CN109802847A - A kind of analysis method of network transmission service quality, device - Google Patents
A kind of analysis method of network transmission service quality, device Download PDFInfo
- Publication number
- CN109802847A CN109802847A CN201711146847.0A CN201711146847A CN109802847A CN 109802847 A CN109802847 A CN 109802847A CN 201711146847 A CN201711146847 A CN 201711146847A CN 109802847 A CN109802847 A CN 109802847A
- Authority
- CN
- China
- Prior art keywords
- node
- scene
- transport services
- network
- matrix
- 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.)
- Pending
Links
Landscapes
- Data Exchanges In Wide-Area Networks (AREA)
Abstract
A kind of analysis method of network transmission service quality, device, analytical equipment is according to the constraint condition of the scene of setting, the transport services record of the node of offer transport services under this scenario is obtained, the scene refers to the transport services environment that the constraint condition limits；The network transmission service quality information of the node under this scenario is determined according to transport services record, and the transport services information includes the statistical value of network transfer speeds.It can also be to node sequencing according to determining network transmission service quality.The application can reflect the network transmission service quality of node, can also carry out node scheduling according to ranking results.
Description
Technical field
The present invention relates to network transmissions, more particularly, to the analysis method and device of a kind of network transmission quality.
Background technique
CDN (Content Delivery Network, content distributing network) is constructed on network, each by being deployed in
The Edge Server on ground obtains user nearby by functional modules such as the load balancing, content distribution, scheduling of central platform
Required content reduces network congestion, improves user's access response speed and hit rate.CDN node be provided for CDN user it is all kinds of
The node of content (picture, audio-video, webpage etc.) download service is one group in a certain operator room in a certain geographic area
The set of server.
In the related technology, the network transmission shape at CDN node is reflected using indexing sections such as round-trip delay, packet loss
Condition, can not be well reflected CDN node offer download service quality, be not suitable as CDN node scheduling according to
According to.
Summary of the invention
The embodiment of the invention provides a kind of analysis methods of network transmission service quality, including calculation method below:
According to the constraint condition of the scene of setting, the network of the node of offer transport services under this scenario is obtained
Service log is transmitted, the scene refers to the transport services environment that the constraint condition limits；
The network transmission service quality letter of the node under this scenario is determined according to transport services record
Breath, the transport services information includes the statistical value of network transfer speeds.
The embodiment of the invention also provides a kind of analytical equipment of network transmission quality, including computing module, the calculating
Module includes:
Acquiring unit is recorded, for the constraint condition according to the scene of setting, obtains and the node of transport services is provided
Transport services record under this scenario, the scene refer to the transport services environment that the constraint condition limits；
Quality determination unit, for determining the net of the node under this scenario according to transport services record
Network transmission service quality information, the transport services information include the statistical value of network transfer speeds.
The embodiment of the invention also provides a kind of analytical equipment of network transmission quality, including memory, processor and deposit
The computer program that can be run on the memory and on the processor is stored up, the processor executes the computer journey
The processing of method as described above is realized when sequence.
The embodiment of the invention also provides a kind of computer readable storage mediums, are stored thereon with computer program, described
The processing of method as described above is realized when computer program is executed by processor.
Above-described embodiment branch scape determines the statistical value of the network transfer speeds of node, and the network for carrying out concentrated expression node passes
Defeated service quality,
The problem of present invention mainly solves the scorings of the network transmission quality of CDN nodes numerous in CDN system, to detect
The performance good job of CDN node, and then foundation is provided for CDN request scheduling.
Detailed description of the invention
Fig. 1 is the flow chart of one method of the embodiment of the present invention；
Fig. 2 is the cellular construction figure of one computing module of the embodiment of the present invention；
Fig. 3 is the flow chart of two method of the embodiment of the present invention；
Fig. 4 is the cellular construction figure of 2 first sorting module of the embodiment of the present invention；
Fig. 5 is the flow chart of three method of the embodiment of the present invention；
Fig. 6 is the cellular construction figure of 3 second sorting module of the embodiment of the present invention.
Specific embodiment
To make the objectives, technical solutions, and advantages of the present invention clearer, below in conjunction with attached drawing to the present invention
Embodiment be described in detail.It should be noted that in the absence of conflict, in the embodiment and embodiment in the application
Feature can mutual any combination.
Embodiment one
In CDN network, the quality for the download service that CDN node provides (is also referred to as the download service matter of CDN node in text
Amount) situation of network transmission is depended not only on, and also (whether such as node there is CPU, memory provides with CDN node resource situation
Source is nervous), the variant factors decision such as the scheduling strategy (suitable scheduling can more efficiently utilize resource) of CDN node.
