CN109769020A - CDN cluster accelerates serve resources management method and device, electronic equipment and storage medium - Google Patents
CDN cluster accelerates serve resources management method and device, electronic equipment and storage medium Download PDFInfo
- Publication number
- CN109769020A CN109769020A CN201910019458.4A CN201910019458A CN109769020A CN 109769020 A CN109769020 A CN 109769020A CN 201910019458 A CN201910019458 A CN 201910019458A CN 109769020 A CN109769020 A CN 109769020A
- Authority
- CN
- China
- Prior art keywords
- client
- bandwidth
- cluster
- service
- accelerates
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Abstract
The present embodiments relate to content distributions to accelerate service technology field, discloses a kind of CDN cluster and accelerates serve resources management method and device, electronic equipment and storage medium.This method comprises: the practical acceleration Service Source behaviour in service of monitoring client；Determine whether the influence by client if being influenced by client controls the use of acceleration Service Source of client to inhibit to influence for the use of the common source in cluster according to the practical acceleration Service Source behaviour in service of client.Embodiment of the present invention can guarantee to be substantially unaffected with the Service Source of other clients of its common source when portions of client access exception and amount of access are uprushed.
Description
Technical field
The present invention relates to content distributions to accelerate service technology field, in particular to a kind of CDN cluster accelerates Service Source pipe
Manage method and device, electronic equipment and storage medium.
Background technique
Service provider is when planning customer resources by CDN (Content Delivery Network, i.e. content distributing network), greatly
Partial picture can all consider resource multiplex, and to achieve the purpose that save cost, i.e. a server is embodied as by configuration file
Multiple clients, which provide, accelerates service.If being ready for exclusive server to each acceleration client, the wasting of resources will cause, especially
It is the smaller client of bandwidth.
Inventor has found the relevant technologies the prior art has at least the following problems: the existing same server provides acceleration for multiple clients
When service, when abnormal access occur in certain clients or amount of access is uprushed, not only the client itself will receive influence, Er Qiehe
The client of its common source also will receive larger impact.
Summary of the invention
Embodiment of the present invention is designed to provide a kind of acceleration of CDN cluster serve resources management method and device, electricity
Sub- equipment and storage medium, can when portions of client access exception and amount of access are uprushed, guarantee with its common source its
The Service Source of his client is substantially unaffected.
In order to solve the above technical problems, embodiments of the present invention provide a kind of CDN cluster acceleration serve resources management
Method, comprising:
Monitor the practical acceleration Service Source behaviour in service of client；
The use of the common source in the cluster is determined according to the practical acceleration Service Source behaviour in service of the client
Whether influenced by the client, if being influenced by the client, controls making for the acceleration Service Source of the client
To inhibit the influence.
Embodiments of the present invention additionally provide a kind of CDN cluster acceleration serve resources management device, comprising:
Monitoring modular, for monitoring the practical acceleration Service Source behaviour in service of client；
Determining module, for determining being total in the cluster according to the practical acceleration Service Source behaviour in service of the client
Whether influenced by the client with the use of resource, if being influenced by the client, trigger control module,
The control module is used to control the use of the acceleration Service Source of the client to inhibit the influence.
Embodiments of the present invention additionally provide a kind of electronic equipment, comprising: memory and processor, memory storage meter
Calculation machine program, processor run the computer program to realize that CDN cluster as described above accelerates serve resources management method.
Embodiments of the present invention additionally provide a kind of storage medium, for storing computer-readable program, the calculating
Machine readable program, which is used to execute CDN cluster as described above for computer, accelerates serve resources management method.
Embodiment of the present invention in terms of existing technologies, passes through the practical acceleration Service Source according to the client of monitoring
Behaviour in service determine the common source in cluster use whether the influence by some client, and when being affected, control
The client, to inhibit the influence, thereby may be ensured that the service of each client of common service resource to the use of acceleration Service Source
Quality.
It is described that the cluster is determined according to the practical acceleration Service Source behaviour in service of the client as one embodiment
Whether interior common source is influenced by the client, specifically: it is used according to the practical acceleration Service Source of the client
Situation determines whether the resource occupation type of the client is abnormal access, if abnormal access, it is determined that in the cluster
Common source is influenced by the client；And/or institute is determined according to the practical acceleration Service Source behaviour in service of the client
Whether the practical acceleration bandwidth for stating client, which exceeds pre-set client, ensures bandwidth, if ensureing bandwidth beyond the client,
Determine influence of the common source by the client in the cluster.
As one embodiment, the client ensures that bandwidth is calculated according to following data: cluster maximum bandwidth and
Client priority level (PRI).
As one embodiment, the client ensures that bandwidth is calculated by using the following formula: Band_GN=(1+PN/P)*
Band_KN；Wherein, N is the number of client, Band_GNBandwidth, Band_K are ensured for the client of n-th clientNFor n-th client
Account planning bandwidth, and the sum of account planning bandwidth of all clients of cluster be less than or equal to predetermined fraction the cluster
Maximum bandwidth；P is the sum of the priority of all clients of cluster, PNFor the priority of n-th client, and n-th client's is preferential
Higher, the P of gradeNIt is bigger.
It is described that the client is determined according to the practical acceleration Service Source behaviour in service of the client as one embodiment
Resource occupation type whether be abnormal access, specifically include: following one or any combination thereof if it exists, it is determined that the visitor
The resource occupation type at family is abnormal access: client's number of request is greater than number of request threshold value in the unit time or the client asks
The ratio between number and all number of requests of cluster is asked to be greater than first threshold；It is default greater than first that the same domain name request number of client increases multiple
The ratio between total number of request of client of client's domain name request number and same domain name is greater than second threshold in multiple or predetermined period, and
Client's domain name request number is greater than the historical high number of request of the second presupposition multiple；Or the rate of people logging in of the same URL of client
Greater than default rate of people logging in.
