CN106790482B - Resource scheduling method and resource scheduling system - Google Patents
Resource scheduling method and resource scheduling system Download PDFInfo
- Publication number
- CN106790482B CN106790482B CN201611143148.6A CN201611143148A CN106790482B CN 106790482 B CN106790482 B CN 106790482B CN 201611143148 A CN201611143148 A CN 201611143148A CN 106790482 B CN106790482 B CN 106790482B
- Authority
- CN
- China
- Prior art keywords
- machine
- data
- decision
- cdn
- scheduling
- 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.)
- Expired - Fee Related
Links
- 238000000034 method Methods 0.000 title claims abstract description 34
- 230000002159 abnormal effect Effects 0.000 claims description 12
- 238000004364 calculation method Methods 0.000 claims description 12
- 238000004422 calculation algorithm Methods 0.000 claims description 9
- 238000012544 monitoring process Methods 0.000 claims description 9
- 239000000126 substance Substances 0.000 claims description 4
- 238000012545 processing Methods 0.000 description 5
- 238000005457 optimization Methods 0.000 description 4
- 230000003068 static effect Effects 0.000 description 4
- 238000007619 statistical method Methods 0.000 description 2
- 230000005540 biological transmission Effects 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 238000013468 resource allocation Methods 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
Images
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L67/00—Network arrangements or protocols for supporting network services or applications
- H04L67/50—Network services
- H04L67/60—Scheduling or organising the servicing of application requests, e.g. requests for application data transmissions using the analysis and optimisation of the required network resources
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L67/00—Network arrangements or protocols for supporting network services or applications
- H04L67/50—Network services
- H04L67/60—Scheduling or organising the servicing of application requests, e.g. requests for application data transmissions using the analysis and optimisation of the required network resources
- H04L67/61—Scheduling or organising the servicing of application requests, e.g. requests for application data transmissions using the analysis and optimisation of the required network resources taking into account QoS or priority requirements
Abstract
The invention provides a resource scheduling method and a resource scheduling system, wherein the resource scheduling system comprises at least two decision machines and an allocation machine, each decision machine controls at least one CDN resource, and the resource scheduling method comprises the following steps: the distributor generates a scheduling task according to the acquired configuration parameters; the distributor sends a scheduling task to each decision machine; and each decision machine draws CDN resources controlled by the decision machine according to the scheduling task. The invention can reduce the risk of resource traction, improve the efficiency of resource scheduling, more efficiently and quickly improve the response to emergency and has better system stability.
Description
Technical Field
The embodiment of the invention relates to the technical field of internet, in particular to a resource scheduling method and a resource scheduling system.
Background
With the increasing popularization of the CDN technology, the service of the CDN is more and more complex and huge, the quality requirement of a customer is more and more high, the current CDN traffic scheduling platform controls the traffic of the CDN whole network through the scheduling of a single machine, and there is no way to meet the quality requirement of the customer already because there are few data dimensions to consider, and when a quality of service problem occurs, the efficiency and accuracy of problem processing are poor, so that fluctuation is easily caused to customer access, resulting in customer complaints.
Specifically, the servers of the existing scheduling platform are deployed on a single machine, only the hardware configuration of the single machine is upgraded when the servers need to be expanded, and all scheduling services need to be migrated when the machines fail, so that the expandability, stability and security of the scheduling system are relatively weak.
The dispatching platform has the advantages that the server is deployed in a single machine, so that certain bottleneck can exist in the processing capacity of the single machine, the business form is more and more complex along with the larger and larger business volume, and the processing efficiency of the single machine can gradually not meet the requirements of customers.
When the scheduling is carried out, too complicated services are not considered when the services are calculated, the timeliness of the collected real-time data can only be transmitted to the scheduling server, the handling capacity and the network problem are obvious, and therefore the scheduling accuracy and timeliness are poor.
Disclosure of Invention
The invention aims to overcome the defects of poor expandability, stability and safety of a scheduling platform of CDN resources in the prior art, and provides a resource scheduling method and a resource scheduling system which improve the usability and expandability of the CDN platform and can efficiently and accurately operate.
The invention solves the technical problems through the following technical scheme:
a resource scheduling method for a resource scheduling system, the resource scheduling system including at least two decision machines and an allocation machine, each decision machine controlling at least one CDN resource, the resource scheduling method comprising:
the distributor generates a scheduling task according to the acquired configuration parameters;
the distributor sends a scheduling task to each decision machine;
and each decision machine draws CDN resources controlled by the decision machine according to the scheduling task.
The method utilizes a resource scheduling system to pull CDN resources. The resource scheduling system is divided into two main modules, namely a distribution machine and a decision machine, wherein the distribution machine monitors and distributes tasks of the decision machines, and the decision machines perform scheduling control decisions on actual CDN resources. Because the decision machine adopts a distributed mode to deploy a plurality of decision machines and can carry out division deployment according to needs, the overall availability and the flexibility of expansion of the system are greatly improved.
