CN106301905B - A kind of rational method and device of assessment CDN deployment - Google Patents
A kind of rational method and device of assessment CDN deployment Download PDFInfo
- Publication number
- CN106301905B CN106301905B CN201610653743.8A CN201610653743A CN106301905B CN 106301905 B CN106301905 B CN 106301905B CN 201610653743 A CN201610653743 A CN 201610653743A CN 106301905 B CN106301905 B CN 106301905B
- Authority
- CN
- China
- Prior art keywords
- cdn
- fringe node
- network
- hit rate
- cdn fringe
- 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.)
- Active
Links
- 238000000034 method Methods 0.000 title claims abstract description 29
- 238000003860 storage Methods 0.000 claims abstract description 114
- 239000002699 waste material Substances 0.000 abstract description 8
- 201000004569 Blindness Diseases 0.000 abstract description 2
- 238000004891 communication Methods 0.000 abstract description 2
- 238000005516 engineering process Methods 0.000 abstract description 2
- 238000010586 diagram Methods 0.000 description 6
- 238000004220 aggregation Methods 0.000 description 4
- 230000002776 aggregation Effects 0.000 description 4
- 238000009826 distribution Methods 0.000 description 4
- 210000002230 centromere Anatomy 0.000 description 2
- 238000009795 derivation Methods 0.000 description 1
- 230000006870 function Effects 0.000 description 1
- 230000014759 maintenance of location Effects 0.000 description 1
- 238000005303 weighing Methods 0.000 description 1
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
- H04L41/12—Discovery or management of network topologies
-
- 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
- H04L41/145—Network analysis or design involving simulating, designing, planning or modelling of a network
Landscapes
- Engineering & Computer Science (AREA)
- Computer Networks & Wireless Communication (AREA)
- Signal Processing (AREA)
- Data Exchanges In Wide-Area Networks (AREA)
Abstract
The embodiment of the present invention provides a kind of assessment CDN deployment rational method and device, is related to field of communication technology, is assessed with before CDN is disposed, treating the reasonability of CDN of deployment, disposes the problem of investment resources caused by CDN wastes to avoid existing blindness.This method may include: the relevant parameter for obtaining the network for being deployed with CDN, according to the relational model between the relevant parameter of network and CDN fringe node hit rate and CDN fringe node stored ratio, draw the relationship rectilinear between the storage of CDN fringe node and CDN fringe node hit rate, relationship rectilinear of the CDN fringe node the drawn out storage between CDN fringe node hit rate is compared with optimum proportioning point of the preset CDN fringe node hit rate between CDN fringe node stored ratio, the reasonability of CDN deployment is assessed according to comparison result.
Description
Technical field
The present embodiments relate to field of communication technology more particularly to a kind of assessment content distributing network (Content
Delivery Network, CDN) the rational method and device of deployment.
Background technique
With flourishing for internet, to guarantee that user can be with quick obtaining to Web content, telecom operators are existing
CDN is disposed in some the Internet infrastructures, the internet content distribution service refined end to end is provided for all types of user.
Currently, the mode for generalling use " storage tape swapping is wide " when telecom operators dispose deployment CDN in Metropolitan Area Network (MAN) obtains
Optimal user content distributes service experience, and when the service traffics of user are smaller, usually only disposing a CDN node can expire
Sufficient users service needs；When service traffics increase, can be guaranteed QoS of customer by two kinds of means: first way,
Continue the capacity of the dilatation CDN node, while dilatation CDN is to the network bandwidth end to end of user；The second way, by CDN
Node sinks, i.e., disposes Distributed C DN node in network edge, thus can save backbone bandwidth.
However, telecom operators are in CDN actual deployment, carrying out that deployment can be achieved using which kind of above mode actually can
It guarantees QoS of customer, to obtain optimal investment cost, there is presently no reasonable programmes, are carrying out CDN deployment
When, the problem of being easy to cause investment waste.
Summary of the invention
The embodiment of the present invention provides a kind of assessment CDN and disposes rational method and device, right with before CDN is disposed
The reasonability of CDN to be disposed is assessed, and disposes the problem of investment resources caused by CDN wastes to avoid existing blindness.
In order to achieve the above objectives, the embodiment of the present invention adopts the following technical scheme that
In a first aspect, providing a kind of assessment CDN deployment rational method, may include:
The relevant parameter for being deployed with the network of CDN is obtained, the network includes: N number of CDN fringe node, and the N is not small
Include in the relevant parameter of 1 integer, the network: equivalent number of concurrent, the film source of user's access of the film source of user's access
Store duration, unit storage cost and unit bandwidth cost；
According between the relevant parameter of the network and CDN fringe node hit rate and CDN fringe node stored ratio
Relational model, draw CDN fringe node storage CDN fringe node hit rate between relationship rectilinear, the edge CDN
Relational model between node hit rate and CDN fringe node stored ratio: it is thrown in the bandwidth for the network for being deployed with CDN and storage
Money and do not dispose between the bandwidth of network of CDN and the investment of storage reach balance in the case where derive；
By the CDN fringe node drawn out storage CDN fringe node hit rate between relationship rectilinear with it is preset
CDN fringe node hit rate is assessed compared with the optimum proportioning point between CDN fringe node stored ratio according to comparison result
The reasonability of CDN deployment.