As one embodiment, the use of acceleration Service Source of the client is controlled to inhibit the influence, is specifically wrapped
It includes: if the resource occupation type of the client is abnormal access, the resource of the client being forbidden to access to eliminate the influence；
If the practical acceleration bandwidth of the client ensures bandwidth beyond pre-set client, bandwidth increments transfer operation is executed.
As one embodiment, bandwidth increments transfer operation is executed, specifically include: what pre-set bandwidths increment exceeded allows to surpass
Duration threshold value out；When reaching described allows beyond duration threshold value, the bandwidth increments that will exceed are transferred to standby resources pond.
As one embodiment, the bandwidth increments that will exceed are transferred to standby resources pond, specifically include: from described standby
With filtering out target available resources in resource pool；The bandwidth increments that will exceed are transferred to the target available resources.
It is described to filter out target available resources from the standby resources pond as one embodiment, it specifically includes: according to
Predetermined order rule is ranked up the available resources in the standby resources pond；According to default screening rule from the institute after sequence
It states and filters out the target available resources in available resources.
As one embodiment, the predetermined order rule includes: the operation according to operator and area and the cluster
Quotient divides the priority of the available resources whether identical with area；Wherein, the identical available resources of operator and area is excellent
First grade is higher than operator and available resources from different places, the available resources in same priority according to available bandwidth big float
Sequence.
As one embodiment, the default screening rule includes: to filter out priority high and resource extent and described super
The comparable available resources of bandwidth increments out are as the target available resources.
As one embodiment, after the bandwidth increments that will exceed are transferred to standby resources pond, further includes: described in determining
Whether the practical acceleration bandwidth of client meets switchback condition, if meeting the switchback condition, cancels the standby resources pond
Detailed description of the invention
Fig. 1 is that the CDN cluster in first embodiment accelerates serve resources management method flow diagram according to the present invention；
Fig. 2 is that the CDN cluster in second embodiment accelerates serve resources management method flow diagram according to the present invention；
Fig. 3 is that the CDN cluster in third embodiment accelerates serve resources management method flow diagram according to the present invention；
Fig. 4 is that the CDN cluster in the 4th embodiment accelerates serve resources management apparatus structure block diagram according to the present invention；
Fig. 5 is the structural schematic diagram of the electronic equipment in the 5th embodiment according to the present invention.
Specific embodiment
In order to make the object, technical scheme and advantages of the embodiment of the invention clearer, below in conjunction with attached drawing to the present invention
Each embodiment be explained in detail.However, it will be understood by those skilled in the art that in each embodiment party of the present invention
In formula, many technical details are proposed in order to make reader more fully understand the present invention.But even if without these technical details
And various changes and modifications based on the following respective embodiments, claimed technical solution of the invention also may be implemented.
The first embodiment of the present invention is related to a kind of CDN clusters to accelerate serve resources management method, referring to Fig. 1, should
Method includes step 101 to step 103.
Step 101: monitoring the practical acceleration Service Source behaviour in service of client.
CDN accelerates service provider when planning customer resources, it will usually consider resource multiplex, that is, plan that several clients share
The acceleration Service Source of CDN cluster.In practical applications, it can plan what a CND cluster can service according to customer demand
Client's number, such as a CDN cluster can provide acceleration service simultaneously for 30 clients.In step 101, respectively to each client
Acceleration Service Source actual use situation be monitored, including but not limited to: monitoring the acceleration service of each client actual use
Bandwidth, the number of request to the page, the number of request to domain name, URL (Uniform Resource Locator, unified resource positioning
Symbol) number of request etc., by monitoring the practical acceleration Service Source behaviour in service of each client, some client can be assisted in whether
Influence use of other clients to common source.
Step 102: the use of the common source in cluster is determined according to the practical acceleration Service Source behaviour in service of client
Whether influenced by client, if being influenced by the client, then follow the steps 103, if being not affected by the influence of the client, is tied
Beam.
It for example, is 300MB (Mbytes) if the account planning bandwidth of certain account planning, but the client is practical to account for
Accelerating service bandwidth has been more than the 30% of account planning bandwidth, then can determine that the client influences whether other clients to shared
The use of resource.Alternatively, attack occurs in certain client, Service Source is largely requested, collection group delay is caused to respond other visitors
Family or when cannot respond to the request of other clients, can also determine that the client has influenced other clients and made to common source
With.
Step 103: controlling the use of acceleration Service Source of client to inhibit to influence.
For example, when monitoring that the practical acceleration service bandwidth occupied of the client has been more than the 30% of account planning bandwidth,
The bandwidth increments that part can be will exceed are transferred out.Alternatively, forbidding the client to visit when monitoring that attack occurs in client
Ask cluster.It is so without being limited thereto, as long as can guarantee that each client can normal use common source.
Present embodiment is determining client's actual use by the practical acceleration Service Source behaviour in service of monitoring client
Acceleration Service Source it is excessive, when being adversely affected to the client of other common service resources, to the client to accelerate service
The use of resource is controlled, to can guarantee use of other clients to common source, and then ensures that more clients share CDN and add
The service quality of each client when fast Service Source.
Second embodiment of the present invention is related to a kind of CDN cluster acceleration serve resources management method, as shown in Fig. 2, this
Embodiment includes step 201 to step 205.
Step 201: monitoring the practical acceleration Service Source behaviour in service of client.
