flow direction of each user area The accounting in the output port region of local dns determines the total flow in the output port region of each local dns.
The flow collection unit 71, is specifically used for：
According to the corresponding flow of User IP, IP and the mapping relations in region, it is useful that acquisition belongs to institute in same user area The total flow of family IP.
The flow accounting computing unit 72, is specifically used for：
Obtain first flow ratio；The first flow ratio represents input of the single user region to single local dns The access times of port areas account for ratio of the user region to the access times in the input port region of all local dns；
According to the mapping relations and the first flow ratio in input port region and output port region, second is obtained Flow proportional, using the accounting in the output port region of each local dns of flux and flow direction as the user region；The second flow Ratio represents that single user region accounts for the user region to all to the access times in the output port region of single local dns The ratio of the access times in the output port region of ground DNS.
When the flow accounting computing unit 72 is used to obtain first flow ratio, including：
Access times of each User IP to the input port IP of each local dns are counted, same user area will be belonged to User IP is clustered, and obtains access times of each user area for the input port IP of each local dns, to be visited as first Ask number；
Based on first access times, the input port in the input port region of same local dns will be belonged to IP is clustered, and obtains access times of each user area to the input port region of each local dns, secondary to be accessed as second Number；
According to the first access times and the second access times, first flow ratio is obtained.
In a specific embodiment, described device further includes：
Dispatching zone name flow rate calculation unit 74, total flow and area for the output port region according to each local dns Domain and the mapping relations of scheduling domain name obtain the flow of scheduling domain name.
In a specific embodiment, described device further includes：
Dispatching zone name flow predicting unit 75, for according to it is described scheduling domain name flow and, with the scheduling domain name Relevant historical traffic record, obtains and the relevant predicted flow rate of dispatching zone name.
In this programme embodiment, after flow collection unit acquires the flow from each user area, just obtain Belong to the total flow of all users generation in the same area, then flow accounting computing unit is being gathered from each use In the flow in family region, determine the accounting in the output port region of each local dns of flux and flow direction in the user region, at this time can With determine the user in each region generate flow each local dns flow export ratio, traffic statistics unit further according to The accounting in the output port region of each local dns of flux and flow direction of flow and each user area from each user area, really It is each to define the flux and flow direction that the user in all areas generates for the total flow in the output port region of fixed each local dns The specific flow in local dns output port region, further, according to the output port region for each local dns determined Total flow understands that the output port region of each local dns is the flow that corresponding CDN node server introduces, due to CDN node The flowing of access of server is introduced according to the output port region of local dns, therefore according to the output port of each local dns The total flow in region carrys out the accuracy that predicted flow rate improves the flow predicted.
Embodiment seven
Figure 10 is the simplified block diagram of CDN dispatch servers 100.The CDN dispatch servers 100 can include and one or more The processor 101 of a data storage facility connection, the data storage facility can include storage medium 102 and internal storage location 103. CDN dispatch servers 100 can also include input interface 104 and output interface 105, for being led to another device or system Letter.It is storable in by the CPU of processor 101 program codes performed in internal storage location 103 or storage medium 102.
Processor 101 in CDN dispatch servers 100 calls the program for being stored in internal storage location 103 or storage medium 102 Code can perform following each step：
Gather the flow from each user area；
In the flow from each user area gathered, each local dns of flux and flow direction in the user region are determined Output port region accounting；
According to the defeated of each local domain name system DNS of the flux and flow direction of the flow from each user area and each user area The accounting in exit port region determines the total flow in the output port region of each local dns.
In a concrete implementation scheme, the processor 101 is additionally operable to according to the corresponding flow of User IP, IP and area The mapping relations in domain obtain the total flow for belonging to all User IPs in same user area.