Second aspect provides a kind of assessment CDN deployment rational device, for executing method described in first aspect,
The apparatus may include:
Acquiring unit obtains the relevant parameter for being deployed with the network of CDN, and the network includes: N number of CDN fringe node, institute
Stating N is the integer not less than 1, and the relevant parameter of the network includes: the equivalent number of concurrent of the film source of user's access, user's access
Film source storage duration, unit storage cost and unit bandwidth cost；
Drawing unit, relevant parameter and CDN fringe node for network according to the information acquired by the acquisition unit are ordered
Relational model between middle rate and CDN fringe node stored ratio is drawn the storage of CDN fringe node and is hit with CDN fringe node
Relationship rectilinear between rate, the relational model between the CDN fringe node hit rate and CDN fringe node stored ratio:
It invests and does not dispose between the bandwidth of network of CDN and the investment of storage in the bandwidth for the network for being deployed with CDN and storage and reach
It is derived in the case where balance；
Assessment unit, the storage of CDN fringe node and CDN fringe node hit rate for drawing out the drawing unit
Between relationship rectilinear and the optimum proportioning point between preset CDN fringe node hit rate and CDN fringe node stored ratio
It compares, the reasonability of CDN deployment is assessed according to comparison result.
From the foregoing, it will be observed that the embodiment of the present invention provides a kind of assessment CDN deployment rational method and device, acquisition is deployed with
The relevant parameter of the network of CDN is stored according to the relevant parameter of network and CDN fringe node hit rate and CDN fringe node
Relational model between ratio draws the relationship rectilinear between the storage of CDN fringe node and CDN fringe node hit rate, will
Relationship rectilinear and preset CDN fringe node between the CDN fringe node storage drawn out and CDN fringe node hit rate
Hit rate assesses the reasonable of CDN deployment compared with the optimum proportioning point between CDN fringe node stored ratio, according to comparison result
Property.In this way, the appraisal procedure proposed by the application, so that whether operator can prejudge cost of investment when CDN is invested in deployment
It is reasonably, to avoid investment waste.
Detailed description of the invention
In order to illustrate the technical solution of the embodiments of the present invention more clearly, below will be in embodiment or description of the prior art
Required attached drawing is briefly described, it should be apparent that, the accompanying drawings in the following description is only some realities of the invention
Example is applied, it for those of ordinary skill in the art, without creative efforts, can also be according to these attached drawings
Obtain other attached drawings.
Fig. 1 is the network architecture schematic diagram for not disposing CDN；
Fig. 2 is the network architecture schematic diagram for disposing CDN；
Fig. 3 is a kind of storage provided in an embodiment of the present invention and bandwidth investment critical point schematic diagram；
Fig. 4 is the flow chart that a kind of assessment CDN provided in an embodiment of the present invention disposes rational method；
Fig. 5 is storage provided in an embodiment of the present invention and bandwidth investment critical point schematic diagram；
Fig. 6 is the structure chart that a kind of assessment CDN provided in an embodiment of the present invention disposes rational device.
Specific embodiment
Due under certain userbase and business scale, when the backbone of CDN fringe node cost to be disposed and saving
When network bandwidth cost reaches balance, the investment of CDN to be disposed can be made to reach zone of reasonableness, be based on this, the application is mainly former
Reason is: when deriving that the backbone bandwidth cost of CDN fringe node cost and saving wait dispose reaches balance, the edge CDN
Computation model between node storage investment and bandwidth investment, the reasonable of the investment of CDN to be disposed is analyzed according to the computation model
Property, to assess the reasonability of CDN deployment before disposing CDN, avoid the problem that investment waste.
With reference to the attached drawing in the embodiment of the present invention, technical solution in the embodiment of the present invention carries out clear, complete
Ground description, it is clear that described embodiments are only a part of the embodiments of the present invention, instead of all the embodiments.Based on this
Embodiment in invention, every other reality obtained by those of ordinary skill in the art without making creative efforts
Example is applied, shall fall within the protection scope of the present invention.
In the description of the present invention, it is to be understood that, the system of instructions such as term " first ", " second ", " another " or
Element is the system or element with certain function described based on embodiment, is merely for convenience of the description present invention and simplification is retouched
It states, rather than the system of indication or suggestion meaning or element must have this name, therefore be not considered as limiting the invention.
It is described in detail before this programme, to facilitate the understanding of the present invention the technical solution, to skill of the present invention
Art noun is explained in detail, it is to be understood that following technical terms be only invention technician for convenience into
Capable description does not represent or implies that signified system or element must have this name, therefore should not be understood as to of the invention
Limitation:
Film source: the big file of the user stored in CDN node video to be visited (such as film, TV play video source file)
It is referred to as film source.
Hot spot film source: the bigger video source file of user's amount of access is referred to as hot spot film source, referred to as: backing.
Hit is concurrently flowed: being referred to as to hit for the Media Stream that user provides service by CDN fringe node and concurrently be flowed.
Ability is concurrently flowed in hit: being the streaming media service ability that user provides by CDN fringe node.
Miss concurrently flows: being when the film source of user's access is cold spot film source (the i.e. smaller video source of user's amount of access
File) when, without storage the contents of the section in CDN fringe node, service can not be provided for user, and by CDN central node
Service is provided for user, CDN central node stores user's whole piece source contents to be accessed (comprising cold spot film source and hot spot piece
Source).Accordingly, with respect to hit, concurrently stream is to provide the Media Stream of service by CDN fringe node for user, and concurrently stream is miss
The Media Stream of service is provided by CDN central node for user.
Miss concurrently flows ability: being the streaming media service ability that user provides by CDN central node.
Total stream handling capacity: entire CDN can provide total concurrent flow of streaming media service, including CDN central node for user
It can be the concurrent flow that user provides with CDN fringe node.
Total stream handling capacity cost: entire CDN can provide the stream service equipment of total concurrent flow of streaming media service for user
Cost, the cost of the stream service equipment including CDN central node and all CDN fringe nodes.
CDN fringe node hit rate: CDN fringe node is that account for whole CDN be use to the streaming media service ability that user provides
The ratio for the streaming media service ability that family provides, referred to as: edge hit rate.
CDN fringe node stored ratio: CDN fringe node memory capacity accounts for whole CDN memory capacity (comprising the center CDN
The memory capacity of node and all CDN fringe nodes) ratio, referred to as: edge stored ratio.