And for download service, most can synthesis to show the index of downloading service quality be exactly speed of download, the shape of above-mentioned network transmission
The superiority and inferiority of condition, CDN node resource situation and scheduling strategy can be finally embodied in speed of download.But the present application
People's discovery, although for the service request of single, speed of download can be calculated, for providing the CDN of magnanimity service
For node, how to reflect that the downloading service quality of node is to need the problem of studying using the speed of download of CDN node.This
It is because the speed of download of node can be influenced by many factors.For example, for same node, morning 3:00-5:00 this
The period of this many people's online of period and the evening 7:00-8:00 of the few's online, speed of download may differ by very
It is more.For another example, request the terminal of downloading at different regions, for example, CDN node is at Beijing, the terminal and one in a Shenzhen
When the download service that the terminal in a Beijing is provided using the CDN node, speed of download difference is also very big.Other factors are also
Internet Service Provider, the domain name that terminal is accessed and type of terminal that terminal is contracted etc..
Constraint condition of the present embodiment by the scene of setting, the statistical value of the speed of download of branch scape calculating CDN node,
Downloading service quality of the CDN node under each scene is determined according to calculated result, so as to more precisely to CDN node
Downloading service quality analyzed.But the application is not limited to the analysis to the downloading service quality of CDN node, at other
In embodiment, the node of transport services can also be provided for other, for example, providing the node of upload service, Huo Zhetong
When provide upload service and download service node, the network transmission service quality of node is determined by the way of the scape of branch, is removed
It may include outside downloading service quality, can also include upload service quality.
The analysis method for present embodiments providing a kind of network transmission service quality, for determining the net of node by analyzing
Network transmission service quality.Node therein is content delivery network node, and transport services are download service, network transmission speed
Degree is speed of download.But the application is not limited to this, can be used for providing other nodes of transport services, the network
Transmission service is also possible to upload service, or simultaneously includes upload service and download service.The network transfer speeds of statistics
It can be uploading speed, or simultaneously include uploading speed and speed of download.
The analysis method of the present embodiment network transmission service quality includes the network transmission service quality for determining node
Calculation method, as shown in Figure 1, the calculation method includes:
Step 110, according to the constraint condition of the scene of setting, the node that transport services are provided is obtained in the scene
Under transport services record, the scene refers to the transport services environment that the constraint condition limits；
In the present embodiment, the constraint condition of the scene includes the constraint to the value of one or more of parameter:
Period；
Use the region where the terminal of the transport services；
The Internet Service Provider that the terminal is contracted；
The domain name that the terminal is accessed；
The type of the terminal.
In an example of the present embodiment, this step is completed in the following manner:
The first step configures the constraint condition of scene for CDN node；
For example, being configured as needed to following field in tables of data to some CDN node: saving name, ISP
Title, domain name, terminal type.Wherein, saving name is in above-mentioned constraint condition using the ground where the terminal of the transport services
One example in domain, ISP title refer to the Internet Service Provider that the terminal is contracted, and the value of the two fields can configure
The title of the province's name and the affiliated ISP of the CDN node that are saved where the CDN node, but other values can also be taken.Domain name refers to terminal institute
The domain name of access, such as " top news ", " quick worker " etc..The type of terminal can for example be divided into apple terminal and Android (Android) eventually
End, the downloading behavior of different types of terminal may have differences.The same same operator in province may have multiple nodes, example
If Guangdong Telecom has 10 nodes, each node services multiple domain names again, for example services " top news " and " quick worker " simultaneously, because
This, a CDN node may correspond to one or more record.It should be noted that above-mentioned field can be combined differently,
For example, terminal type field, etc. can not be configured.In combination difference, the constraint of scene has the difference loosened, tighten, but
Certain interference can be excluded.
In above-mentioned tables of data, it also is configured with log derived data library name and log collection time field for CDN node,
Wherein, log derived data library name is used to identify the database as data source, to obtain from the log of the database
The Download History of CDN node.And the log collection time configured in log collection time field then define to obtain which section when
Interior log.
In one tables of data above-mentioned configuration can be completed for multiple CDN nodes.
Second step generates log based on above-mentioned configuration information and extracts sentence, executes the log extraction sentence and extracts phase
The daily record data answered；
Based on the record in the tables of data postponed, corresponding sentence can be automatically generated, here by taking SQL statement as an example,
For extracting corresponding daily record data.In constructing SQL statement, every row record of above-mentioned tables of data can automatically generate one
SQL statement, province's name of first step configuration, ISP title, domain name, terminal type constitute the querying condition of SQL statement, the first step
The log collection time of configuration defines the daily record data that needs to acquire and generate in which time range.The SQL statement of generation can
It is spare in a new database table to be stored in.Required daily record data can be extracted by executing these SQL statements,
In include CDN node under the constraint condition of configuration download service record.