CDN accelerates service provider when planning customer resources, it will usually consider resource multiplex, that is, plan that several clients share
The acceleration Service Source of CDN cluster.In practical applications, it can plan what a CND cluster can service according to customer demand
Client's number, such as a CDN cluster can provide acceleration service simultaneously for 30 clients.In step 201, respectively to each client
Acceleration Service Source actual use situation be monitored, including but not limited to: monitoring the acceleration service of each client actual use
Bandwidth, the number of request to the page, the number of request to domain name, URL (Uniform Resource Locator, unified resource positioning
Symbol) etc., by monitoring the practical acceleration Service Source behaviour in service of each client, it can assist in whether some client influences it
His use of the client to common source.
Step 202: according to the practical acceleration Service Source behaviour in service of client determine client resource occupation type whether
It thens follow the steps 203 if abnormal access for abnormal access and the resource of client is forbidden to access to eliminate the effects of the act, if not abnormal
Access, thens follow the steps 204.In practical applications, client can be notified to debug after the resource access for forbidding client,
It can restore the service of the client after Failure elimination.
Step 202 specifically includes: following one or any combination thereof if it exists, it is determined that the resource occupation type of client is
Abnormal access: client IP (Internet Protocol, the agreement interconnected between network) centrality, cluster access domain name are concentrated
Property or cluster access set of URL it is neutral.Any one of the above situation if it does not exist, it is determined that the resource occupation type of client is not
Abnormal access.Whether present embodiment is that the specific method of determination of abnormal access does not limit for the resource occupation type of client
System.
Wherein, client IP centrality specifically includes: client's number of request is greater than number of request threshold value or client in the unit time
The ratio between number of request and all number of requests of cluster are greater than first threshold.
Client's number of request for example refers to visit of the client in certain time with some IP address for the page of homepage etc.
Ask that number of request, number of request threshold value can consider setting according to service feature and cluster service performance synthesis.Number of request threshold value example
For example 10,000 times, i.e., when the client's number of request for monitoring some client is greater than 10,000 times, determine that the client is abnormal access, this
Embodiment is not particularly limited number of request threshold value.Wherein, client's number of request can be the request to some specific webpage
Number, is also possible to client to the number of request of all pages of cluster.
The resource of client is being determined by the way that whether the ratio between client's number of request and all number of requests of cluster are greater than first threshold
When whether occupancy type is abnormal access, first threshold can be determined according to the quantity of the client of cluster service.For example,
One threshold value can be 50%, i.e., when the ratio between client's number of request of some client and all number of requests of cluster are greater than 50%, determine
The resource occupation type of the client is abnormal access.
Cluster access domain name centrality specifically includes: the same domain name request number of client increases multiple and is greater than first default times
The ratio between total number of request of client of client's domain name request number and same domain name is greater than second threshold in several or predetermined period, and same
The number of request of one domain name is greater than the historical high number of request of the second presupposition multiple.
Specifically, same domain name request number increase multiple be in the current statistic period same domain name request number with before unify
Count the quotient of the difference of same domain name request number and same domain name request number in previous measurement period in the period.Same domain name request number increases
Doubling number can be calculated by statistical history data.First presupposition multiple is, for example, 5, i.e. the same domain name request of certain client
When number is higher than 5 times in a measurement period in the past, that is, it can determine that the resource occupation type of the client is abnormal access.First is pre-
If multiple can be set according to practical business feature, present embodiment is not particularly limited the first presupposition multiple, as long as energy
Enough accurately identify abnormal access.
In addition, total number of request of client refers in certain time, such as in one minute, the institute of common source in CDN cluster
There is client to total number of request of certain domain name, second threshold is, for example, 90%, and historical high number of request is each client couple in history
The largest request number of certain domain name, the second presupposition multiple are, for example, 1.5 times.For example, 30 clients share some CDN cluster
Service Source, it is assumed that each client is little to the demand difference of Service Source, then the domain name request number of a client is generally in client
Within the 10%~20% of total number of request, if the ratio between the domain name request number of some client and the total number of request of client are greater than the second threshold
Value is greater than 90%, and the domain name request number of the client is greater than the historical high number of request of 150% (i.e. 1.5 times), then can determine
The resource occupation type of the client is abnormal access.
Cluster access set of URL neutrality specifically includes: the rate of people logging in of the same URL of client is greater than default rate of people logging in.Specifically,
Default rate of people logging in refers to the access times of same URL in the unit time, and default rate of people logging in can be set according to business demand,
Present embodiment is not particularly limited default rate of people logging in.
Step 204: according to the practical acceleration Service Source behaviour in service of client determine client practical acceleration bandwidth whether
Bandwidth is ensured beyond pre-set client, if ensureing bandwidth beyond client, thens follow the steps 205, if ensureing without departing from client
Bandwidth then terminates.
Wherein, client ensure bandwidth can be calculated according to following data: cluster maximum bandwidth, client priority level (PRI) and
Client's magnitude.For example, client ensures that bandwidth can be calculated by using the following formula:
Band_GN=(1+PN/P)*Band_KN(1)；
Wherein, N is the number of client, Band_GNBandwidth, Band_K are ensured for the client of n-th clientNFor n-th client
Account planning bandwidth, and the sum of account planning bandwidth of all clients of cluster be less than or equal to predetermined fraction cluster it is maximum
Bandwidth；P is the sum of the priority of all clients of cluster, PNPriority for the priority of n-th client, and n-th client is got over
Height, PNIt is bigger.
As shown in the table, it is assumed that client's number of cluster is 4, and the number of each client is respectively K1~K4, and each client's is preferential
Grade is respectively P1~P4, and the account planning bandwidth of each client is respectively Band_K1~Band_K4, can be obtained respectively based on formula (one)
The client of client ensures bandwidth.
| Client | Priority | Account planning bandwidth | Client ensures bandwidth |
| K1 | P1 | Band_K1 | (1+P1/P)*Band_K1 |
| K2 | P2 | Band_K2 | (1+P2/P)*Band_K2 |
| K3 | P3 | Band_K3 | (1+P3/P)*Band_K3 |
| K4 | P4 | Band_K4 | (1+P4/P)*Band_K4 |
Where it is assumed that the value of P1~P4 is respectively 1,2,3,4, then the client of each client ensures that bandwidth difference is as follows:
Band_G1=1.1*Band_K1；
Band_G2=1.2*Band_K2；
Band_G3=1.3*Band_K3；
Band_G4=1.4*Band_K4.