High definition code rate: the code rate that high definition film source plays is provided in CDN, usually 8Mbps.
SD code rate: the code rate that SD film source plays is provided in CDN, usually 2Mbps.
Equivalent number of concurrent: since the code rate that user accesses the film source in CDN is different, there is SD, high definition film source, in order to
Facilitate and calculate CDN concurrent capability, high definition code rate concurrent capability can be converted into the code rate of multiple SD, the film source after converting into is made
To convert into number of concurrent.Such as the code rate of 1 high definition film source is 8Mbps, the code rate of 1 SD film source is 2Mbps, then can be by 1
The code rate of high definition film source is converted into the code rate of 5 SD film sources.
Average bit rate: since the code rate of the film source stored in CDN is different, it there are SD, high definition film source, count for convenience
CDN memory capacity is calculated, summation can be weighted according to the ratio of high definition film source and SD film source, by the film source after weighted sum
Code rate as average bit rate.For example, CDN stores the SD film source of 40% 2Mbps code rate, while storing 60%
The high definition film source of 8Mbps code rate, then the average bit rate of film source is 5.6Mbps in CDN.
Memory capacity: for the product of the storage duration of film source, the number of film source and average bit rate in CDN.Such as: it is all
The duration of film source is 2 hours, totally 10000 film sources, wherein SD film source 4000, and high definition film source 6000, film source in CDN
Average bit rate be 5.6Mbps, then the total memory capacity of the CDN central node be (2*4000+8*6000) * 2*3600/ (8*
1024*1024)=48.1TB.
It is adapted to inside Metropolitan Area Network (MAN) it should be noted that assessment CDN provided by the invention disposes rational method
The scene for affixing one's name to CDN, the scene for being readily applicable to local area network, disposing CDN in backbone network, the embodiment of the present invention is to this without limit
Fixed, for ease of description, the present invention only for disposing the scene of CDN in Metropolitan Area Network (MAN), disposes assessment CDN provided by the invention
Rational method is illustrated.
It referring to Fig.1, is the network architecture schematic diagram for not disposing CDN, as shown in Figure 1, being deployed in the network architecture:
CDN central node, bearer network core node, bearer network aggregation node, business control nodes and terminal.In the network structure
In, the content of all film sources of user's access is stored to CDN central node, and the whole for being accessed user by CDN central node
Content passes sequentially through bearer network core node, bearer network aggregation node, business control nodes and is sent to terminal, i.e., in the network
100% customer flow is occupied from bearer network core node to bearer network aggregation node, bearer network aggregation node to business control
Node processed, business control nodes to the Metropolitan Area Network (MAN) backbone bandwidth between terminal.At this point, when the flow of user's access is bigger
When, in order to make user obtain optimal user content distribution service experience, existing technical staff is disposed using CDN, by CDN node
Sink, CDN fringe node is disposed in network edge (place i.e. from terminal relatively), by from the closer side CDN of user
The hot spot film source that user accesses is forwarded to user by edge node, so that user is obtained optimal content distribution service experience with this, together
When, save Metropolitan Area Network (MAN) backbone bandwidth.
For example, Fig. 2 is the network architecture schematic diagram for disposing CDN, as shown in Fig. 2, the network architecture shown in original Fig. 1 is basic
The upper multiple CDN fringe nodes of deployment, CDN fringe node are connect with business control nodes.In the network architecture, CDN centromere
Point is stored with the content of all film sources of user's access, and by the content distribution of the hot spot film source of user's access to the edge CDN section
Point is stored by CDN fringe node, and the content of the hot spot film source of storage is sent to user by business control nodes, and CDN
The cold spot film source that least a portion of user accesses then is passed sequentially through bearer network core node by central node, business control nodes are sent
To user, in this way, bearer network core node to the Metropolitan Area Network (MAN) backbone network between business control nodes is made only to carry fraction
The bandwidth of customer flow (do not order and concurrently flow), is greatly saved backbone bandwidth.It should be noted that for the ease of retouching
It states, a CDN fringe node is only deployed in Fig. 2, but this can not be to the CDN fringe node in the application when deployment CDN
Number causes to limit.
Compare Fig. 1 and Fig. 2 it is found that the content of all film sources of user's access is by CDN centromere when not disposing CDN
Point offer service, CDN central node will provide full content storage and whole concurrent stream process abilities；After disposing CDN,
CDN central node provides full content memory capacity and miss concurrently flows ability, and multiple CDN fringe nodes provide hot spot film source
Ability is concurrently flowed in content storage and hit.So when Fig. 1 does not dispose CDN, the center CDN needs to provide from cost of investment
Storage user accesses the investment of the storage equipment of the content of all film sources and supports all concurrently to spread defeated bandwidth investment,
Bearer network core node to the Metropolitan Area Network (MAN) backbone network between business control nodes need to provide support and all concurrently spread defeated bandwidth
Investment, the network of business control nodes to terminal room, which need to provide, supports all concurrently to spread defeated bandwidth investment；When being disposed in Fig. 2
When CDN, Metropolitan Area Network (MAN) backbone network, which only needs to provide, supports that miss concurrently spreads defeated bandwidth investment, and business control nodes are to end
Network between end, which still needs to provide, supports all concurrently to spread defeated bandwidth investment, and CDN central node needs to provide storage user and visits
It asks the investment of the storage equipment of the content of all film sources and miss is supported concurrently to spread defeated bandwidth investment, the edge CDN section
Point, which does not only need provide, supports hit concurrently to spread defeated bandwidth investment, and it is also required to provide the hot spot piece of storage user's access
The investment of the storage equipment of the content in source, i.e., the major investment of whole network when not disposing CDN are as follows: central node investment (C)=
The bandwidth cost that total storage cost+total stream handling capacity cost+whole concurrently flows；The major investment of whole network after deployment CDN are as follows:
CDN central node investment (C1)=always store the bandwidth cost that cost+miss stream handling capacity cost+miss concurrently flows, CDN
Fringe node invests (E1)=N × (fringe node stores cost+hit stream handling capacity cost), wherein N is CDN fringe node
Quantity.