Step 120, determine that the network transmission of the node under this scenario takes according to transport services record
Business quality information, the transport services information includes the statistical value of network transfer speeds.
In the examples described above, it is based on SQL statement, after the daily record data of corresponding log collection time is extracted, by this
The log collection time is divided into multiple periods (or being time slice), for example, the log collection time is configured to daily 17:00
~21:00, can by the log collection time again score at multiple periods, the duration of each time slice can for 10 seconds,
20 seconds or other values.Because of the segmentation of time, under primary complete download service process may be divided under multiple scenes
Carry service.
It is province, ISP, domain name, terminal type and the time slice of CDN node configuration together as about in this example
Beam condition, defines an a scene i.e. transport services environment, under the transport services environment, requests the end of service
End is in identical province, the identical ISP of signing, the identical domain name of access, terminal type having the same, and time interval is very
Small (less than the duration such as 10 seconds or so of a period), therefore, the download service under Same Scene is regarded as basic isomorphism
, the comparison that the network transmission service quality of node is carried out under conditions of an isomorphism can exclude very more other factors and do
It disturbs.
To a CDN node, many download services may be performed in one scenario, record under corresponding in log
Record can achieve thousands of to tens of thousands of.According to the byte number and download time downloaded in each Download History, can calculate
Multiple speeds of download.In the present embodiment, according to the multiple speeds of download being calculated under a scene, determination is compared at a slow speed, as
The statistical value of the CDN node speed of download under the scene, to embody VELOCITY DISTRIBUTION situation of the CDN node under the scene.Slowly
Speed ratio refers in the Download History of a node that speed of download is less than the record proportion of a certain threshold value (such as 256KBps).
At a slow speed than smaller, node quality-of-service is better；Otherwise it is poorer.
Because the log of acquisition generally comprises many time slices, and a CDN node is configured with a plurality of record,
The thus log based on extraction, can be calculated comparing at a slow speed under multiple scenes to a CDN node.And according to being multiple
The tables of data of CDN node configuration carries out log to each CDN node and extracts and above-mentioned calculating, so that it may be calculated described more
A CDN node comparing at a slow speed under multiple scenes.
In other embodiments, average transmission speed or quartile spot speed can also be used as the CDN node under the scene
The statistical value of speed of download.In addition, statistical value also can be used multiple, such as compare at a slow speed, average transmission speed and quartile spot speed
In it is a variety of.Wherein, above-mentioned quartile spot speed can be 50% quartile spot speed i.e. median, be also possible to 10% quartile spot speed
Degree, 20% quartile spot speed ..., 90% quartile spot speed etc..For example, for the set of a speed of download, it is assumed that there is 100
Record, sorts from small to large to speed of download, the value for coming the 20th speed of download is exactly 20% quartile spot speed.Together
Reason, coming the 50th is exactly 50% quartile spot speed, and so on.
The present embodiment additionally provides a kind of analytical equipment of network transmission quality, can be set in CDN network for holding
In the equipment of row CDN node scheduling, but also it can be set in other equipment such as CDN node, or be independently arranged.The analysis
Device is connect with data source (such as the database for saving daily record data), is obtained from the data source and is recorded under CDN node
Record, branch scape determine the downloading service quality of node.The analytical equipment can use distributed deployment mode, such as each region
Setting one, or it is arranged one for one group of CDN node；It can also be by the way of concentrating deployment, as set in a CDN network
One is set, mixed deployment mode can also be used, the application, which does not do this, to be limited to.
The analytical equipment of the present embodiment network transmission quality includes computing module, as shown in Fig. 2, the computing module packet
It includes:
Acquiring unit 10 is recorded, for the constraint condition according to the scene of setting, obtains and the section of transport services is provided
The transport services record of point under this scenario, the scene refer to the transport services ring that the constraint condition limits
Border；
Quality determination unit 20, for determining the node under this scenario according to transport services records
Network transmission service quality information, the transport services information include the statistical value of network transfer speeds.
In the present embodiment,
According to the constraint condition of the scene of setting, the node for obtaining offer transport services exists the acquiring unit that records
Transport services record under the scene, wherein the constraint condition of the scene includes to one or more of parameter
Value constraint:
Period；
Use the region where the terminal of the transport services；
The Internet Service Provider (ISP:Internet Service Provider) that the terminal is contracted；
The domain name that the terminal is accessed；
The type of the terminal.