It should be noted that the account planning bandwidth of client can according to cluster maximum bandwidth, client's magnitude of client with
And the priority level initializing of client, client's magnitude is bigger and priority is higher, then the account planning bandwidth of client is bigger.Also, this
In embodiment, according to the height of client priority level (PRI), further account planning bandwidth is weighted to obtain the visitor of client
Family ensures bandwidth, so that the protection of the available bandwidth of the high client of priority is stronger.Right present embodiment is for client
Ensure that the calculation of bandwidth is not particularly limited.
Moreover, it is assumed that cluster maximum bandwidth is Band_MAX, machine service bandwidth B and_FLUX=0.8*Band_ is planned
MAX, also, the Band_FLUX, i.e. Band_K1+Band_K2+ that the sum of account planning bandwidth of all clients of cluster is 80%
Band_K3+Band_K4=0.8*Band_FLUX.The sum of the account planning bandwidth of right present embodiment clients all for cluster
The ratio for accounting for planning machine service bandwidth is not particularly limited.In this way, even if ensureing that band is wider than account planning bandwidth in client
In the case where, it can still ensure that cluster service ability has certain redundancy, and then it is abundant not only to enable each client to enjoy
Bandwidth, and the stability of cluster work can be ensured simultaneously.
In step 204, when the practical acceleration bandwidth of client ensures that bandwidth (has bandwidth increasing beyond pre-set client
Amount is uprushed) when, then 205 are entered step, this processing is otherwise terminated.
Step 205: executing bandwidth increments transfer operation.
In step 205, client can be ensured that the bandwidth increments part of bandwidth is transferred to resource standby pool beyond client.
Wherein, standby resources pond can be located on other clusters in CDN network, and standby resources pond can be common CDN resource, so
It is without being limited thereto.
Cluster caused by the situation that step 202 to step 205 shows in particular abnormal access and bandwidth increments are uprushed etc.
The judgment mode that the use of interior common source is influenced by some client, and give respective acceleration Service Source and make
With control mode, so as to greatly mitigate the influence to other client's common sources.It should be appreciated that in some instances,
The single situation that can be uprushed for abnormal access or bandwidth increments etc. is handled, and present embodiment does not do specific limit to this
System.
Present embodiment is determining client's actual use by the practical acceleration Service Source behaviour in service of monitoring client
Acceleration Service Source it is excessive, when being adversely affected to the client of other common service resources, to the client to accelerate service
The use of resource is controlled, to can guarantee use of other clients to common source, and then ensures that more clients share CDN and add
The service quality of each client when fast Service Source.Also, by provide various ways can accurately identify abnormal access so as to and
When cut off attack, by will exceed client ensure bandwidth bandwidth increments be transferred out, can ensure that client enjoys
Abundant bandwidth and it will not influence the use of cluster other clients to common source.
Third embodiment of the present invention is related to a kind of CDN cluster acceleration serve resources management method, as shown in figure 3, this
Embodiment includes step 301 to step 311.
Step 301 respectively corresponded to step 304 with the step 201 in second embodiment to step 204 it is identical, herein not
It repeats again.
Step 305: what pre-set bandwidths increment exceeded allows beyond duration threshold value.
Allow to be set according to the business characteristic of each client of cluster beyond duration threshold value, allow beyond duration threshold value
For example, 20 minutes, each client's allowed to may be the same or different beyond duration threshold value, for example, client ensure bandwidth compared with
Big client's allows that the client guarantee lesser client of bandwidth can be less than beyond duration threshold value, to avoid bandwidth increments larger
Client influence the use of other clients.Present embodiment is for allowing to be not particularly limited beyond duration threshold value.
Step 306: determining that the practical acceleration bandwidth of client ensures whether the duration of bandwidth reaches permission beyond client
Beyond duration threshold value, if reaching allows to then follow the steps 307 beyond duration threshold value, if not up to allowing beyond duration threshold value,
Terminate.
For example, it is assumed that allowing for some client is 20 minutes beyond duration threshold value, bandwidth is accelerated to exceed when the client is practical
When client ensures that the duration of bandwidth is greater than 20 minutes, 307 are thened follow the steps, accelerates bandwidth beyond client if the client is practical
Ensure that the duration of bandwidth less than 20 minutes, then terminates this processing.This way it is possible to avoid sporadic ground bandwidth increments are uprushed
Caused bandwidth increments transfer bring additionally consumes, and is conducive to the stability of cluster work.
Step 307: target available resources are filtered out from standby resources pond.
Specifically, the available resources in standby resources pond can be ranked up according to predetermined order rule, according to default
Screening rule filters out target available resources from the available resources after sequence.
Wherein, predetermined order rule includes but is not limited to: operator and regional phase according to operator and area with cluster
Same or not divide the priority of the available resources；Wherein, the priority of the identical available resources of operator and area is higher than fortune
Seek quotient and available resources from different places, the available resources in same priority according to available bandwidth size sequence.At one
In example, the sequence of available resources is exemplified below:
1. same operator, same to area, cluster residue service ability are big (i.e. cluster available bandwidth is big)
2. same operator, same to area, cluster residue service ability are small (i.e. cluster available bandwidth is small)
3. same operator, trans-regional, cluster residue service ability is big (i.e. cluster available bandwidth is big)
4. same operator, trans-regional, cluster residue service ability is small (i.e. cluster available bandwidth is small).