At this point, the gross investment of the bandwidth of network and storage is less than not when needing to dispose CDN if wishing CDN deployment rationally
The gross investment of bandwidth and storage when deployment CDN, i.e., emphasis of the invention are to analyze the bandwidth of the network of CDN to be disposed and deposit
Bandwidth when storing up investment and not disposing CDN and the size relation between the investment of storage, the portion CDN is assessed according to the size relation
The reasonability of administration.
Firstly, when the investment of the bandwidth of the network wait dispose CDN and storage investment and bandwidth and storage when not disposing CDN
Between when reaching balance, i.e., both reach identical value, when (C)=(C1)+(E1), the critical point of deployment CDN can be obtained.
Investing (C) according to central node ,=always storing cost+always flows the bandwidth cost that handling capacity cost+whole concurrently flows；
CDN central node investment (C1)=always store the bandwidth cost that cost+miss stream handling capacity cost+miss concurrently flows, CDN
Fringe node invests (E1)=N × (fringe node stores cost+hit stream handling capacity cost) it is found that as (C)=(C1)+(E1)
When, following formula (1) can be obtained by abbreviation:
Hit bandwidth cost (HB)=N × CDN fringe node storage cost (ES) formula (1) concurrently flowed
The service ability of i.e. N number of CDN fringe node can save the bandwidth that hit is concurrently flowed when not disposing CDN after being formed.
Wherein, bandwidth cost (HB)=total bandwidth cost (the TB) × CDN fringe node hit rate concurrently flowed is hit
(HR) (formula 2)
CDN fringe node storage cost (ES)=always store cost (TS) × CDN fringe node stored ratio
(ER) (formula 3)
Total storage cost (TS)=unit storage cost (US) × average bit rate (AM) × storage duration (TT) (formula 4)
Total bandwidth cost (TB)=unit bandwidth cost (UB) × equivalent number of concurrent (TC) × average bit rate (AM) (formula 5)
Formula (2), formula (3) are substituted into formula (1) and can obtained:
CDN fringe node hit rate (HR)=(N × always store cost (TS) × CDN fringe node stored ratio (ER))/
Total bandwidth cost (TB) formula (6)
Formula (4), formula (5), which are substituted into formula (6), can be obtained following formula (7):
CDN fringe node hit rate (HR)=(unit stores cost (US)/unit bandwidth cost (UB)) × side N × CDN
Edge node stored ratio (ER) × storage duration (TT)/equivalent number of concurrent (TC) (formula 7)
It can be by the throwing of the bandwidth of the network wait dispose CDN and storage investment and bandwidth and storage when not disposing CDN
Relation formula (C)=(C1)+(E1) when reaching balance between money is converted into formula (7), by formula (7) it is found that according to deployment
Number of concurrent (TC), storage duration (TT), unit bandwidth and storage cost ratio (US/UB) and fringe node quantity are converted into after CDN
(N) difference, it can be deduced that the relationship between CDN fringe node hit rate (HR) and CDN fringe node stored ratio (ER) is bent
Line chart disposes the equalization point of CDN.
For example, telecom operators are in actual deployment, determine convert into after deployment CDN number of concurrent (TC), storage duration (TT),
After unit bandwidth and storage these parameters such as cost ratio (US/UB) and fringe node quantity (N), it can be obtained according to formula (7)
Relationship rectilinear between CDN fringe node hit rate and CDN fringe node stored ratio, the straight line are also when not disposing CDN
It hits after the bandwidth cost that concurrently flows and deployment CDN between the storage cost of CDN fringe node in graph of relation when balancing.
Since the storage cost of CDN fringe node should be less than hitting the bandwidth cost concurrently flowed on Metropolitan Area Network (MAN) backbone network,
It is just reasonable CDN deployment when hitting bandwidth cost (HB) > N × CDN fringe node storage cost (ES) concurrently flowed,
So according to the above-mentioned derivation process from formula (1) to formula (7), it can be deduced that: when CDN fringe node hit rate (HR) >
(unit stores cost (US)/unit bandwidth cost (UB)) × N × CDN fringe node stored ratio (ER) × storage duration
(TT)/equivalent number of concurrent (TC) when, be just reasonable CDN deployment, CDN should be disposed；As CDN fringe node hit rate (HR) <
(unit stores cost (US)/unit bandwidth cost (UB)) × N × CDN fringe node stored ratio (ER) × storage duration
(TT)/equivalent number of concurrent (TC) when, DCN should not be disposed.
I.e. as shown in figure 3, if indicating that CDN fringe node stored ratio (ER), ordinate of orthogonal axes indicate the side CDN with horizontal axis coordinate
The relationship rectilinear that edge node hit rate (HR) is drawn between CDN fringe node hit rate and CDN fringe node stored ratio (is deposited
Storage invests critical line with bandwidth), then the region on the relationship rectilinear indicates the Metropolitan Area Network (MAN) backbone network band for not disposing CDN
Width investment is greater than investment when deployment CDN, is the region that dispose CDN；Region under relationship rectilinear indicates non-portion
The investment of Metropolitan Area Network (MAN) backbone bandwidth is less than investment when disposing CDN when affixing one's name to CDN, is the region that dispose CDN, in relationship
Point on rectilinear indicates investment when Metropolitan Area Network (MAN) backbone bandwidth investment when not disposing CDN is equal to deployment CDN, in facing
Boundary's point.