In a concrete implementation scheme, the processor 101 is additionally operable to obtain first flow ratio；It is described first-class Amount ratio represents that single user region accounts for the access times in the input port region of single local dns in the user region to all The ratio of the access times in the input port region of local dns；It is closed according to the mapping in input port region and output port region System and the first flow ratio obtain second flow ratio, using each local dns of flux and flow direction as the user region The accounting in output port region；The second flow ratio represents output terminal mouth region of the single user region to single local dns The access times in domain account for ratio of the user region to the access times in the output port region of all local dns.
In a concrete implementation scheme, the processor 101 is additionally operable to count each User IP to the defeated of each local dns The access times of inbound port IP cluster the User IP for belonging to same user area, obtain each user area for each The access times of the input port IP of local dns, using as the first access times；It, will based on first access times The input port IP for belonging to the input port region of same local dns is clustered, and obtains each user area to each local dns Input port region access times, using as the second access times；According to the first access times and the second access times, obtain Obtain first flow ratio.
In a concrete implementation scheme, the processor 101 is additionally operable to the output port region according to each local dns Total flow and region with scheduling domain name mapping relations, obtain scheduling domain name flow.
In a concrete implementation scheme, the processor 101 is additionally operable to the flow according to the scheduling domain name, with And historical traffic record relevant with the dispatching zone name, it obtains and the relevant predicted flow rate of dispatching zone name.
In above-described embodiment six, storage medium can be read-only memory (Read-Only Memory, ROM) or can Read-write, such as hard disk, flash memory.Internal storage location can be random access memory (Random Access Memory, RAM).It is interior Memory cell can be with processor physical integration or integrated in memory or being configured to individual unit.
Processor is the control centre of above-mentioned CDN dispatch servers, and provides processing unit, for executing instruction, carries out Interrupt operation provides clocking capability and various other functions.Optionally, processor includes one or more central processing unit (CPU), the CPU 0 and CPU 1 such as in Fig. 9 shown.Above equipment includes one or more processor.Processor can For monokaryon (single CPU) processor or multinuclear (multi -CPU) processor.Unless otherwise stated, it is described as performing task for example The component of processor or memory can be realized as universal component, be temporarily used for performing task in given time or be embodied as specially Door manufactures to perform the particular elements of the task.Terminology used herein " processor " refers to one or more devices, circuit and/ Or process cores, for handling data, such as computer program instructions.
It is storable in by the CPU of the processor program codes performed in internal storage location or storage medium.Optionally, it is stored in Program code in storage medium can be copied into internal storage location and be performed so as to the CPU of processor.Processor can perform at least One kernel (such as LINUXTM、UNIXTM、WINDOWSTM、ANDROIDTM、IOSTM), it is well known that the kernel is used to pass through control Execution, control and the communication of peripheral unit and the use of control computer device resource of other programs or process are made to control The operation of above equipment.
Said elements in above equipment can be connected to each other by bus, bus such as data/address bus, address bus, control One of bus, expansion bus and local bus or its any combination.
In the several embodiments provided in this programme, it should be understood that disclosed apparatus and method can pass through it Its mode is realized.For example, the apparatus embodiments described above are merely exemplary, for example, the division of the unit, only Only a kind of division of logic function, can there is other dividing mode in actual implementation, for example, multiple units or component can be tied It closes or is desirably integrated into another system or some features can be ignored or does not perform.It is another, it is shown or discussed Mutual coupling, direct-coupling or communication connection can be the INDIRECT COUPLING or logical by some interfaces, device or unit Letter connection can be electrical, machinery or other forms.
The unit illustrated as separating component may or may not be physically separate, be shown as unit The component shown may or may not be physical location, you can be located at a place or can also be distributed to multiple In network element.Some or all of unit therein can be selected to realize the mesh of this embodiment scheme according to the actual needs 's.
In addition, each functional unit in each embodiment of this programme can be integrated in a processing unit, it can also That unit is individually physically present, can also two or more units integrate in a unit.Above-mentioned integrated list The form that hardware had both may be employed in member is realized, can also be realized in the form of hardware adds SFU software functional unit.
The foregoing is merely the preferred embodiment of this programme, not limiting this programme, all essences in this programme God and any modification, equivalent substitution, improvement and etc. within principle, done, should be included within the scope of this programme protection.