Wherein, cluster available bandwidth=planning machine service bandwidth-utilized bandwidth.
Default screening rule includes but is not limited to: filtering out priority height and resource extent is suitable with the bandwidth increments exceeded
Available resources as target available resources.
Step 308: the bandwidth increments that will exceed are transferred to target available resources.
For example, if the bandwidth increments that certain client needs to shift are 100MB, the cluster A's in resource standby pool is preferential
Grade highest, and have the available resource block of 1GB, 200MB and 80MB etc. in cluster A, then it can be by the available money of 200MB in cluster A
The client is distributed to as target available resources in source.This way it is possible to avoid being split to bulk available resources, kept for
The bigger client of bandwidth demand uses.
Step 309: determining whether the practical acceleration bandwidth of client meets switchback condition, if meeting switchback condition, execute
Step 310 cancels standby resources pond, if being unsatisfactory for switchback condition, return step 309.
In one example, step 309 determines whether the practical acceleration bandwidth of client meets switchback condition and specifically include: if
The practical acceleration bandwidth of client is down to client and ensures that the duration within bandwidth reaches preset switchback waiting time, it is determined that
The practical acceleration bandwidth of client meets switchback condition, continues within bandwidth if the practical acceleration bandwidth of client is down to client and is ensured
Duration is not up to preset switchback waiting time, determines that the practical acceleration bandwidth of client is unsatisfactory for switchback condition.Wherein, different visitors
The switchback waiting time at family can be identical, also cannot be different.Switchback waiting time is, for example, 20 minutes, is added when certain client is practical
When the wide duration being down within client's guarantee bandwidth of speed belt is more than or equal to 20 minutes, then by the task in standby resources pond
It switches back, if the practical duration for accelerating bandwidth to be down within client's guarantee bandwidth of the client was returned less than 20 minutes
Step 309.This way it is possible to avoid instantaneity bandwidth increments decline caused by bandwidth increments transfer bring additionally consume, have
Conducive to the stability of cluster work.
Present embodiment is determining client's actual use by the practical acceleration Service Source behaviour in service of monitoring client
Acceleration Service Source it is excessive, when being adversely affected to the client of other common service resources, to the client to accelerate service
The use of resource is controlled, to can guarantee use of other clients to common source, and then ensures that more clients share CDN and add
The service quality of each client when fast Service Source.Also, by provide various ways can accurately identify abnormal access so as to and
When cut off attack, by will exceed client ensure bandwidth bandwidth increments be transferred out, can ensure that client enjoys
Abundant bandwidth and it will not influence the use of cluster other clients to common source.Meanwhile present embodiment can also avoid antithesis
Hair property bandwidth increments uprush or bust is shifted, and distribute suitable available resources for bandwidth increments to be transferred,
Ensure overall quality of service.
4th embodiment of the invention is related to a kind of CDN cluster acceleration serve resources management device, referring to Fig. 4, this
The CDN cluster of embodiment accelerates the serve resources management device 400 to include:
Monitoring modular 401, for monitoring the practical acceleration Service Source behaviour in service of client；
Determining module 402, for determining the shared money in cluster according to the practical acceleration Service Source behaviour in service of client
Whether the use in source is influenced by client, if being influenced by client, trigger control module 403, control module 403 is used for
The use of the acceleration Service Source of client is controlled to inhibit to influence.
In one example, determining module 402 is specifically used for true according to the practical acceleration Service Source behaviour in service of client
Whether the resource occupation type for determining client is abnormal access, if abnormal access, it is determined that the common source in cluster is by visitor
The influence at family, and trigger control module 403, control module 403 are specifically used for forbidding the resource of the client to access to eliminate the effects of the act；
If not abnormal access, determining module 402 then determines the reality of client according to the practical acceleration Service Source behaviour in service of client
Accelerate bandwidth whether to exceed pre-set client and ensure bandwidth, if ensureing bandwidth beyond client, it is determined that sharing in cluster
Resource is influenced by client, and trigger control module 403, at this point, control module 403 then executes bandwidth increments transfer operation.
Wherein it is determined that module 402 specifically when there are following one or any combination thereof, determines the resource occupation class of client
Type is abnormal access, there is no it is following any one or any combination thereof when, determine that the resource occupation type of client is not different
Frequentation is asked:
Client's number of request is greater than number of request threshold value or the ratio between client's number of request and all number of requests of cluster in unit time
Greater than first threshold；
The same domain name request number of client increases multiple asks greater than client's domain name in the first presupposition multiple or predetermined period
The ratio between several and same domain name total number of request of client is asked to be greater than second threshold, and client's domain name request number is greater than the second presupposition multiple
Historical high number of request；Or
The same URL rate of people logging in of client is greater than default rate of people logging in.
In present embodiment, client ensures that bandwidth is calculated according to following data: cluster maximum bandwidth and client are excellent
First grade.In one example, client ensures that bandwidth can be calculated by using the following formula:
Band_GN=(1+PN/P)*Band_KN；
Wherein, N is the number of client, Band_GNBandwidth, Band_K are ensured for the client of n-th clientNFor n-th client
Account planning bandwidth, and the sum of account planning bandwidth of all clients of cluster be less than or equal to predetermined fraction the cluster
Maximum bandwidth；P is the sum of the priority of all clients of cluster, PNFor the priority of n-th client, and n-th client's is preferential
Higher, the P of gradeNIt is bigger.
Control module 403 can specifically carry out bandwidth increments transfer operation in the following ways: pre-set bandwidths increment exceeds
Allow the practical acceleration bandwidth for determining client beyond duration threshold value to ensure whether the duration of bandwidth reaches fair beyond client
Perhaps exceed duration threshold value, if reaching allows beyond duration threshold value, the bandwidth increments that will exceed are transferred to standby resources pond, if not
Reaching allows beyond duration threshold value, then without bandwidth increments transfer operation.Further, control module 403 specifically can be with
The bandwidth increments that will exceed in the following ways are transferred to standby resources pond: target is filtered out from standby resources pond can use money
Source, the bandwidth increments that will exceed are transferred to target available resources.Specifically, predetermined order rule may include: according to operator
With area and the priority of division of available resources whether the operator of cluster and identical area；Wherein, operator is identical with area
The priority of available resources be higher than operator and available resources from different places, the available resources in same priority are according to can
It is sorted with the size of bandwidth.Default screening rule may include: to filter out that priority is high and resource extent increases with the bandwidth exceeded
Comparable available resources are measured as target available resources.In practical applications, control module 403 can also be in the reality for determining client
When border accelerates bandwidth to meet switchback condition, cancel standby resources pond.
Present embodiment is determining client's actual use by the practical acceleration Service Source behaviour in service of monitoring client
Acceleration Service Source it is excessive, when being adversely affected to the client of other common service resources, to the client to accelerate service
The use of resource is controlled, to can guarantee use of other clients to common source, and then ensures that more clients share CDN and add
The service quality of each client when fast Service Source.Also, by provide various ways can accurately identify abnormal access so as to and
When cut off attack, by will exceed client ensure bandwidth bandwidth increments be transferred out, can ensure that client enjoys
Abundant bandwidth and it will not influence the use of cluster other clients to common source.Meanwhile present embodiment can also avoid antithesis
Hair property bandwidth increments uprush and shifted, and distribute suitable available resources for bandwidth increments to be transferred, ensure whole
Service quality.
5th embodiment of the invention is related to a kind of electronic equipment.The electronic equipment of present embodiment can be independent
Server is also possible to server cluster.As shown in figure 5, the electronic equipment further includes memory 502 and processor 501；
Wherein, the memory 502 is stored with the instruction that can be executed by least one described processor 501, described instruction
It is executed by least one described processor 501 to realize above-mentioned all or part of embodiment of the method.
One or more processors 501 and memory 502, in Fig. 5 by taking a processor 501 as an example.Processor 501,
Memory 502 can be connected by bus or other modes, in Fig. 5 for being connected by bus.Memory 502 is used as one
Kind non-volatile computer readable storage medium storing program for executing, it is executable to can be used for storing non-volatile software program, non-volatile computer
Program and module.Non-volatile software program, instruction and the mould that processor 501 is stored in memory 502 by operation
Block realizes that above-mentioned CDN cluster accelerates serve resources management side thereby executing the various function application and data processing of equipment
Method.
Memory 502 may include storing program area and storage data area, wherein storing program area can store operation system
Application program required for system, at least one function.In addition, memory 502 may include high-speed random access memory, may be used also
To include nonvolatile memory, a for example, at least disk memory, flush memory device or the storage of other nonvolatile solid states
Device.In some embodiments, it includes the memory remotely located relative to processor 501 that memory 502 is optional, these are remote
Journey memory can pass through network connection to external equipment.The example of above-mentioned network includes but is not limited to internet, enterprises
Net, local area network, mobile radio communication and combinations thereof.
One or more module is stored in memory 502, when being executed by one or more processor 501, is held
CDN cluster in the above-mentioned any means embodiment of row accelerates serve resources management method.
Above equipment can be performed embodiment of the present invention provided by method, have the corresponding functional module of execution method and
Beneficial effect, the not technical detail of detailed description in the present embodiment, reference can be made to method provided by embodiment of the present invention.
Present embodiment is determining client's actual use by the practical acceleration Service Source behaviour in service of monitoring client
Acceleration Service Source it is excessive, when being adversely affected to the client of other common service resources, to the client to accelerate service
The use of resource is controlled, to can guarantee use of other clients to common source, and then ensures that more clients share CDN and add
The service quality of each client when fast Service Source.Also, by provide various ways can accurately identify abnormal access so as to and
When cut off attack, by will exceed client ensure bandwidth bandwidth increments be transferred out, can ensure that client enjoys
Abundant bandwidth and it will not influence the use of cluster other clients to common source.Meanwhile present embodiment can also avoid antithesis
Hair property bandwidth increments uprush and shifted, and distribute suitable available resources for bandwidth increments to be transferred, ensure whole
Service quality.
Sixth embodiment of the invention is related to a kind of non-volatile memory medium, for storing computer-readable program,
The computer-readable program is used to execute above-mentioned all or part of embodiment of the method for computer.
That is, it will be understood by those skilled in the art that implement the method for the above embodiments be can be with
Relevant hardware is instructed to complete by program, which is stored in a storage medium, including some instructions are to make
It obtains an equipment (can be single-chip microcontroller, chip etc.) or processor (processor) executes side described in each embodiment of the present invention
The all or part of the steps of method.And storage medium above-mentioned includes: USB flash disk, mobile hard disk, read-only memory (ROM, Read-Only
Memory), random access memory (RAM, Random Access Memory), magnetic or disk etc. are various can store journey
The medium of sequence code.
It will be understood by those skilled in the art that the respective embodiments described above are to realize specific embodiments of the present invention,
And in practical applications, can to it, various changes can be made in the form and details, without departing from the spirit and scope of the present invention.
Claims (15)
1. a kind of CDN cluster accelerates serve resources management method characterized by comprising
Monitor the practical acceleration Service Source behaviour in service of client；
According to the practical acceleration Service Source behaviour in service of the client determine the common source in the cluster use whether
Influenced by the client, if being influenced by the client, control the client acceleration Service Source use with
Inhibit the influence.