Since in practical applications, according to Zipf's law, in the backing of CDN fringe node storage 20%, maximum can
Reach 80% hit rate, it is impossible to the hit rate more than 80%, this be theoretically optimal CDN fringe node hit rate and
The proportion relation (as shown in black dot in Fig. 3) of CDN fringe node stored ratio, i.e. storage are invested on critical line with bandwidth
CDN fringe node stored ratio is 20%, if the corresponding CDN fringe node hit rate of the value is less than or equal to 80%, then it represents that CDN
Rationally, an even CDN carries out configuration deployment not in accordance with 80/20, and deployment is inevitable unreasonable, and resource is unable to get for deployment
It adequately and reasonably utilizes, necessarily causes the waste of resource, in other words, if CDN fringe node hit rate and CDN fringe node
The optimum proportioning point (80/20 point) of stored ratio is in CDN fringe node hit rate and CDN fringe node stored ratio relationship straight line
When on (storage investing critical line with bandwidth), then it represents that CDN deployment is reasonable, otherwise, it means that CDN deployment investment is greater than non-portion
Metropolitan Area Network (MAN) backbone bandwidth is invested when affixing one's name to CDN, should not dispose CDN.
From the foregoing, it will be observed that in embodiments of the present invention, it can be first based on the bandwidth of network and the gross investment of storage when disposing CDN
With the balance of the gross investment of the bandwidth and storage of network when not disposing CDN, CDN fringe node hit rate and CDN are derived
Network parameter when disposing CDN is substituted into the relational model, obtains deployment by the relational model between fringe node stored ratio
Relationship when CDN between CDN fringe node hit rate and CDN fringe node stored ratio, according to the relationship and CDN fringe node
Comparison between hit rate and the optimum proportioning point of CDN fringe node stored ratio, the reasonability of assessment CDN deployment.
Clear following for describing, embodiment one shows assessment provided by the embodiment of the present invention in the form of step
CDN disposes rational method, it should be noted that this method can be in addition to assessment CDN disposes rational device
It is executed on any computer.Although, in some cases, can be to be different from addition, be shown in figure logical order
Sequence herein executes shown or described step.
Embodiment one
Fig. 4 is the flow chart that a kind of assessment CDN provided in an embodiment of the present invention disposes rational method, by assessing CDN
It disposes rational device to execute, it can be any service being deployed in Metropolitan Area Network (MAN) that assessment CDN, which disposes rational device,
Device, as shown in figure 4, the method may include following steps:
S101: the relevant parameter for being deployed with the network of CDN is obtained, which includes: N number of CDN fringe node, the network
Relevant parameter includes: equivalent number of concurrent, storage duration, the unit of the film source of user's access of the film source of user's access store cost
And unit bandwidth cost.
Wherein, N is the integer not less than 1.
Optionally, the relevant parameter of network can be stored in advance in assessment CDN and dispose in rational device.Wherein,
The equivalent number of concurrent of the film source of user's access can be with are as follows: the high definition film source of user's access is converted into number and use after SD film source
The summation of the number of the SD film source of family access.
The storage duration of the film source of user's access can be with are as follows: storage equipment of the film source of user's access in CDN
Unit stores cost can be with are as follows: the price of the corresponding storage equipment of the memory capacity of each TB, unit bandwidth cost
It can be with are as follows: the price of the corresponding equipment of the bandwidth of each Gbps；Optionally, unit stores cost, the value of unit bandwidth cost
Median can be taken according to the comprehensive cost in actual operation；Under normal conditions, classification, the disk battle array of unit storage cost and disk
Caching it is related, unit bandwidth cost can be related according to IP metropolitan area network equipment.Such as: unit stores cost can be with value
Are as follows: 1.2 ten thousand yuan/TB；Unit bandwidth cost can be with value are as follows: 50,000 yuan/Gbps.
S102: according to the relevant parameter for the network for being deployed with CDN and CDN fringe node hit rate and CDN fringe node
Relational model between stored ratio draws the relationship straight line between the storage of CDN fringe node and CDN fringe node hit rate
Figure.
Wherein, the relational model between CDN fringe node hit rate and CDN fringe node stored ratio can be deployed with
The feelings for reaching balance between the bandwidth of network of CDN and the investment of storage are invested and are not disposed in the bandwidth of the network of CDN and storage
It is derived under condition.Preferably, which can be as shown in above-mentioned formula (7).
At this point it is possible to first substitute into the relevant parameter of network in above-mentioned formula (7), determine to be deployed in the network of CDN
Then relationship between CDN fringe node hit rate and CDN fringe node stored ratio calculates different CDN fringe nodes and deposits
CDN fringe node hit rate under storage ratio draws the storage of CDN fringe node according to calculated result in plane right-angle coordinate
With the relationship rectilinear between CDN fringe node hit rate.
S103: by the relationship rectilinear between the CDN fringe node drawn out storage and CDN fringe node hit rate and in advance
If CDN fringe node hit rate compared with optimum proportioning point between CDN fringe node stored ratio, commented according to comparison result
Estimate the reasonability of CDN deployment.
Wherein, optimum proportioning point can be above-mentioned 80/20 point, i.e. the backing of CDN fringe node storage 20%, and maximum can
With 80% fringe node hit rate.
Optionally, if relationship rectilinear between CDN fringe node hit rate and CDN fringe node stored ratio this most
Good proportion point is following, it is determined that the network C DN deployment for being deployed with CDN is reasonable；If CDN fringe node hit rate and the side CDN
Relationship rectilinear between edge node stored ratio is on the optimum proportioning point, it is determined that is deployed with the portion CDN in the network of CDN
It affixes one's name to unreasonable.
It will be appreciated that in embodiments of the present invention, CDN fringe node hit rate and CDN fringe node stored ratio it
Between relationship rectilinear can refer to below the optimum proportioning point: vertical line is done using the optimum proportioning point as starting point, if this is vertical
Relationship rectilinear between the part that line extends downwardly and CDN fringe node hit rate and CDN fringe node stored ratio intersects,
Then determine the relationship rectilinear between CDN fringe node hit rate and CDN fringe node stored ratio under the optimum proportioning point
Face；If anyway between the part that the vertical line upwardly extends and CDN fringe node hit rate and CDN fringe node stored ratio
Relationship rectilinear intersection, it is determined that the relationship rectilinear between CDN fringe node hit rate and CDN fringe node stored ratio exists
Above the optimum proportioning point.