In the present embodiment,
The statistical value of the network transfer speeds includes: the ratio at a slow speed counted to multiple network transfer speeds, average
One of transmission speed and quartile spot speed are a variety of.
The present embodiment just obtains by the constraint condition of the scene of setting and is based on tool to the statistics of the network transfer speeds of node
The scene of body carries out, and more accurately can reflect to more fine granularity the network transmission service quality of node, meanwhile, also for not
It lays a good foundation with the comparison for carrying out network transmission service quality between node.
Embodiment two
When being scheduled to CDN node, it is sometimes desirable to it is sorted according to the downloading service quality of CDN node to CDN node,
To meet the needs of scheduling strategy, for example, the service priority that the user of high priority requests is dispatched to the fast CDN of speed of download
Node.But if not dividing scene, may because of scene difference and cause the distortions of ranking results, such as under X node
The Download History that the statistical value of load speed is mostly based under A scene obtains, and the statistical value of Y node speed of download is then mostly based on
Download History under B scene obtains, and the speed of download of A scene itself is generally higher than B scene, may obtain X section in this way
Point speed of download is better than the conclusion of Y node, but if the speed of download of Y node may be better than X node under Same Scene.
A kind of method for present embodiments providing branch scape sequence, the network transmission based on multiple nodes under Same Scene
Service quality to the multiple node sequencing, ranking results can be used as under the scene to the multiple node be scheduled according to
According to.
A kind of analysis method of network transmission service quality is present embodiments provided, for carrying out branch scape to multiple nodes
Sequence processing, as shown in Figure 3, comprising:
Step 210, the statistical value of the network transfer speeds of multiple nodes to be sorted under the first scene is calculated separately；
In the present embodiment, the statistical value of network transfer speeds of the node under the first scene can be according to the side of embodiment one
Method is calculated.Here the first scene is used for a certain transport services environment for indicating to limit according to constraint condition, and
It is not representing a kind of special scene.Here multiple nodes can be the multiple nodes being uniformly scheduled in CDN network.
In the present embodiment, the first scene is by one of province, ISP, domain name, terminal type and time slice or a variety of ginsengs
Several values limits.Thus under identical scene, it can calculate separately to obtain according to the log of multiple CDN nodes multiple
The statistical value of the speed of download of CDN node.
The node of the present embodiment is also by taking CDN node as an example, and transport services are download service, under network transfer speeds are
Carry speed.But the application is not limited to this.
Step 220, according to the size of corresponding statistical value, to network transmission of the multiple node under first scene
Quality of service ordered.
In the present embodiment, according to the size of corresponding statistical value, to network of the multiple node under first scene
When transmission service quality sorts, it can only consider that the size of statistical value is come such as according to the sequence of the size of ratio at a slow speed, but be not required
In this way, in other embodiments, other parameters such as round-trip delay, packet loss etc. can also be added, based on the value of multiple parameters come
It is integrated ordered.One value can be calculated by modes such as weighted average, weighted accumulations to the value of the multiple parameter, further according to
The size of the value is to the multiple node sequencing.In one example, by CDN node each in the first scene according to corresponding slow
It after the ascending sequence of speed ratio, is stored in a database table, is used for node scheduling.
In another embodiment, the constraint condition of scene is defined using domain name and period, it is assumed that A node serve is
Quick worker and the two top domain names；B node service is microblogging and the two top domain names；And C node serve be quick worker and
The two domain names of microblogging.Since these three nodes do not service same domain name, the speed of download of direct comparison node can exist very
Large deviation.But A, B node can access this scene of top domain name based on the same period to compare；B, C node can be with base
This scene of microblogging domain name is accessed to compare in the same period, and such accuracy and precision are with regard to relatively high.
The analytical equipment of the present embodiment network transmission service quality includes the computing module in embodiment one, in addition to this,
It further include the first sorting module, as shown in figure 4, first sorting module includes:
Receiving unit 40 is passed for obtaining the network of multiple nodes to be sorted under the first scene from the computing module
The statistical value of defeated speed；
Sequencing unit 50, for the size according to corresponding statistical value, to the multiple node under first scene
Network transmission service quality sequence.
In the present embodiment, after carrying out the sequence of branch scape to multiple nodes, ranking results can be used as institute under corresponding scene
State the scheduling foundation of multiple nodes.For example, to multiple nodes after the network transmission service quality sequence under the first scene, row
Sequence result is then used as under the first scene, the foundation being scheduled to the multiple node.If receiving the clothes under the first scene
Business request, then search the ranking results of the multiple node under the first scene, meets service quality according to scheduling strategy selection and wants
The node asked, service request is dispatched on the node of selection.