Claims (12)

1. a kind of method for predicting, which is characterized in that the described method includes：
Gather the flow from each user area；
In the flow from each user area gathered, determine the user region each local dns of flux and flow direction it is defeated The accounting in exit port region；
According to the output terminal of each local domain name system DNS of the flux and flow direction of the flow from each user area and each user area The accounting in mouth region domain determines the total flow in the output port region of each local dns.
2. according to the method described in claim 1, it is characterized in that, acquisition the flow from each user area, including：
According to the corresponding flow of User IP, IP and the mapping relations in region, acquisition belongs to all User IPs in same user area Total flow.
3. according to the method described in claim 1, it is characterized in that, in the flow from each user area gathered, Determine the accounting in the output port region of each local dns of flux and flow direction in the user region, including：
Obtain first flow ratio；The first flow ratio represents input port of the single user region to single local dns The access times in region account for ratio of the user region to the access times in the input port region of all local dns；
According to the mapping relations and the first flow ratio in input port region and output port region, second flow is obtained Ratio, using the accounting in the output port region of each local dns of flux and flow direction as the user region；The second flow ratio Represent that single user region accounts for the user region to all locals to the access times in the output port region of single local dns The ratio of the access times in the output port region of DNS.
4. according to the method described in claim 3, it is characterized in that, obtain first flow ratio, including：
Access times of each User IP to the input port IP of each local dns are counted, the user of same user area will be belonged to IP is clustered, and obtains access times of each user area for the input port IP of each local dns, secondary to be accessed as first Number；
Based on first access times, will belong to same local dns input port region input port IP into Row cluster, obtains access times of each user area to the input port region of each local dns, using as the second access times；
According to the first access times and the second access times, first flow ratio is obtained.
5. according to the method described in claim 1, it is characterized in that, the method further includes：
According to the total flow in the output port region of each local dns and region and the mapping relations of scheduling domain name, dispatched The flow of domain name.
6. according to the method described in claim 5, it is characterized in that, the method further includes：
And, relevant with dispatching zone name historical traffic record obtains and the tune according to the flow of the scheduling domain name Spend the relevant predicted flow rate of domain name.
7. a kind of volume forecasting device, which is characterized in that described device includes：
Flow collection unit, for gathering the flow from each user area；
Flow accounting computing unit, in the flow from each user area gathered, determining the user region The accounting in the output port region of each local dns of flux and flow direction；
Traffic statistics unit, for each local of flux and flow direction according to flow and each user area from each user area The accounting in the output port region of DNS determines the total flow in the output port region of each local dns.
8. device according to claim 7, which is characterized in that the flow collection unit is specifically used for：
According to the corresponding flow of User IP, IP and the mapping relations in region, acquisition belongs to all User IPs in same user area Total flow.
9. device according to claim 7, which is characterized in that the flow accounting computing unit is specifically used for：
Obtain first flow ratio；The first flow ratio represents input port of the single user region to single local dns The access times in region account for ratio of the user region to the access times in the input port region of all local dns；
According to the mapping relations and the first flow ratio in input port region and output port region, second flow is obtained Ratio, using the accounting in the output port region of each local dns of flux and flow direction as the user region；The second flow ratio Represent that single user region accounts for the user region to all locals to the access times in the output port region of single local dns The ratio of the access times in the output port region of DNS.
10. device according to claim 9, which is characterized in that when the flow accounting computing unit is for acquisition first During flow proportional, including：
Access times of each User IP to the input port IP of each local dns are counted, the user of same user area will be belonged to IP is clustered, and obtains access times of each user area for the input port IP of each local dns, secondary to be accessed as first Number；
Based on first access times, will belong to same local dns input port region input port IP into Row cluster, obtains access times of each user area to the input port region of each local dns, using as the second access times；
According to the first access times and the second access times, first flow ratio is obtained.
11. device according to claim 7, which is characterized in that described device further includes：
Dispatching zone name flow rate calculation unit, for the total flow in the output port region according to each local dns and region and tune The mapping relations of domain name are spent, obtain the flow of scheduling domain name.
12. according to the devices described in claim 11, which is characterized in that described device further includes：
Dispatching zone name flow predicting unit, for according to it is described scheduling domain name flow and, it is relevant with the dispatching zone name Historical traffic records, and obtains and the relevant predicted flow rate of dispatching zone name.
CN201610997850.2A 2016-11-11 2016-11-11 Flow prediction method and device Active CN108075909B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201610997850.2A CN108075909B (en) 2016-11-11 2016-11-11 Flow prediction method and device

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201610997850.2A CN108075909B (en) 2016-11-11 2016-11-11 Flow prediction method and device

Publications (2)

Publication Number Publication Date
CN108075909A true CN108075909A (en) 2018-05-25
CN108075909B CN108075909B (en) 2020-12-18

Family

ID=62161674

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201610997850.2A Active CN108075909B (en) 2016-11-11 2016-11-11 Flow prediction method and device

Country Status (1)

Country Link
CN (1) CN108075909B (en)

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110138684A (en) * 2019-04-01 2019-08-16 贵州力创科技发展有限公司 A kind of flux monitoring method and system based on DNS log
CN111327495A (en) * 2020-02-25 2020-06-23 网宿科技股份有限公司 Method and device for detecting IP service area of visitor
CN116016220A (en) * 2022-12-23 2023-04-25 天翼安全科技有限公司 Method, device and equipment for predicting service traffic based on DNS traffic

Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20080019269A1 (en) * 2006-07-18 2008-01-24 Chatschik Bisdikian Proactive mechanism for supporting the global management of vehicle traffic flow
CN102868550A (en) * 2012-08-17 2013-01-09 新浪网技术(中国)有限公司 Total network flow scheduler and method for querying domain name resolution record by using total network flow scheduler
CN103650569A (en) * 2013-07-22 2014-03-19 华为技术有限公司 Fault diagnosis method and device of wireless network
US20140115704A1 (en) * 2012-10-24 2014-04-24 Hewlett-Packard Development Company, L.P. Homoglyph monitoring
CN105871734A (en) * 2015-01-22 2016-08-17 阿里巴巴集团控股有限公司 Method and device for acquiring traffic distribution information of local domain name system
CN106027516A (en) * 2016-05-17 2016-10-12 中国互联网络信息中心 Domain name service security event evaluation method and system

Patent Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20080019269A1 (en) * 2006-07-18 2008-01-24 Chatschik Bisdikian Proactive mechanism for supporting the global management of vehicle traffic flow
CN102868550A (en) * 2012-08-17 2013-01-09 新浪网技术(中国)有限公司 Total network flow scheduler and method for querying domain name resolution record by using total network flow scheduler
US20140115704A1 (en) * 2012-10-24 2014-04-24 Hewlett-Packard Development Company, L.P. Homoglyph monitoring
CN103650569A (en) * 2013-07-22 2014-03-19 华为技术有限公司 Fault diagnosis method and device of wireless network
CN105871734A (en) * 2015-01-22 2016-08-17 阿里巴巴集团控股有限公司 Method and device for acquiring traffic distribution information of local domain name system
CN106027516A (en) * 2016-05-17 2016-10-12 中国互联网络信息中心 Domain name service security event evaluation method and system

Cited By (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110138684A (en) * 2019-04-01 2019-08-16 贵州力创科技发展有限公司 A kind of flux monitoring method and system based on DNS log
CN110138684B (en) * 2019-04-01 2022-04-29 贵州力创科技发展有限公司 Traffic monitoring method and system based on DNS log
CN111327495A (en) * 2020-02-25 2020-06-23 网宿科技股份有限公司 Method and device for detecting IP service area of visitor
CN111327495B (en) * 2020-02-25 2021-08-27 网宿科技股份有限公司 Method and device for detecting IP service area of visitor
CN116016220A (en) * 2022-12-23 2023-04-25 天翼安全科技有限公司 Method, device and equipment for predicting service traffic based on DNS traffic
CN116016220B (en) * 2022-12-23 2024-06-18 天翼安全科技有限公司 Method, device and equipment for predicting service traffic based on DNS traffic

Also Published As

Publication number Publication date
CN108075909B (en) 2020-12-18

Similar Documents

Publication Publication Date Title
EP1482712B1 (en) Virtual network addresses
US5974237A (en) Communications network monitoring
CN105279027B (en) A kind of virtual machine deployment method and device
JP2002024192A (en) Device and method for dividing computer resources
CN105144109B (en) Distributive data center technology
EP0849912A2 (en) Communications network monitoring
CN106170947A (en) A kind of alarm information processing method, relevant device and system
CN112333290A (en) Data access control method, device, storage medium and content distribution network system
CN108075909A (en) A kind of method for predicting and device
CN101778013A (en) Website testing system and method
CN109921925A (en) A kind of dial testing method and device
JP2008003907A (en) Method for controlling information processing distribution and computer therefor
CN109639796A (en) A kind of implementation of load balancing, device, equipment and readable storage medium storing program for executing
CN112491719A (en) Network node selection method, equipment and storage medium
EP0849911A2 (en) Communications network monitoring
CN109246257A (en) Flow allocating method, apparatus, computer equipment and storage medium
CN109710447A (en) For the method, apparatus of data access, medium and calculate equipment
CN102929693A (en) Performance evaluation method and device for servers of whole equipment cabinet
EP0849910A2 (en) Communications network monitoring
CN111913784B (en) Task scheduling method and device, network element and storage medium
CN108347465A (en) A kind of method and device of selection network data center
CN111447282A (en) Method and apparatus for determining transmission path
CN112732451A (en) Load balancing system in cloud environment
CN112383453B (en) Data processing method and system
RU2798799C1 (en) Method for collecting and storing network data obtained from broadcasting an ip address

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
REG Reference to a national code

Ref country code: HK

Ref legal event code: DE

Ref document number: 1255824

Country of ref document: HK

GR01 Patent grant
GR01 Patent grant