Further alternative, the method can also include:
If drawing out the first relationship rectilinear in the optimum proportioning point according to the first equivalent number of concurrent, the first storage duration
Under, the second relationship rectilinear is drawn out under the optimum proportioning point according to the first equivalent number of concurrent, the first storage duration,
And first relationship rectilinear between the optimum proportioning point at a distance from less than between the second relationship rectilinear and optimum proportioning point
Distance, then using the first equivalent number of concurrent, first storage duration be deployed with as described in CDN network best configuration.
To make the object, technical solutions and advantages of the present invention clearer, below for different high definition code rates (8M) and mark
The quantity of clear code rate (2M) film source, storage needed for calculating 50000 hours, 70000 hours, 100000 hours and 150000 hours
Capacity (as shown in table 1).
Table 1
Wherein, film source amount of storage is 50000 hours, when the equivalent number of concurrent of user is 5000, show that edge storage is different
Ratio is as shown in table 2 below in storage and the relationship of CDN fringe node hit rate corresponding to bandwidth investment critical point:
Table 2
It is 50000 hours that film source amount of storage as shown in Figure 5, which can be obtained, according to upper table, when user concurrent number is 5000
In the case of, storage invests critical line chart with bandwidth, wherein the Fig. 5 using CDN fringe node stored ratio as abscissa line, with
CDN fringe node hit rate is as the coordinate longitudinal axis.Similarly, can also obtain film source amount of storage be 70000 hours, 100000 hours and
150000 small storages at present and bandwidth invest critical line chart (as shown in Figure 5).
As seen from Figure 5, when equivalent number of concurrent is 5000 users, storage is interior greater than 100,000 hours (such as 150,000 hours)
The optimum proportioning point (80/20 point, in Fig. 5 shown in black dot) of Rong Shi, edge order rate and edge stored ratio storage with
Under bandwidth investment matching critical line, CDN deployment is unreasonable；When equivalent number of concurrent is 5000 users, storage is less than 10
When the content of ten thousand hours (such as 50,000 hours, 70,000 hours, 100,000 hours), the optimum proportioning of edge order rate and edge stored ratio
Point (80/20 point, in Fig. 5 shown in five-pointed star) matches on critical line in storage with bandwidth investment, and CDN deployment is reasonable.
Further, as shown in figure 5, equivalent number of concurrent is 5000, when storing 100,000 hours contents, storage and bandwidth
Therefore the optimum proportioning point (80/20 point) of investment critical line most approaching edge order rate and edge stored ratio can will convert into
Number of concurrent is 5000, best configuration of the content of storage 100,000 hours as the network for disposing CDN.
From the foregoing, it will be observed that the embodiment of the present invention provides a kind of assessment CDN deployment rational method, acquisition is deployed with CDN's
The relevant parameter of network, according to the relevant parameter of network and CDN fringe node hit rate and CDN fringe node stored ratio
Between relational model, draw CDN fringe node storage CDN fringe node hit rate between relationship rectilinear, will draw
The relationship rectilinear between the storage of CDN fringe node and CDN fringe node hit rate out is hit with preset CDN fringe node
Rate assesses the reasonability of CDN deployment according to comparison result compared with the optimum proportioning point between CDN fringe node stored ratio.
In this way, by the application propose appraisal procedure so that operator deployment invest CDN when, can prejudge cost of investment whether be
Reasonably, investment waste is avoided.
Embodiment two
Fig. 6 shows the structure chart that a kind of assessment CDN provided in an embodiment of the present invention disposes rational device 20, is used for
The method for executing embodiment one, as shown in fig. 6, the device 20 may include:
Acquiring unit 201 obtains the relevant parameter for being deployed with the network of CDN, and network includes: N number of CDN fringe node, N are
The relevant parameter of integer not less than 1, network includes: equivalent number of concurrent, the film source of user's access of the film source of user's access
Store duration, unit storage cost and unit bandwidth cost；
Drawing unit 202, the relevant parameter and CDN fringe node of the network for being got according to acquiring unit 201
Relational model between hit rate and CDN fringe node stored ratio is drawn the storage of CDN fringe node and is ordered with CDN fringe node
Relationship rectilinear between middle rate, the relational model between CDN fringe node hit rate and CDN fringe node stored ratio:
The bandwidth and storage for being deployed with the network of CDN, which are invested and do not disposed between the bandwidth of network of CDN and the investment of storage, reaches flat
It is derived in the case where weighing apparatus；
Assessment unit 203, the CDN fringe node storage for drawing out drawing unit 202 are hit with CDN fringe node
The optimum proportioning between relationship rectilinear and preset CDN fringe node hit rate and CDN fringe node stored ratio between rate
Point compares, and the reasonability of CDN deployment is assessed according to comparison result.
Wherein, the optimum proportioning point between preset CDN fringe node hit rate and CDN fringe node stored ratio are as follows: CDN
The coordinate points that CDN fringe node hit rate is 80% when fringe node stored ratio is 20%.
Optionally, assessment unit 203 specifically can be used for:
If the relationship rectilinear between the CDN fringe node hit rate drawn out and CDN fringe node stored ratio is most
Under good proportion point, it is determined that it is reasonable to be deployed with CDN deployment in the network of CDN；If CDN fringe node hit rate and the edge CDN
Relationship rectilinear between node stored ratio is on optimum proportioning point, it is determined that is deployed with CDN in the network of CDN and disposes not
Rationally.