2. CDN cluster according to claim 1 accelerates serve resources management method, which is characterized in that
The practical acceleration Service Source behaviour in service according to the client determine the common source in the cluster whether by
To the influence of the client, specifically:
Whether the resource occupation type that the client is determined according to the practical acceleration Service Source behaviour in service of the client is different
Frequentation is asked, if abnormal access, it is determined that the common source in the cluster is influenced by the client；And/or
Determine whether the practical acceleration bandwidth of the client exceeds according to the practical acceleration Service Source behaviour in service of the client
Pre-set client ensures bandwidth, if ensureing bandwidth beyond the client, it is determined that common source in the cluster by
The influence of the client.
3. CDN cluster according to claim 2 accelerates serve resources management method, which is characterized in that the client ensures
Bandwidth is calculated according to following data:
Cluster maximum bandwidth and client priority level (PRI).
4. CDN cluster according to claim 2 accelerates serve resources management method, which is characterized in that the client ensures
Bandwidth is calculated by using the following formula:
Band_GN=(1+PN/P)*Band_KN；
Wherein, N is the number of client, Band_GNBandwidth, Band_K are ensured for the client of n-th clientNFor the visitor of n-th client
Bandwidth is planned at family, and the sum of account planning bandwidth of all clients of cluster is less than or equal to the cluster maximum of predetermined fraction
Bandwidth；P is the sum of the priority of all clients of cluster, PNPriority for the priority of n-th client, and n-th client is got over
Height, PNIt is bigger.
5. CDN cluster according to claim 2 accelerates serve resources management method, which is characterized in that described according to
The practical acceleration Service Source behaviour in service of client determines whether the resource occupation type of the client is abnormal access, specific to wrap
It includes:
Following one or any combination thereof if it exists, it is determined that the resource occupation type of the client is abnormal access:
Client's number of request is greater than number of request threshold value or the ratio between client's number of request and all number of requests of cluster in unit time
Greater than first threshold；
The same domain name request number of client increases multiple and is greater than client's domain name request number in the first presupposition multiple or predetermined period
It is greater than second threshold with the ratio between the total number of request of client of same domain name, and client's domain name request number is greater than the second presupposition multiple
Historical high number of request；Or
The same URL rate of people logging in of client is greater than default rate of people logging in.
6. CDN cluster according to claim 2 accelerates serve resources management method, which is characterized in that described in the control
The use of the acceleration Service Source of client is specifically included with inhibiting the influence:
If the resource occupation type of the client is abnormal access, the resource of the client is forbidden to access to eliminate the shadow
It rings；
If the practical acceleration bandwidth of the client ensures bandwidth beyond pre-set client, bandwidth increments transfer behaviour is executed
Make.
7. CDN cluster according to claim 6 accelerates serve resources management method, which is characterized in that the execution bandwidth
Incremental transfer operation, specifically includes:
What pre-set bandwidths increment exceeded allows beyond duration threshold value；
Determine that the practical acceleration bandwidth of client ensures whether the duration of bandwidth reaches described and allow to exceed beyond the client
Duration threshold value, if reaching described allows beyond duration threshold value, the bandwidth increments that will exceed are transferred to standby resources pond.
8. CDN cluster according to claim 7 accelerates serve resources management method, which is characterized in that described to will exceed
Bandwidth increments are transferred to standby resources pond, specifically include:
Target available resources are filtered out from the standby resources pond；
The bandwidth increments that will exceed are transferred to the target available resources.
9. CDN cluster according to claim 8 accelerates serve resources management method, which is characterized in that described from described standby
With target available resources are filtered out in resource pool, specifically include:
The available resources in the standby resources pond are ranked up according to predetermined order rule；
The target available resources are filtered out from the available resources after sequence according to default screening rule.
10. CDN cluster according to claim 9 accelerates serve resources management method, which is characterized in that the predetermined order
Rule includes:
According to operator and area and the priority that divides the available resources whether operator of the cluster and identical area；
Wherein, the available resources of the priority of the identical available resources of operator and area higher than operator and from different places, it is same excellent
Available resources in first grade sort according to the size of available bandwidth.
11. CDN cluster according to claim 9 accelerates serve resources management method, which is characterized in that the default screening
Rule includes:
It filters out priority height and resource extent and the comparable available resources of bandwidth increments exceeded can as the target
Use resource.
12. CDN cluster according to claim 7 accelerates serve resources management method, which is characterized in that in the band that will exceed
After wide incremental transfer to standby resources pond, further includes:
Determine whether the practical acceleration bandwidth of the client meets switchback condition, if meeting the switchback condition, described in cancellation
Standby resources pond.
13. a kind of CDN cluster accelerates serve resources management device characterized by comprising
Monitoring modular, for monitoring the practical acceleration Service Source behaviour in service of client；
Determining module, for determining the shared money in the cluster according to the practical acceleration Service Source behaviour in service of the client
Whether the use in source is influenced by the client, if being influenced by the client, trigger control module,
The control module is used to control the use of the acceleration Service Source of the client to inhibit the influence.
14. a kind of electronic equipment characterized by comprising memory and processor, memory store computer program, processing
Device runs the computer program to realize that the CDN cluster as described in any one of claims 1 to 12 accelerates Service Source pipe
Reason method.
15. a kind of storage medium, which is characterized in that for storing computer-readable program, the computer-readable program is used for
The CDN cluster as described in any one of claims 1 to 12, which is executed, for computer accelerates serve resources management method.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910019458.4A CN109769020B (en) | 2019-01-09 | 2019-01-09 | CDN cluster acceleration service resource management method and device, electronic equipment and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910019458.4A CN109769020B (en) | 2019-01-09 | 2019-01-09 | CDN cluster acceleration service resource management method and device, electronic equipment and storage medium |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109769020A true CN109769020A (en) | 2019-05-17 |
| CN109769020B CN109769020B (en) | 2021-09-24 |
Family
ID=66453949
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910019458.4A Expired - Fee Related CN109769020B (en) | 2019-01-09 | 2019-01-09 | CDN cluster acceleration service resource management method and device, electronic equipment and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109769020B (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114095443A (en) * | 2020-08-24 | 2022-02-25 | 百度在线网络技术（北京）有限公司 | Method and device for managing and controlling flow, electronic equipment and readable storage medium |
| CN114448809A (en) * | 2021-12-22 | 2022-05-06 | 网宿科技股份有限公司 | CDN acceleration resource planning method, device, equipment and storage medium |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20130151702A1 (en) * | 2008-03-31 | 2013-06-13 | Amazon Technologies, Inc. | Request routing based on class |
| CN103186536A (en) * | 2011-12-27 | 2013-07-03 | 中兴通讯股份有限公司 | Method and system for scheduling data shearing devices |
| CN104111876A (en) * | 2014-07-10 | 2014-10-22 | 深圳供电局有限公司 | Dynamic resource management device and method based on Oracle resource plan |
| CN104410691A (en) * | 2014-11-27 | 2015-03-11 | 网宿科技股份有限公司 | Method and system for website acceleration on basis of content delivery network |
| CN105471995A (en) * | 2015-12-14 | 2016-04-06 | 山东省农业机械科学研究院 | High-availability implementation method for large-scale Web server cluster based on SOA |
| CN108021450A (en) * | 2017-12-04 | 2018-05-11 | 北京小度信息科技有限公司 | Job analysis method and apparatus based on YARN |
| CN109167812A (en) * | 2018-08-02 | 2019-01-08 | 网宿科技股份有限公司 | Evaluation services quality, the method for determining adjustable strategies, server and storage medium |
-
2019
- 2019-01-09 CN CN201910019458.4A patent/CN109769020B/en not_active Expired - Fee Related
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20130151702A1 (en) * | 2008-03-31 | 2013-06-13 | Amazon Technologies, Inc. | Request routing based on class |
| CN103186536A (en) * | 2011-12-27 | 2013-07-03 | 中兴通讯股份有限公司 | Method and system for scheduling data shearing devices |
| CN104111876A (en) * | 2014-07-10 | 2014-10-22 | 深圳供电局有限公司 | Dynamic resource management device and method based on Oracle resource plan |
| CN104410691A (en) * | 2014-11-27 | 2015-03-11 | 网宿科技股份有限公司 | Method and system for website acceleration on basis of content delivery network |
| CN105471995A (en) * | 2015-12-14 | 2016-04-06 | 山东省农业机械科学研究院 | High-availability implementation method for large-scale Web server cluster based on SOA |
| CN108021450A (en) * | 2017-12-04 | 2018-05-11 | 北京小度信息科技有限公司 | Job analysis method and apparatus based on YARN |
| CN109167812A (en) * | 2018-08-02 | 2019-01-08 | 网宿科技股份有限公司 | Evaluation services quality, the method for determining adjustable strategies, server and storage medium |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114095443A (en) * | 2020-08-24 | 2022-02-25 | 百度在线网络技术（北京）有限公司 | Method and device for managing and controlling flow, electronic equipment and readable storage medium |
| CN114095443B (en) * | 2020-08-24 | 2023-11-03 | 百度在线网络技术（北京）有限公司 | Method, device, electronic equipment and readable storage medium for controlling flow |
| CN114448809A (en) * | 2021-12-22 | 2022-05-06 | 网宿科技股份有限公司 | CDN acceleration resource planning method, device, equipment and storage medium |
Also Published As
| Publication number | Publication date |
|---|---|
| CN109769020B (en) | 2021-09-24 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110276182B (en) | API distributed current limiting realization method | |
| US9497139B2 (en) | Client-allocatable bandwidth pools | |
| US9154589B1 (en) | Bandwidth-optimized cloud resource placement service | |
| CN109062658A (en) | Realize dispatching method, device, medium, equipment and the system of computing resource serviceization | |
| DE60221019T2 (en) | MANAGING SERVER DEVICES FOR HOST APPLICATIONS | |
| CN104335182A (en) | Method and apparatus for single point of failure elimination for cloud-based applications | |
| Rabbani et al. | On achieving high survivability in virtualized data centers | |
| CN105872079B (en) | A kind of link equalization method based on domain name system DNS | |
| JP2018518744A (en) | Automatic scaling of resource instance groups within a compute cluster | |
| CN107241384A (en) | A kind of content distribution service priority scheduling of resource method based on many cloud frameworks | |
| DE112009000834T5 (en) | Express virtual channels in a packet-switched on-chip connection network | |
| CN106201722A (en) | The load regulation method of server and system | |
| CN104580524A (en) | Resource scaling method and cloud platform with same | |
| CN103607424B (en) | Server connection method and server system | |
| CN104756444A (en) | CDN load balancing in the cloud | |
| CN103916396A (en) | Method for automatic expansion of application examples of cloud platform based on load self-adaption | |
| CN108574645B (en) | Queue scheduling method and device | |
| CN111432247B (en) | Traffic scheduling method, traffic scheduling device, server and storage medium | |
| CN103986766A (en) | Self-adaptation load balancing job task scheduling method and device | |
| CN108845874A (en) | The dynamic allocation method and server of resource | |
| CN106685724A (en) | Node-server management method and device based on election and system | |
| CN109769020A (en) | CDN cluster accelerates serve resources management method and device, electronic equipment and storage medium | |
| CN110191160A (en) | A kind of concurrency control method and device | |
| CN106936877A (en) | A kind of content distribution method, apparatus and system | |
| CN108924203B (en) | Data copy self-adaptive distribution method, distributed computing system and related equipment |
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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| CF01 | Termination of patent right due to non-payment of annual fee | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20210924 Termination date: 20220109 |