In the present embodiment, can there is multiple comparison between multiple nodes, but be more all each time under Same Scene into
Capable, it that is to say that carrying out the service quality between different nodes in fine-grained scene level compares；Thus comparison result accuracy
It is good, the distortion of ranking results caused by avoiding because of scene difference, so as to effectively realize scheduling strategy.
Embodiment three
The method that the present embodiment determines the network transmission service quality of node also based on one branch scape of embodiment, embodiment two
A kind of branch scape is provided to the method for node sequencing, and the present embodiment is then there is provided a kind of integrated ranking method of node,
That is, comprehensive analysis is carried out according to network transmission service quality of multiple nodes under several scenes, to the multiple section
Point sequence, ranking results can be used as the foundation being scheduled to the multiple node, and when scheduling does not subdivide scene.
According to embodiment one, the network transmission service quality of node can be determined with branch scape.But according under multiple scenes
The speed of download of multiple nodes is difficult the multiple node sequencing, because of the downloading of first node in some scenes
Speed may be better than second node, and then be inferior to second node under other scenes, and the section to be compared in practical application
Point is numerous, and the intersection that can exist between many nodes compares, the deep-rooted situation of the superiority and inferiority relationship between each node, this is just needed
A kind of method is wanted to realize effective analysis.The present embodiment proposes building and state transition probability matrix and normalizes, pass through by
The mode of initial vector and matrix iteration obtains the scoring of each node so that the sequence between each node become can
Energy.
The analysis method for present embodiments providing a kind of network transmission service quality, for carrying out comprehensive row to multiple nodes
Sequence processing, as shown in Figure 5, comprising:
Step 310, the statistical value of the network transfer speeds of N number of node to be sorted under K scene is calculated separately, N >=
2, K >=2；
In the present embodiment, the statistical value of network transfer speeds of each node under each scene can be according to embodiment one
Method be calculated.
Step 320, N × N-dimensional state transition probability matrix M is constructed, wherein N row and the N column of matrix M respectively correspond institute
N number of node is stated, the value for the element that the i-th row jth arranges in matrix M is equal in the K scene, and the network of the i-th row corresponding node passes
The statistical value of defeated speed is inferior to the number of the scene of the statistical value of the network transfer speeds of jth column corresponding node, 1≤i, j≤N；
Step 330, the initial score vector of preset 1 × N-dimensional and matrix M ' are iterated operation, obtain the section of stable state
Point scoring vector, matrix M ' are the matrixes obtained to matrix M by row normalization；
Step 340, according to the size of N number of node value of corresponding element in node scoring vector, to the N
The network transmission service quality of a node sorts.
It is similar with embodiment two, according to N number of node the node scoring vector in corresponding element value it is big
It is small, when sorting to the network transmission service quality of N number of node, it can only consider the big of the value of corresponding element in scoring vector
It is small, other parameters can also be added, based on the value of multiple parameters come integrated ordered.
During calculated above, the change of some operational forms can be made, for example, " row " therein and " column " is exchanged,
By the initial score vector of 1 × N-dimensional be changed to N × 1 dimension vector " being inferior to " therein is for another example changed to " being better than ", then originally from
It is excellent to become the sequence, etc. from bad to excellent to bad sequence, be substantially it is the same, only in an operational form
Equivalents, such transformation, which still should belong to, to be defined within the scope of protection of the claims of this embodiment scheme.
In one example, all scenes for including within the log collection time of configuration to N number of node calculate, and obtain
To at a slow speed ratio of the N number of node under all scenes, constructed to obtain the shape of following N × N-dimensionals according to the comparison result of ratio at a slow speed
State transition probability matrix:
| Node 1 | Node 2 | …… | Node N | |
| Node 1 | 23 | 35 | 68 | |
| Node 2 | 12 | 43 | 38 | |
| …… | 32 | 17 | 22 | |
| Node N | 42 | 35 | 23 |
If in one scenario, node 1 at a slow speed than being inferior to second node, then in that the corresponding a line of node 1, will
The value of element in the corresponding column of node 2 adds 1.Thus in above-mentioned probability transfer matrix each element concrete meaning are as follows: its
Corresponding row node is inferior to its respective column degree of node at a slow speed frequently relatively.Since than smaller, node quality-of-service is got at a slow speed
It is good, thus its correspond to row node be inferior at a slow speed frequently relatively its respective column node namely its correspond to row node at a slow speed than big
In its respective column node.And if quartile spot speed or average speed of download is used to correspond to row node as statistical value and exist
It is inferior to its respective column node in the comparison of quartile spot speed or average speed of download, then it represents that its quartile spot speed for corresponding to row node
Degree or average speed of download are less than its respective column node.In addition, the node number marked in upper table is intended merely to facilitate understanding, do not make
For the element of matrix.