Further alternative, assessment unit 203 can be also used for:
If according to the first equivalent number of concurrent, first the first relationship rectilinear for drawing out of storage duration optimum proportioning point it
Under, the second relationship rectilinear is drawn out under optimum proportioning point according to the second equivalent number of concurrent, the second storage duration, and first
Relationship rectilinear between optimum proportioning point at a distance from less than the second relationship rectilinear between optimum proportioning point at a distance from, then by first
The best configuration of equivalent number of concurrent, the first storage duration as the network for being deployed with CDN.
From the foregoing, it will be observed that the embodiment of the present invention provides a kind of assessment CDN deployment rational device, acquisition is deployed with CDN's
The relevant parameter of network, according to the relevant parameter of network and CDN fringe node hit rate and CDN fringe node stored ratio
Between relational model, draw CDN fringe node storage CDN fringe node hit rate between relationship rectilinear, will draw
The relationship rectilinear between the storage of CDN fringe node and CDN fringe node hit rate out is hit with preset CDN fringe node
Rate assesses the reasonability of CDN deployment according to comparison result compared with the optimum proportioning point between CDN fringe node stored ratio.
In this way, by the application propose appraisal procedure so that operator deployment invest CDN when, can prejudge cost of investment whether be
Reasonably, investment waste is avoided.
Finally, it should be noted that the above embodiments are merely illustrative of the technical solutions of the present invention, rather than its limitations；Although
Present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that: it still may be used
To modify the technical solutions described in the foregoing embodiments or equivalent replacement of some of the technical features；
And these are modified or replaceed, the range for technical solution of various embodiments of the present invention that it does not separate the essence of the corresponding technical solution.
Claims (6)
1. a kind of assessment content distributing network CDN disposes rational method characterized by comprising
The relevant parameter for being deployed with the network of CDN is obtained, the network includes: N number of CDN fringe node, and the N is not less than 1
Integer, the relevant parameter of the network includes: user access film source equivalent number of concurrent, user access film source storage
Duration, unit storage cost and unit bandwidth cost；The equivalent number of concurrent of the film source is the high definition film source folding of user's access
The summation of the number of the SD film source of number and user's access after being combined into SD film source；
According to the pass between the relevant parameter of the network and CDN fringe node hit rate and CDN fringe node stored ratio
It is model, draws the relationship rectilinear between CDN fringe node stored ratio and CDN fringe node hit rate, the edge CDN
Relational model between node hit rate and CDN fringe node stored ratio: it is thrown in the bandwidth for the network for being deployed with CDN and storage
Money and do not dispose between the bandwidth of network of CDN and the investment of storage reach balance in the case where derive；
By between the CDN fringe node stored ratio drawn out and CDN fringe node hit rate relationship rectilinear with it is preset
CDN fringe node hit rate is assessed compared with the optimum proportioning point between CDN fringe node stored ratio according to comparison result
The reasonability of CDN deployment；
Optimum proportioning point between the preset CDN fringe node hit rate and CDN fringe node stored ratio are as follows: the edge CDN
The coordinate points that CDN fringe node hit rate is 80% when node stored ratio is 20%.
2. the method according to claim 1, wherein it is described according to comparison result assess CDN deployment reasonability,
Include:
If relationship rectilinear between the CDN fringe node hit rate drawn out and CDN fringe node stored ratio it is described most
Under good proportion point, it is determined that described to be deployed in the network of CDN CDN deployment reasonable；
If the relationship rectilinear between the CDN fringe node hit rate and CDN fringe node stored ratio is most preferably matched described
On point, it is determined that it is unreasonable to be deployed with CDN deployment in the network of CDN.
3. -2 described in any item methods according to claim 1, which is characterized in that the method also includes:
If according to the first equivalent number of concurrent, first the first relationship rectilinear for drawing out of storage duration the optimum proportioning point it
Under, the second relationship rectilinear is drawn out under the optimum proportioning point according to the second equivalent number of concurrent, the second storage duration, and
The first relationship rectilinear between the optimum proportioning point at a distance from be less than the second relationship rectilinear most preferably match with described
Than the distance between point, then the described first equivalent number of concurrent, the first storage duration are deployed with to the network of CDN as described in
Best configuration.
4. a kind of assessment content distributing network CDN disposes rational device characterized by comprising
Acquiring unit obtains the relevant parameter for being deployed with the network of CDN, and the network includes: N number of CDN fringe node, the N
Relevant parameter for the integer not less than 1, the network includes: the equivalent number of concurrent of the film source of user's access, user access
Storage duration, unit storage cost and the unit bandwidth cost of film source；The equivalent number of concurrent of the film source is user's access
High definition film source is converted into the summation of the number and the number of the SD film source of user's access after SD film source；
Drawing unit, relevant parameter and CDN fringe node hit rate for network according to the information acquired by the acquisition unit
With the relational model between CDN fringe node stored ratio, draws CDN fringe node stored ratio and CDN fringe node is hit
Relationship rectilinear between rate, the relational model between the CDN fringe node hit rate and CDN fringe node stored ratio:
It invests and does not dispose between the bandwidth of network of CDN and the investment of storage in the bandwidth for the network for being deployed with CDN and storage and reach
It is derived in the case where balance；
Assessment unit, CDN fringe node stored ratio and CDN fringe node hit rate for drawing out the drawing unit
Between relationship rectilinear and the optimum proportioning point between preset CDN fringe node hit rate and CDN fringe node stored ratio
It compares, the reasonability of CDN deployment is assessed according to comparison result；
Optimum proportioning point between the preset CDN fringe node hit rate and CDN fringe node stored ratio are as follows: the edge CDN
The coordinate points that CDN fringe node hit rate is 80% when node stored ratio is 20%.