Tasks of scheduling decisions of a single machine are divided into a plurality of decision machines, so that the task amount of each scheduling decision is greatly reduced, and each decision machine performs scheduling decisions on the distributed tasks in parallel, so that the scheduling decision time is greatly reduced, and the throughput of the whole system is improved. And because the framework has good expandability, when the business develops to a certain degree, the development can be met by stacking machines.
Because the decision machine is deployed in a distributed manner, different decision machines control different CDN resources, and when the decision machines and the resources are deployed, the different CDN resources are deployed on the same network in principle according to the proximity of the decision machines and the resources, and the different CDN resources only need to upload data such as quality and bandwidth related to the different CDN resources to the decision machine for controlling the use of the different CDN resources, the effectiveness and the accuracy of the data are greatly improved.
Preferably, the resource scheduling system further includes data acquisition machines, the decision machines correspond to the data acquisition machines one to one, and each decision machine draws CDN resources controlled by the decision machine according to the scheduling task includes:
the data acquisition machine acquires service capacity data of CDN resources connected with the corresponding decision machine;
and the decision machine receives the service capacity data and generates scheduling data for pulling CDN resources according to the service capacity data and the scheduling task.
In the invention, the distributor collects configuration data and working state data fed back by the decision machine to generate a scheduling task, and the decision machine executes the scheduling task after receiving the scheduling task.
Furthermore, the decision machine is not only used for executing the scheduling task, but also used for collecting data of the CDN resources, including the data of the CDN resources and the operation data of the CDN server, and performing corresponding adjustment and further optimization of resource traction under the large frame of the scheduling task allocated by the decision machine. The decision machine has certain scheduling capability, so that the burden of the distribution machine can be relieved, and the corresponding speed of the decision machine to emergencies can be accelerated. And the decision machine only needs to feed back the working state after the self-scheduling to the distribution machine, so that the step of specific scheduling of each CDN resource by the distribution machine is omitted, and the effectiveness of the system is further improved.
Preferably, the service capability data includes CDN resource node service capability data and CDN server service capability data, and the resource scheduling method includes:
the decision machine sends decision machine service capability data to the distribution machine.
Preferably, the resource scheduling method includes:
the decision machine acquires the network quality and bandwidth data of the CDN resources controlled by the data acquisition machine, calculates the network quality, bandwidth data and CDN resource configuration parameters to generate CDN resource node service capacity data, acquires hardware indexes and software operation data of a CDN server by the data acquisition machine, and generates CDN server service capacity data by the hardware indexes and the software operation data.
Preferably, the dispenser includes an input interface, and the dispenser generating the scheduling task according to the collected configuration parameters includes:
providing operator information, geographical position information and hardware configuration information of each decision machine through the input interface;
and the distributor acquires the configuration parameters through the input interface and generates a scheduling task.
Preferably, the distributor collecting the configuration parameters and generating the scheduling task through the input interface includes:
generating a scheduling task according to the matching degree of operator information, geographical position information and CDN resources of the decision machine;
and when the matching degrees of the two decision machines are the same, generating a scheduling task according to the hardware configuration information.
Preferably, the generating the scheduling task by the distributor according to the collected configuration parameters includes:
the distributor collects decision machine service capacity data;
and calculating the score of the decision machine according to the service capacity data of the decision machine through a preset algorithm, and generating a scheduling task according to the score.
Preferably, the decision machine service capability data includes hardware configuration data of loads and machines, task number and total task number, data integrity, network status data,
the decision machine score is equal to the sum of the operator and physical location score, the machine service capability score, the machine task score, the machine data integrity score and the network state score;
wherein the content of the first and second substances,
the operator and physical location score is equal to the matching degree of the operator information, the geographic location information and the CDN resources multiplied by the weight of the operator and the physical location;
the machine serviceability score is equal to the hardware configuration data of the load and the machine multiplied by the machine serviceability weight;
the machine task score is equal to the number of tasks and the total number of tasks multiplied by the machine task weight;
the machine data integrity score is equal to the data integrity multiplied by the machine data integrity weight;
the network status score is equal to the network status data multiplied by the network status weight.
Preferably, the data integrity weight is greater than the machine service capability weight, greater than the network status weight, greater than the machine mission weight, and greater than the operator and physical location weight.
Preferably, the decision machines backup each other, and the resource scheduling method includes:
the distributor monitors the working states of all decision machines;
and when the decision machine works abnormally, the decision machine containing the backup of the abnormal decision machine is used for replacing the abnormal decision machine to work.
Preferably, the backup of the decision machines to each other includes:
and numbering the decision machines, and for any decision machine, backing up the decision machine contents of the adjacent numbers of the decision machine numbers.
The invention also includes a resource scheduling system, characterized in that the resource scheduling system includes at least two decision machines and a distribution machine, each decision machine controls at least one CDN resource, the distribution machine includes a calculation module, a transmission module, the decision machine includes a control module,
the calculation module is used for generating a scheduling task according to the configuration parameters acquired by the distributor;
the sending module is used for sending the scheduling task to each decision machine;
and the control module is used for towing CDN resources controlled by the decision machine according to the scheduling task.
Preferably, the resource scheduling system further comprises a data collector, the decision machine and the data collector are in one-to-one correspondence,
the data acquisition machine is used for acquiring service capacity data of CDN resources connected with the corresponding decision machine;
and the control module is used for generating scheduling data for pulling CDN resources according to the service capacity data and the scheduling task.
Preferably, the service capability data includes CDN resource node service capability data and CDN server service capability data;
the CDN resource node service capacity data are generated by calculating the network quality, the bandwidth data and CDN resource configuration parameters, wherein the network quality and the bandwidth data are acquired from CDN resources controlled by a decision machine by a data acquisition machine;
the CDN server service capacity data is generated through calculation of hardware indexes and software operation data of the CDN server, wherein the hardware indexes and the software operation data are acquired from CDN resources controlled by a decision machine through a data acquisition machine.
Preferably, the dispenser further comprises a collection module,
the acquisition module is used for acquiring service capacity data of the decision machine;
the calculation module is also used for calculating the score of the decision machine according to the service capacity data of the decision machine through a preset algorithm and generating a scheduling task according to the score;
the service capability data of the decision machine comprises hardware configuration data of loads and machines, the number of tasks, the total number of tasks, data integrity and network state data,
the decision machine score is equal to the sum of the operator and physical location score, the machine service capability score, the machine task score, the machine data integrity score and the network state score;
wherein the content of the first and second substances,
the operator and physical location score is equal to the matching degree of the operator information, the geographic location information and the CDN resources multiplied by the weight of the operator and the physical location;
the machine serviceability score is equal to the hardware configuration data of the load and the machine multiplied by the machine serviceability weight;
the machine task score is equal to the number of tasks and the total number of tasks multiplied by the machine task weight;
the machine data integrity score is equal to the data integrity multiplied by the machine data integrity weight;
the network status score is equal to the network status data multiplied by the network status weight.
Preferably, the decision machines are provided with numbers, for any decision machine, said any decision machine being adapted to back up adjacently numbered decision machine content, said distribution machine further comprising a monitoring module,
the monitoring module is used for monitoring the working states of all the decision machines, and when the decision machines work abnormally, the decision machines containing the backup of the abnormal decision machines are used for replacing the abnormal decision machines to work.
On the basis of the common knowledge in the field, the above preferred conditions can be combined randomly to obtain the preferred embodiments of the invention.
The positive progress effects of the invention are as follows:
scheduling resources for hierarchical control, wherein the distributor controls a decision machine which controls all CDN resources; different decision machines control different CDN resources; the scheme can greatly reduce the risk of single machine deployment and improve the scheduling decision efficiency.
Decision machines can be backed up mutually, when an upper layer distributor finds that the service of a certain decision machine is problematic, the upper layer distributor intelligently judges according to the physical geographic position among the decision machines, the integrity of basic data, the network condition of the network environment where the decision machine is located and the currently distributed scheduling task amount and then transfers the tasks of the decision machine to other decision machines for processing, and the whole system has high availability.
When the decision machine carries out scheduling switching on CDN resources, multidimensional data are combined: the bandwidth of the machine, the bandwidth of the node, the service capacity of the machine, the network condition of the node, the coverage requirement of the client and the like are comprehensively scheduled, and used data are acquired and transmitted in a distributed manner, so that the scheduling accuracy and effectiveness are greatly improved.
According to the scheme, the usability and the expandability of the CDN platform are greatly improved, the scheduling efficiency and the scheduling accuracy are improved, and the quality of a customer can be guaranteed by performing accurate and rapid scheduling when the customer service has problems.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings needed to be used in the description of the embodiments or the prior art will be briefly introduced below, and it is obvious that the drawings in the following description are some embodiments of the present invention, and it is obvious for those skilled in the art to obtain other drawings based on these drawings without creative efforts.
Fig. 1 is a schematic structural diagram of a resource scheduling system according to embodiment 1 of the present invention.
Fig. 2 is a flowchart of a resource scheduling method according to embodiment 1 of the present invention.
Detailed Description
In order to make the objects, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are some, but not all, embodiments of the present invention. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
Example 1
Referring to fig. 1, the present embodiment provides a resource scheduling system, which includes 5 decision machines 11, one allocation machine 12, 5 data collectors 13, and 10 CDN resources 14, where each decision machine controls at least one CDN resource, and the decision machines and the data collectors are in one-to-one correspondence.
The dispenser further comprises an input module, a calculation module and a sending module, and the decision machine comprises a control module.
The input module comprises an input interface, and the operator information, the geographical position information and the hardware configuration information of each decision machine are acquired through the input interface, wherein the operator information, the geographical position information and the hardware configuration information can be manually input by a user and can be manually and automatically adjusted, so that the static control tasks of the decision machines are generated.
The calculation module is used for generating a scheduling task according to the configuration parameters collected by the distributor.
And the sending module is used for sending the scheduling task to each decision machine.
The data acquisition machine periodically acquires network quality and bandwidth data on the CDN resource node, then performs statistical analysis according to configuration data of the node, and calculates CDN resource node service capacity data of the node, namely CDN resource node service capacity data is generated by calculation of the network quality, the bandwidth data and CDN resource configuration parameters, wherein the network quality and the bandwidth data are acquired from CDN resources controlled by the decision machine through the data acquisition machine.
In addition, the data acquisition machine can also periodically acquire hardware indexes (CPU, internal memory, IO and the like), bandwidth data and software operation data on the CDN server, and then perform statistical analysis by combining configuration data of the machine to calculate the CDN server service capacity data, namely the CDN server service capacity data is generated by calculating the hardware indexes and the software operation data of the CDN server, wherein the hardware indexes and the software operation data are acquired by the data acquisition machine from CDN resources controlled by the decision machine.
And the CDN resource node service capacity data and the CDN server service capacity data are both the service capacity data.
The control module is used for generating scheduling data for pulling CDN resources according to the service capacity data and the scheduling task, wherein the service capacity data comprises CDN resource node service capacity data and CDN server service capacity data.
And the decision machine generates DNS configuration data after scheduling decision according to the scheduling data, and then deploys the DNS configuration data to a DNS server for traction of final flow.
The decision machine of the embodiment is not only used for executing the scheduling task, but also used for collecting data of the CDN resources, including the data of the CDN resources and the operation data of the CDN server, and performing corresponding adjustment and further optimization of resource traction under a large framework of the scheduling task allocated by the decision machine. The decision machine has certain scheduling capability, so that the burden of the distribution machine can be relieved, and the corresponding speed of the decision machine to emergencies can be accelerated. And the decision machine only needs to feed back the working state after the self-scheduling to the distribution machine, so that the step of specific scheduling of each CDN resource by the distribution machine is omitted, and the effectiveness of the system is further improved.
Referring to fig. 2, with the resource scheduling system, this embodiment further provides a resource scheduling method, where the resource scheduling method includes:
The configuration parameters include operator information, geographical location information, and hardware configuration information for each decision machine.
And 103, acquiring the CDN resource node service capacity data and the CDN server service capacity data by the data acquisition machine.
And the CDN resource node service capacity data and the CDN server service capacity data are both the service capacity data.
In the above steps, after the decision machine obtains the scheduling task, further scheduling optimization is performed in the framework of the scheduling task. The framework of scheduling task is an overall rule that specifies which CDN resources are controlled by a decision machine, i.e. a decision machine handles several CDN resource nodes and those several resource nodes, etc. During specific control, the decision machine generates specific DNS configuration data according to service capability data fed back by the data acquisition machine, the decision machine is not only used for executing scheduling tasks, but also used for acquiring data of CDN resources, including the data of the CDN resources and operation data of CDN servers, and the data are utilized to make corresponding adjustment and further optimization of resource traction under a large frame of self-distributed scheduling tasks. The decision machine has certain scheduling capability, so that the burden of the distribution machine can be relieved, and the corresponding speed of the decision machine to emergencies can be accelerated. And the decision machine only needs to feed back the working state after the self-scheduling to the distribution machine, so that the step of specific scheduling of each CDN resource by the distribution machine is omitted, and the effectiveness of the system is further improved.
And step 104, generating scheduling data for pulling CDN resources by the decision machine according to the scheduling task and the service capacity data.
And 105, the decision machine utilizes the scheduling data to pull CDN resources, and sends decision machine service capacity data to the distribution machine.
In this embodiment, after the allocation machine allocates the decision task, although a general resource allocation rule of the decision machine is given, the specific pulling policy is made by the decision machine, that is, the scheduling data. And the decision machine directly runs the scheduling data and feeds back a running result and the current running state to the distributor. The dispenser may again adjust the scheduling tasks as needed.
In addition, when the distributor of this embodiment generates the scheduling task, the operator information, the geographic location information, and the hardware configuration information of each decision machine are provided through the input interface to generate the scheduling task. And further, generating a scheduling task according to the matching degree of operator information and geographical position information of the decision machines and CDN resources, and generating the scheduling task according to the hardware configuration information when the matching degree of the two decision machines is the same.
And the decision machine periodically updates the CDN resource basic configuration data. And the decision-making opportunity receives real-time data which is uploaded from the agent data acquisition machine and is quantized by the CDN resources in real time. For the controlled CDN resources, platform basic configuration data and quantized real-time data are combined, and for CDN server resources with problems in service or poor quality, available resources are selected according to a service quality scheduling mode for replacement, so that the high quality of customers is guaranteed. And performing special scheduling according to the service characteristics of the client, and ensuring the requirement and quality of the client to finally generate data of the scheduling decision.
The resource scheduling method and the resource scheduling system of the embodiment can realize hierarchical control of resource scheduling, and the distributor controls the decision machine which controls all CDN resources; different decision machines control different CDN resources; the scheme can greatly reduce the risk of single machine deployment and improve the scheduling decision efficiency.
Example 2
This embodiment is substantially the same as embodiment 1 except that:
the distributor further comprises a collection module for collecting decision machine service capability data.
And the calculation module is also used for calculating the score of the decision machine according to the service capacity data of the decision machine through a preset algorithm and generating a scheduling task according to the score.
The service capability data of the decision machine comprises hardware configuration data of loads and machines, task quantity and total task quantity, data integrity and network state data.
The decision machine score is obtained by the following formula,
N＝A+B+C+D+E。
wherein, N is the score of the decision machine, A is the score of the operator and the physical position, B is the score of the service capability of the machine, C is the score of the task of the machine, D is the score of the integrity of the data of the machine, and E is the score of the network state.
Specifically, a is a × r, where a is the matching degree of the operator information, the geographic location information, and the CDN resources, and r is the weight of a, that is, the operator and physical location weight.
B is the hardware configuration data of the load and the machine, and s is the weight of B, i.e. the machine service capability weight.
C is C x t, where C is the number of tasks and the total number of tasks, and t is the weight of C, i.e., the machine task weight.
D ═ D × u, where D is the data integrity and u is the weight of D, i.e., the machine data integrity weight.
E ═ E × v, where E is the network state data and v is the weight of E, i.e., the network state weight.
The weight of the data integrity is greater than the weight of the machine service capability, greater than the weight of the network state, greater than the weight of the machine task and greater than the weight of the operator and the physical location.
The allocation opportunity obtains the decision machine basic platform configuration data from the outside, and the data comprises the physical position, the operator, the service available state and relevant hardware configuration information of the decision machine. And manually allocating the resources of the CDN through a UI configuration interface according to the basic platform configuration data of the decision machine and the requirements on services to form a decision machine static allocation task. And the static tasks of the decision machine are transmitted to a dynamic allocation processing link through a memory structure. And the distributor receives real-time running data uploaded from the decision machine in real time. And the distributor generates final scheduling tasks of the decision machines by combining the static allocation tasks and the real-time data operated by the decision machines.
The embodiment provides a specific algorithm for generating the scheduling task, and the scheduling task can be intelligently and automatically generated by using the algorithm, so that the working efficiency of the system is improved. When the decision machine carries out scheduling switching on CDN resources, multidimensional data are combined: the bandwidth of the machine, the bandwidth of the node, the service capacity of the machine, the network condition of the node, the coverage requirement of the client and the like are comprehensively scheduled, and used data are acquired and transmitted in a distributed manner, so that the scheduling accuracy and effectiveness are greatly improved.
With the resource scheduling system, the resource scheduling method provided by this embodiment includes the following specific algorithm for generating the scheduling task,
and collecting service capacity data of the decision machine. And calculating the score of the decision machine according to the service capacity data of the decision machine through a preset algorithm, and generating a scheduling task according to the score.
The service capability data of the decision machine comprises hardware configuration data of loads and machines, task quantity and total task quantity, data integrity and network state data.
The decision machine score is obtained by the following formula,
N＝A+B+C+D+E。
wherein, N is the score of the decision machine, A is the score of the operator and the physical position, B is the score of the service capability of the machine, C is the score of the task of the machine, D is the score of the integrity of the data of the machine, and E is the score of the network state.
Specifically, a is a × r, where a is the matching degree of the operator information, the geographic location information, and the CDN resources, and r is the weight of a, that is, the operator and physical location weight.
B is the hardware configuration data of the load and the machine, and s is the weight of B, i.e. the machine service capability weight.
C is C x t, where C is the number of tasks and the total number of tasks, and t is the weight of C, i.e., the machine task weight.
D ═ D × u, where D is the data integrity and u is the weight of D, i.e., the machine data integrity weight.
E ═ E × v, where E is the network state data and v is the weight of E, i.e., the network state weight.
The resource scheduling method and the resource scheduling system of the embodiment can more efficiently realize the hierarchical control of resource scheduling, and the distributor controls the decision machine which controls all CDN resources; different decision machines control different CDN resources; the scheme can greatly reduce the risk of single machine deployment and improve the scheduling decision efficiency.
Example 3
This embodiment is substantially the same as embodiment 1 except that:
the decision machines are provided with numbers, and for any decision machine, the decision machine is used for backing up the content of the decision machine with adjacent numbers, the distribution machine also comprises a monitoring module,
the monitoring module is used for monitoring the working states of all the decision machines, and when the decision machines work abnormally, the decision machines containing the backup of the abnormal decision machines are used for replacing the abnormal decision machines to work.
With the resource scheduling system, the resource scheduling method of this embodiment includes:
the decision machines are mutually backed up; and the decision machine comprises a number, and for any decision machine, the decision machine content of the adjacent number of the decision machine number is backed up.
The distributor monitors the operating status of all decision machines.
And when the decision machine works abnormally, the decision machine containing the backup of the abnormal decision machine is used for replacing the abnormal decision machine to work.
Example 4
This embodiment is substantially the same as embodiment 1 except that:
the resource scheduling system also comprises a backup machine which performs data synchronization with the distribution machine in real time, and when the distribution machine is in a state, the backup machine is used for generating scheduling tasks.
The above-described embodiments of the apparatus are merely illustrative, and the units described as separate parts may or may not be physically separate, and parts displayed as units may or may not be physical units, may be located in one place, or may be distributed on a plurality of network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of the present embodiment. One of ordinary skill in the art can understand and implement it without inventive effort.
Through the above description of the embodiments, those skilled in the art will clearly understand that each embodiment can be implemented by software plus a necessary general hardware platform, and certainly can also be implemented by hardware. With this understanding in mind, the above-described technical solutions may be embodied in the form of a software product, which can be stored in a computer-readable storage medium such as ROM/RAM, magnetic disk, optical disk, etc., and includes instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to execute the methods described in the embodiments or some parts of the embodiments.
Finally, it should be noted that: the above examples are only intended to illustrate the technical solution of the present invention, but not to limit it; although the present invention has been described in detail with reference to the foregoing embodiments, it will be understood by those of ordinary skill in the art that: the technical solutions described in the foregoing embodiments may still be modified, or some technical features may be equivalently replaced; and such modifications or substitutions do not depart from the spirit and scope of the corresponding technical solutions of the embodiments of the present invention.
Claims (14)
1. A resource scheduling method is used for a resource scheduling system, the resource scheduling system comprises at least two decision machines and an allocation machine, each decision machine controls at least one CDN resource, the resource scheduling method comprises:
the distributor generates a scheduling task according to the acquired configuration parameters;
the distributor sends a scheduling task to each decision machine;
each decision machine draws CDN resources controlled by the decision machine according to the scheduling task;
the resource scheduling system further comprises data acquisition machines, the decision machines correspond to the data acquisition machines one to one, and each decision machine draws CDN resources controlled by the decision machine according to the scheduling task, and the resource scheduling system comprises:
the data acquisition machine acquires service capacity data of CDN resources connected with the corresponding decision machine;
the decision machine receives the service capacity data and generates scheduling data for pulling CDN resources according to the service capacity data and the scheduling tasks;
the decision machine regularly updates CDN resource basic configuration data; the decision machine receives real-time data which are uploaded by the data acquisition machine and are quantized by CDN resources in real time; and selecting available resources for replacing the controlled CDN resources according to a service quality scheduling mode by combining the basic configuration data and the quantized real-time data, wherein the CDN resources have problems in service or have poor quality.
2. The resource scheduling method of claim 1, wherein the service capability data comprises CDN resource node service capability data and CDN server service capability data, and the resource scheduling method comprises:
the decision machine sends decision machine service capability data to the distribution machine.
3. The resource scheduling method of claim 2, wherein the resource scheduling method comprises:
the decision machine acquires the network quality and bandwidth data of the CDN resources controlled by the data acquisition machine, calculates the network quality, bandwidth data and CDN resource configuration parameters to generate CDN resource node service capacity data, acquires hardware indexes and software operation data of a CDN server by the data acquisition machine, and generates CDN server service capacity data by the hardware indexes and the software operation data.
4. The resource scheduling method of claim 1, wherein the dispensing machine includes an input interface, the dispensing machine generating scheduling tasks according to the collected configuration parameters including:
providing operator information, geographical position information and hardware configuration information of each decision machine through the input interface;
and the distributor acquires the configuration parameters through the input interface and generates a scheduling task.
5. The resource scheduling method of claim 4 wherein a distributor collecting the configuration parameters and generating scheduling tasks via the input interface comprises:
generating a scheduling task according to the matching degree of operator information, geographical position information and CDN resources of the decision machine;
and when the matching degrees of the two decision machines are the same, generating a scheduling task according to the hardware configuration information.
6. The resource scheduling method of claim 1 wherein the distributor generating scheduling tasks according to the collected configuration parameters comprises:
the distributor collects decision machine service capacity data;
and calculating the score of the decision machine according to the service capacity data of the decision machine through a preset algorithm, and generating a scheduling task according to the score.
7. The resource scheduling method of claim 6, wherein the decision machine service capability data comprises hardware configuration data of loads and machines, task number and total task number, data integrity, network status data,
the decision machine score is equal to the sum of the operator and physical location score, the machine service capability score, the machine task score, the machine data integrity score and the network state score;
wherein the content of the first and second substances,
the operator and physical location score is equal to the matching degree of the operator information, the geographic location information and the CDN resources multiplied by the weight of the operator and the physical location;
the machine serviceability score is equal to the hardware configuration data of the load and the machine multiplied by the machine serviceability weight;
the machine task score is equal to the number of tasks and the total number of tasks multiplied by the machine task weight;
the machine data integrity score is equal to the data integrity multiplied by the machine data integrity weight;
the network status score is equal to the network status data multiplied by the network status weight.
8. The method of claim 7, wherein the machine data integrity weight is greater than the machine service capability weight is greater than the network status weight is greater than the machine mission weight is greater than the operator and physical location weight.
9. The resource scheduling method of claim 1, wherein the decision machines backup each other, and the resource scheduling method comprises:
the distributor monitors the working states of all decision machines;
and when the decision machine works abnormally, the decision machine containing the backup of the abnormal decision machine is used for replacing the abnormal decision machine to work.
10. The resource scheduling method of claim 9, wherein the decision machines backup each other, comprising:
and numbering the decision machines, and for any decision machine, backing up the decision machine contents of the adjacent numbers of the decision machine numbers.
11. Resource scheduling system, characterized in that it comprises at least two decision machines, each controlling at least one CDN resource, and a distribution machine comprising a calculation module, a sending module, said decision machines comprising a control module,
the calculation module is used for generating a scheduling task according to the configuration parameters acquired by the distributor;
the sending module is used for sending the scheduling task to each decision machine;
the control module is used for towing CDN resources controlled by the decision machine according to the scheduling task;
the resource scheduling system also comprises a data acquisition machine, the decision machine and the data acquisition machine are in one-to-one correspondence,
the data acquisition machine is used for acquiring service capacity data of CDN resources connected with the corresponding decision machine;
the control module is used for generating scheduling data for pulling CDN resources according to the service capacity data and the scheduling task;
the decision machine regularly updates CDN resource basic configuration data; the decision machine receives real-time data which are uploaded by the data acquisition machine and are quantized by CDN resources in real time; and selecting available resources for replacing the controlled CDN resources according to a service quality scheduling mode by combining the basic configuration data and the quantized real-time data, wherein the CDN resources have problems in service or have poor quality.
12. The resource scheduling system of claim 11 wherein the service capability data comprises CDN resource node service capability data and CDN server service capability data;
the CDN resource node service capacity data are generated by calculating network quality, bandwidth data and CDN resource configuration parameters, wherein the network quality and the bandwidth data are acquired by a data acquisition machine from CDN resources controlled by a decision machine;
the CDN server service capacity data is generated through calculation of hardware indexes and software operation data of the CDN server, wherein the hardware indexes and the software operation data are acquired from CDN resources controlled by a decision machine through a data acquisition machine.
13. The resource scheduling system of claim 11 wherein the distributor further comprises an acquisition module,
the acquisition module is used for acquiring service capacity data of the decision machine;
the calculation module is also used for calculating the score of the decision machine according to the service capacity data of the decision machine through a preset algorithm and generating a scheduling task according to the score;
the decision machine service capability data comprises load and hardware configuration data of the machine, task quantity and total task quantity, data integrity and network state data;
the decision machine score is equal to the sum of the operator and physical location score, the machine service capability score, the machine task score, the machine data integrity score and the network state score;
wherein the content of the first and second substances,
the operator and physical location score is equal to the matching degree of the operator information, the geographic location information and the CDN resources multiplied by the weight of the operator and the physical location;
the machine serviceability score is equal to the hardware configuration data of the load and the machine multiplied by the machine serviceability weight;
the machine task score is equal to the number of tasks and the total number of tasks multiplied by the machine task weight;
the machine data integrity score is equal to the data integrity multiplied by the machine data integrity weight;
the network status score is equal to the network status data multiplied by the network status weight;
the weight of the data integrity is greater than the weight of the machine service capability, greater than the weight of the network state, greater than the weight of the machine task and greater than the weight of the operator and the physical location.
14. The resource scheduling system of claim 11 wherein the decision machine is provided with a number, for any decision machine, the any decision machine is used to back up the contents of the decision machine in the adjacent number, the distribution machine further comprises a monitoring module,
the monitoring module is used for monitoring the working states of all the decision machines, and when the decision machines work abnormally, the decision machines containing the backup of the abnormal decision machines are used for replacing the abnormal decision machines to work.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201611143148.6A CN106790482B (en) | 2016-12-13 | 2016-12-13 | Resource scheduling method and resource scheduling system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201611143148.6A CN106790482B (en) | 2016-12-13 | 2016-12-13 | Resource scheduling method and resource scheduling system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN106790482A CN106790482A (en) | 2017-05-31 |
| CN106790482B true CN106790482B (en) | 2020-05-22 |
Family
ID=58880512
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201611143148.6A Expired - Fee Related CN106790482B (en) | 2016-12-13 | 2016-12-13 | Resource scheduling method and resource scheduling system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN106790482B (en) |
Families Citing this family (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107707378B (en) | 2017-06-29 | 2018-11-13 | 贵州白山云科技有限公司 | A kind of CDN covering scheme generation method and device |
| CN107733681B (en) | 2017-07-28 | 2018-10-30 | 贵州白山云科技有限公司 | A kind of scheduling scheme configuration method and device |
| CN107689930A (en) * | 2017-09-08 | 2018-02-13 | 桂林加宏汽车修理有限公司 | A kind of resource regulating method and system |
| WO2019075876A1 (en) * | 2017-10-20 | 2019-04-25 | 华为技术有限公司 | Resource scheduling method and terminal device |
| CN109120688A (en) * | 2018-08-10 | 2019-01-01 | 北京天安智慧信息技术有限公司 | Distributed acquisition method for industrial real-time data |
| CN109587528B (en) * | 2018-12-24 | 2021-06-29 | 中国移动通信集团江苏有限公司 | Method, device, equipment and medium for distributing CDN resources |
| CN113810435A (en) * | 2020-06-11 | 2021-12-17 | 中央广播电视总台 | CDN allocation method, device, terminal and medium |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102065142A (en) * | 2010-12-23 | 2011-05-18 | 网宿科技股份有限公司 | File downloading based scheduling method and system for content delivery network (CDN) |
| CN102291447A (en) * | 2011-08-05 | 2011-12-21 | 中国电信股份有限公司 | Content distribution network load scheduling method and system |
| CN104320487A (en) * | 2014-11-11 | 2015-01-28 | 网宿科技股份有限公司 | HTTP dispatching system and method for content delivery network |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103024001B (en) * | 2012-11-30 | 2018-07-31 | 中兴通讯股份有限公司 | A kind of business scheduling method and device and fusion device |
-
2016
- 2016-12-13 CN CN201611143148.6A patent/CN106790482B/en not_active Expired - Fee Related
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102065142A (en) * | 2010-12-23 | 2011-05-18 | 网宿科技股份有限公司 | File downloading based scheduling method and system for content delivery network (CDN) |
| CN102291447A (en) * | 2011-08-05 | 2011-12-21 | 中国电信股份有限公司 | Content distribution network load scheduling method and system |
| CN104320487A (en) * | 2014-11-11 | 2015-01-28 | 网宿科技股份有限公司 | HTTP dispatching system and method for content delivery network |
Also Published As
| Publication number | Publication date |
|---|---|
| CN106790482A (en) | 2017-05-31 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106790482B (en) | Resource scheduling method and resource scheduling system | |
| CN107124375B (en) | Off-peak scheduling method, system and server for CDN (content delivery network) network bandwidth resources | |
| US20170331705A1 (en) | Resource Scaling Method on Cloud Platform and Cloud Platform | |
| CN106534318B (en) | A kind of OpenStack cloud platform resource dynamic scheduling system and method based on flow compatibility | |
| WO2021104096A1 (en) | Method and apparatus for task scheduling in container cloud environment, and server and storage apparatus | |
| CN109167674B (en) | Service node scoring method, Domain Name System (DNS) scheduling method and server | |
| CN111444019B (en) | Cloud collaborative deep learning model distributed training method and system | |
| CN108345501A (en) | A kind of distributed resource scheduling method and system | |
| CN112148484B (en) | Coupling degree-based micro-service online distribution method and system | |
| CN106227596A (en) | Mission Monitor method and apparatus for task scheduling server | |
| Rathore et al. | Variable threshold-based hierarchical load balancing technique in Grid | |
| CN106452842B (en) | Network system based on network function virtualization intermediary system architecture | |
| CN107346264A (en) | A kind of method, apparatus and server apparatus of virtual machine load balance scheduling | |
| CN109257399A (en) | Cloud platform application management method and management platform, storage medium | |
| CN105404549B (en) | Scheduling virtual machine system based on yarn framework | |
| CN104683450B (en) | Video service monitors cloud system | |
| CN107992392A (en) | A kind of automatic monitoring repair system and method for cloud rendering system | |
| CN102737159A (en) | Operation of a data processing network with multiple geographically decentralised data centres | |
| Chieu et al. | Dynamic resource allocation via distributed decisions in cloud environment | |
| CN103346978A (en) | Method for guaranteeing fairness and stability of virtual machine network bandwidth | |
| CN107231437A (en) | A kind of task backup management method and device | |
| CN104320433B (en) | Data processing method and distributed data processing system | |
| CN103617083B (en) | Store dispatching method and system, job scheduling method and system and management node | |
| CN114416355A (en) | Resource scheduling method, device, system, electronic equipment and medium | |
| CN111352726B (en) | Stream data processing method and device based on containerized micro-service |
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: 20200522 |