After obtaining above-mentioned transition probability matrix, to each row therein carry out by row normalize, also i.e. by each element divided by
The sum of the row element.
An initial score vector is first provided, is initialized by average value, as follows:
| 1.0/N | 1.0/N | 1.0/N | ……. | 1.0/N |
The scoring of each node is calculated using iterative method, when calculating, by above-mentioned vector and aforementioned normalization transition probability matrix
It is multiplied, obtains the matrix and a N*N matrix multiple of new node scoring vector namely a 1*N, obtain a 1*N
Then matrix will obtain new node scoring vector and be multiplied again with the normalization transition probability matrix, also obtains previous step
The 1*N matrix arrived with N*N matrix multiple, obtains another 1*N matrix again, and so on, until node scoring vector no longer becomes
Until change or the number of iterations have reached a certain threshold value (such as 100,000 times).Stable node scoring vector is finally obtained, wherein
Numerical value just corresponded to the scorings of different nodes, value is between 0 and 1, and bigger to represent meshed network transmission quality better.
The set-up mode of the analytical equipment of the network transmission service quality of the present embodiment can be the same as example 1, including
In addition to this computing module in embodiment one further includes the second sorting module, as shown in fig. 6, the second sorting module packet
It includes:
Receiving unit 60, for obtaining the network transmission of N number of node to be sorted under K scene from the computing module
The statistical value of speed, N >=2, K >=2；
Matrix construction unit 70, for constructing N × N-dimensional state transition probability matrix M, wherein the N row and N of matrix M arranges
N number of node is respectively corresponded, the value for the element that the i-th row jth arranges in matrix M is equal in the K scene, the corresponding section of the i-th row
Point network transfer speeds statistical value be inferior to jth column corresponding node network transfer speeds statistical value scene number, 1
≤i,j≤N；
Vector calculation unit 80, for the initial score vector of preset 1 × N-dimensional and matrix M ' to be iterated operation,
The node scoring vector of stable state is obtained, matrix M ' is the matrix obtained to matrix M by row normalization；
Sequencing unit 90, for according to N number of node the node scoring vector in corresponding element value size,
Network transmission service quality sequence to N number of node.
In the present embodiment, after integrated ordered to the progress of multiple nodes, ranking results can be with the scheduling of the multiple node
Foundation does not subdivide scene at this time, can simplify processing.
The above embodiment of the present invention, which is provided, to be carried out quantitative analysis to the network transmission service quality of CDN node and carries out
The scheme of sequence.According to the ranking results of multiple CDN nodes, more preferably node scheduling can be carried out for the multiple CDN node,
To obtain better CDN system service quality.
The serial number of the above embodiments of the invention is only for description, does not represent the advantages or disadvantages of the embodiments.Pass through above embodiment party
The description of formula, it is required general that those skilled in the art can be understood that above-described embodiment method can add by software
The mode of hardware platform is realized, naturally it is also possible to which by hardware, but in many cases, the former is more preferably embodiment.It is based on
Such understanding, substantially the part that contributes to existing technology can be with soft in other words for the technical solution of the embodiment of the present invention
The form of part product embodies, which is stored in a storage medium (such as ROM/RAM, magnetic disk, CD)
In, including some instructions use is so that a terminal device (can be mobile phone, computer, server or the network equipment etc.)
Execute method described in each embodiment of the present invention.
The foregoing is only a preferred embodiment of the present invention, is not intended to restrict the invention, for the skill of this field
For art personnel, the invention may be variously modified and varied.All within the spirits and principles of the present invention, made any to repair
Change, equivalent replacement, improvement etc., should all be included in the protection scope of the present invention.
Claims (12)
1. a kind of analysis method of network transmission service quality, including calculation method below:
According to the constraint condition of the scene of setting, the network transmission of the node of offer transport services under this scenario is obtained
Service log, the scene refer to the transport services environment that the constraint condition limits；
The network transmission service quality information of the node under this scenario, institute are determined according to transport services record
State the statistical value that transport services information includes network transfer speeds.
2. the method as described in claim 1, it is characterised in that:
The constraint condition of the scene includes the constraint to the value of one or more of parameter:
Period；
Use the region where the terminal of the transport services；
The Internet Service Provider that the terminal is contracted；
The domain name that the terminal is accessed；
The type of the terminal.
3. the method as described in claim 1, it is characterised in that:
The statistical value of the network transfer speeds include: to multiple network transfer speeds count compare at a slow speed, average transmission
One of speed and quartile spot speed are a variety of.
4. the method as described in claim 1, it is characterised in that:
The node is content delivery network node, and the transport services are download service, and the network transfer speeds are
Speed of download.
5. the method as described in any in claim 1-4, it is characterised in that:
The method also includes scape sequence in branch below processing:
According to the calculation method, the statistics of the network transfer speeds of multiple nodes to be sorted under the first scene is calculated separately
Value；
According to the size of corresponding statistical value, network transmission service quality of the multiple node under first scene is arranged
Sequence.
6. the method as described in any in claim 1-4, it is characterised in that:
The method also includes integrated ordered processing below:
According to the calculation method, the statistics of the network transfer speeds of N number of node to be sorted under K scene is calculated separately
Value, N >=2, K >=2；
Construct N × N-dimensional state transition probability matrix M, wherein N row and the N column of matrix M respectively correspond N number of node, square
The value of the element of the i-th row jth column is equal in the K scene in battle array M, the statistics of the network transfer speeds of the i-th row corresponding node
Value is inferior to the number of the scene of the statistical value of the network transfer speeds of jth column corresponding node, 1≤i, j≤N；
The initial score vector of preset 1 × N-dimensional and matrix M ' are iterated operation, obtain the node scoring vector of stable state,
Matrix M ' is the matrix obtained to matrix M by row normalization；
According to the size of N number of node value of corresponding element in node scoring vector, to the network of N number of node
Transmission service quality sequence.
7. a kind of analytical equipment of network transmission quality, which is characterized in that including computing module, the computing module includes:
Acquiring unit is recorded, for the constraint condition according to the scene of setting, obtains the node that transport services are provided in institute
The transport services record under scene is stated, the scene refers to the transport services environment that the constraint condition limits；
Quality determination unit, for determining that the network of the node under this scenario passes according to transport services record
Defeated quality of service information, the transport services information include the statistical value of network transfer speeds.
8. device as claimed in claim 7, it is characterised in that:
The record acquiring unit obtains the node for providing transport services described according to the constraint condition of the scene of setting
Transport services record under scene, wherein the constraint condition of the scene includes taking to one or more of parameter
The constraint of value:
Period；
Use the region where the terminal of the transport services；
The Internet Service Provider that the terminal is contracted；
The domain name that the terminal is accessed；
The type of the terminal.
9. device as claimed in claim 7 or 8, it is characterised in that:
Described device further includes the first sorting module, and first sorting module includes:
Receiving unit, for obtaining the network transfer speeds of multiple nodes to be sorted under the first scene from the computing module
Statistical value；
Sequencing unit passes network of the multiple node under first scene for the size according to corresponding statistical value
Defeated quality of service ordered.
10. device as claimed in claim 7 or 8, it is characterised in that:
Described device further includes the second sorting module, and second sorting module includes:
Receiving unit, for obtaining the network transfer speeds of N number of node to be sorted under K scene from the computing module
Statistical value, N >=2, K >=2；
Matrix construction unit, for constructing N × N-dimensional state transition probability matrix M, wherein N row and the N column of matrix M are right respectively
N number of node is answered, the value for the element that the i-th row jth arranges in matrix M is equal in the K scene, the net of the i-th row corresponding node
The statistical value of network transmission speed is inferior to the number of the scene of the statistical value of the network transfer speeds of jth column corresponding node, 1≤i, j
≤N；
Vector calculation unit obtains steady for the initial score vector of preset 1 × N-dimensional and matrix M ' to be iterated operation
The node scoring vector of state, matrix M ' are the matrixes obtained to matrix M by row normalization；
Sequencing unit, for the size according to N number of node value of corresponding element in node scoring vector, to described
The network transmission service quality of N number of node sorts.
11. a kind of analytical equipment of network transmission quality, including memory, processor and it is stored on the memory and can be
The computer program run on the processor, which is characterized in that the processor is realized such as when executing the computer program
The processing of method of any of claims 1-6.
12. a kind of computer readable storage medium, is stored thereon with computer program, which is characterized in that the computer program
The processing such as method of any of claims 1-6 is realized when being executed by processor.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201711146847.0A CN109802847A (en) | 2017-11-17 | 2017-11-17 | A kind of analysis method of network transmission service quality, device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201711146847.0A CN109802847A (en) | 2017-11-17 | 2017-11-17 | A kind of analysis method of network transmission service quality, device |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN109802847A true CN109802847A (en) | 2019-05-24 |
Family
ID=66555984
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201711146847.0A Pending CN109802847A (en) | 2017-11-17 | 2017-11-17 | A kind of analysis method of network transmission service quality, device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109802847A (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110636109A (en) * | 2019-08-19 | 2019-12-31 | 咪咕视讯科技有限公司 | Node scheduling optimization method, server and computer readable storage medium |
| CN114461305A (en) * | 2022-04-11 | 2022-05-10 | 天聚地合(苏州)科技股份有限公司 | Data source determination method and device |
| CN115379243A (en) * | 2021-05-21 | 2022-11-22 | 北京字跳网络技术有限公司 | CDN scheduling method, device, equipment and storage medium |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2006083694A2 (en) * | 2005-01-28 | 2006-08-10 | United Parcel Service Of America, Inc. | Registration and maintenance of address data for each service point in a territory |
| CN101459614A (en) * | 2008-12-24 | 2009-06-17 | 腾讯科技（深圳）有限公司 | Method, system and equipment for network data transmission |
| CN105897838A (en) * | 2015-12-07 | 2016-08-24 | 乐视云计算有限公司 | Network service node selection method and device |
-
2017
- 2017-11-17 CN CN201711146847.0A patent/CN109802847A/en active Pending
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2006083694A2 (en) * | 2005-01-28 | 2006-08-10 | United Parcel Service Of America, Inc. | Registration and maintenance of address data for each service point in a territory |
| CN101459614A (en) * | 2008-12-24 | 2009-06-17 | 腾讯科技（深圳）有限公司 | Method, system and equipment for network data transmission |
| CN105897838A (en) * | 2015-12-07 | 2016-08-24 | 乐视云计算有限公司 | Network service node selection method and device |
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110636109A (en) * | 2019-08-19 | 2019-12-31 | 咪咕视讯科技有限公司 | Node scheduling optimization method, server and computer readable storage medium |
| CN110636109B (en) * | 2019-08-19 | 2022-05-13 | 咪咕视讯科技有限公司 | Node scheduling optimization method, server and computer readable storage medium |
| CN115379243A (en) * | 2021-05-21 | 2022-11-22 | 北京字跳网络技术有限公司 | CDN scheduling method, device, equipment and storage medium |
| WO2022242498A1 (en) * | 2021-05-21 | 2022-11-24 | 北京字跳网络技术有限公司 | Cdn scheduling method and apparatus, device and storage medium |
| CN115379243B (en) * | 2021-05-21 | 2024-02-27 | 北京字跳网络技术有限公司 | CDN scheduling method, device, equipment and storage medium |
| CN114461305A (en) * | 2022-04-11 | 2022-05-10 | 天聚地合(苏州)科技股份有限公司 | Data source determination method and device |
| CN114461305B (en) * | 2022-04-11 | 2022-08-26 | 天聚地合(苏州)科技股份有限公司 | Data source determination method and device |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106101202B (en) | It analyzes for social graph data to determine internuncial system and method in community | |
| US20150242497A1 (en) | User interest recommending method and apparatus | |
| US9817885B2 (en) | Method and apparatus for grouping network service users | |
| CN109784636A (en) | Fraudulent user recognition methods, device, computer equipment and storage medium | |
| US11250250B2 (en) | Pedestrian retrieval method and apparatus | |
| CN108052639A (en) | Industry user based on carrier data recommends method and device | |
| US20100185619A1 (en) | Sampling analysis of search queries | |
| JP2017512344A (en) | System and method for rapid data analysis | |
| CN104182392B (en) | The method and apparatus of service recommendation processing | |
| CN103150696A (en) | Method and device for selecting potential customer of target value-added service | |
| CN107896153B (en) | Traffic package recommendation method and device based on mobile user internet surfing behavior | |
| CN109802847A (en) | A kind of analysis method of network transmission service quality, device | |
| US20160299903A1 (en) | Path analysis method and apparatus | |
| CN110298687B (en) | Regional attraction assessment method and device | |
| CN105631749A (en) | User portrait calculation method based on statistical data | |
| CN104750760A (en) | Application software recommending method and device | |
| CN103248677A (en) | Internet behavior analysis system and working method thereof | |
| CN113157922B (en) | Network entity behavior evaluation and visualization method based on graph | |
| CN115858168A (en) | Earth application model arrangement system and method based on importance ranking | |
| US20140214826A1 (en) | Ranking method and system | |
| CN110909072B (en) | Data table establishment method, device and equipment | |
| CN105824946A (en) | Method and system for multimedia recommendation on basis of data grading | |
| CN112241820A (en) | Risk identification method and device for key nodes in fund flow and computing equipment | |
| JP6164596B2 (en) | Keyword targeting system, keyword targeting device, server device, keyword targeting method, and program | |
| CN107784363B (en) | Data processing method, device and system |
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 | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20190524 |