5. device according to claim 4, which is characterized in that the assessment unit is specifically used for:
If relationship rectilinear between the CDN fringe node hit rate drawn out and CDN fringe node stored ratio it is described most
Under good proportion point, it is determined that described to be deployed in the network of CDN CDN deployment reasonable；
If the relationship rectilinear between the CDN fringe node hit rate and CDN fringe node stored ratio is most preferably matched described
On point, it is determined that it is unreasonable to be deployed with CDN deployment in the network of CDN.
6. according to the described in any item devices of claim 4-5, which is characterized in that the assessment unit is also used to:
If according to the first equivalent number of concurrent, first the first relationship rectilinear for drawing out of storage duration the optimum proportioning point it
Under, the second relationship rectilinear is drawn out under the optimum proportioning point according to the second equivalent number of concurrent, the second storage duration, and
The first relationship rectilinear between the optimum proportioning point at a distance from be less than the second relationship rectilinear most preferably match with described
Than the distance between point, then the described first equivalent number of concurrent, the first storage duration are deployed with to the network of CDN as described in
Best configuration.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610653743.8A CN106301905B (en) | 2016-08-10 | 2016-08-10 | A kind of rational method and device of assessment CDN deployment |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610653743.8A CN106301905B (en) | 2016-08-10 | 2016-08-10 | A kind of rational method and device of assessment CDN deployment |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN106301905A CN106301905A (en) | 2017-01-04 |
| CN106301905B true CN106301905B (en) | 2019-05-14 |
Family
ID=57668222
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201610653743.8A Active CN106301905B (en) | 2016-08-10 | 2016-08-10 | A kind of rational method and device of assessment CDN deployment |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN106301905B (en) |
Families Citing this family (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109683816B (en) * | 2018-12-14 | 2021-08-27 | 北京奇艺世纪科技有限公司 | Disk configuration method and system for back source tree nodes |
| CN110933140B (en) * | 2019-11-05 | 2021-12-24 | 北京字节跳动网络技术有限公司 | CDN storage allocation method, system and electronic equipment |
| CN114826935B (en) * | 2022-03-16 | 2024-06-07 | 网宿科技股份有限公司 | Model generation method, system, server and storage medium |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102801792A (en) * | 2012-07-26 | 2012-11-28 | 华南理工大学 | Statistical-prediction-based automatic cloud CDN (Content Delivery Network) resource automatic deployment method |
| CN105229624A (en) * | 2013-03-14 | 2016-01-06 | 麻省理工学院 | For realizing the method and apparatus of distributed content cache in content distributing network |
| CN105450535A (en) * | 2015-11-10 | 2016-03-30 | 深圳市网心科技有限公司 | Massive CDN node resource deployment system and method |
| WO2016109297A1 (en) * | 2014-12-31 | 2016-07-07 | Vidscale, Inc. | Methods and systems for an end-to-end solution to deliver content in a network |
| CN105791150A (en) * | 2016-05-09 | 2016-07-20 | 中国联合网络通信集团有限公司 | CDN network deployment method and system |
-
2016
- 2016-08-10 CN CN201610653743.8A patent/CN106301905B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102801792A (en) * | 2012-07-26 | 2012-11-28 | 华南理工大学 | Statistical-prediction-based automatic cloud CDN (Content Delivery Network) resource automatic deployment method |
| CN105229624A (en) * | 2013-03-14 | 2016-01-06 | 麻省理工学院 | For realizing the method and apparatus of distributed content cache in content distributing network |
| WO2016109297A1 (en) * | 2014-12-31 | 2016-07-07 | Vidscale, Inc. | Methods and systems for an end-to-end solution to deliver content in a network |
| CN105450535A (en) * | 2015-11-10 | 2016-03-30 | 深圳市网心科技有限公司 | Massive CDN node resource deployment system and method |
| CN105791150A (en) * | 2016-05-09 | 2016-07-20 | 中国联合网络通信集团有限公司 | CDN network deployment method and system |
Also Published As
| Publication number | Publication date |
|---|---|
| CN106301905A (en) | 2017-01-04 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN103295147B (en) | method, device and system for advertising | |
| CN106301905B (en) | A kind of rational method and device of assessment CDN deployment | |
| CN108737405A (en) | A kind of method, CCL servers and system carrying out water conservancy diversion to live video stream | |
| CN109412977B (en) | Domain name bandwidth adjusting method and related equipment | |
| CN106992938A (en) | A kind of network traffics dynamic dispatching distribution method and system | |
| US20120072526A1 (en) | Method and node for distributing electronic content in a content distribution network | |
| CN100592743C (en) | Operation supporting platform system for supporting stream media business | |
| CN104348798B (en) | A kind of method, apparatus, dispatch server and system for distributing network | |
| CN101132373A (en) | Method of providing qos for a flow | |
| CN104796449B (en) | Content delivery method, device and equipment | |
| CN110308995A (en) | A kind of edge cloud computing service system edges cloud node deployment device | |
| CN103607424B (en) | Server connection method and server system | |
| CN110417676A (en) | A kind of dispatching method, device and dispatch network | |
| CN110071980A (en) | The distribution method and device of agent node | |
| CN106993014A (en) | The method of adjustment of cache contents, apparatus and system | |
| CN109151009A (en) | A kind of CDN node distribution method and system based on MEC | |
| CN106789956B (en) | A kind of P2P order method and system based on HLS | |
| CN102761572A (en) | Method and system for dynamically adjusting data flow | |
| Silvestre et al. | Aren: a popularity aware replication scheme for cloud storage | |
| CN101296512A (en) | Service quality control method, device and system | |
| CN109618003A (en) | A kind of servers' layout method, server and storage medium | |
| CN103416027B (en) | The system of the method, buffer and cache optimization of cache optimization | |
| CN105827454A (en) | Method and system for constructing topologic CDN (content delivery network) model | |
| KR20140004236A (en) | Content delivery architecture and method | |
| CN104871499A (en) | Communication node, control device, method for managing control information entries, and program |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |