WO2020224022A1 - 一种资源调度方法及系统 - Google Patents

一种资源调度方法及系统 Download PDF

Info

Publication number
WO2020224022A1
WO2020224022A1 PCT/CN2019/090318 CN2019090318W WO2020224022A1 WO 2020224022 A1 WO2020224022 A1 WO 2020224022A1 CN 2019090318 W CN2019090318 W CN 2019090318W WO 2020224022 A1 WO2020224022 A1 WO 2020224022A1
Authority
WO
WIPO (PCT)
Prior art keywords
target
bandwidth
node
server
resource 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.)
Ceased
Application number
PCT/CN2019/090318
Other languages
English (en)
French (fr)
Inventor
杨主决
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Wangsu Science and Technology Co Ltd
Original Assignee
Wangsu Science and Technology Co Ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Wangsu Science and Technology Co Ltd filed Critical Wangsu Science and Technology Co Ltd
Priority to EP19874760.2A priority Critical patent/EP3754944B1/en
Priority to US16/867,433 priority patent/US11153370B2/en
Publication of WO2020224022A1 publication Critical patent/WO2020224022A1/zh
Anticipated expiration legal-status Critical
Ceased legal-status Critical Current

Links

Images

Classifications

    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L41/00—Arrangements for maintenance, administration or management of data switching networks, e.g. of packet switching networks
    • H04L41/04—Network management architectures or arrangements
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L41/00—Arrangements for maintenance, administration or management of data switching networks, e.g. of packet switching networks
    • H04L41/04—Network management architectures or arrangements
    • H04L41/045—Network management architectures or arrangements comprising client-server management architectures
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L47/00—Traffic control in data switching networks
    • H04L47/70—Admission control; Resource allocation
    • 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
    • H04L67/01—Protocols
    • H04L67/10—Protocols in which an application is distributed across nodes in the network
    • H04L67/1001—Protocols in which an application is distributed across nodes in the network for accessing one among a plurality of replicated servers
    • 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
    • H04L67/63—Routing a service request depending on the request content or context

Definitions

  • This application relates to the field of Internet technology, and in particular to a resource scheduling method and system.
  • CDN Content Delivery Network
  • CDN service providers can usually allocate multiple node servers in one or more regions for customers according to their actual business bandwidth to meet their business bandwidth requirements. Specifically, because the customer’s business may cover one or more regions, and the customer’s business bandwidth requirements in each region usually have obvious peaks and troughs, in order to meet the customer’s business bandwidth requirements in peak periods in each region, CDN service providers can usually allocate multiple node servers for customers in each region according to the actual business bandwidth of customers during peak periods in each region, and distribute the customer’s business access requests to multiple node servers in each region based on the load balancing strategy. Multiple node servers in each area jointly provide customers with bandwidth resources required by their services.
  • a resource scheduling method includes:
  • the resource scheduling system periodically collects the available bandwidth of the target client of all the node servers corresponding to the target client in the target area, and the real-time bandwidth demand of the target client in the target area;
  • the resource scheduling system determines a target node server among all the node servers according to the real-time bandwidth demand, the bandwidth cost of all the node servers, and the available bandwidth of the target client;
  • the resource scheduling system schedules the service access request of the target client in the target area to the target node server.
  • the resource scheduling system determines the target node server among all the node servers according to the real-time bandwidth demand, the bandwidth cost of all the node servers, and the available bandwidth of the target client, including:
  • the resource scheduling system sequentially selects bandwidth bearer nodes from all the node servers in the order of the bandwidth cost from low to high;
  • the resource scheduling system determines all the bandwidth bearing nodes as the target node server.
  • the method further includes:
  • the resource scheduling system When it is detected that the available bandwidth of the node of the first node server in the target area of the target customer is less than the preset bandwidth warning value, the resource scheduling system sends the target customer's service access request in the target area, Scheduling to all node servers except for the first node server.
  • the method further includes:
  • the resource scheduling system When detecting that the target client has a node server whose available bandwidth of the node is less than the preset bandwidth warning value in the bandwidth bearer server in the target area, the resource scheduling system removes the target node from all the node servers Select the server to be added with the lowest bandwidth cost among the other node servers except the server;
  • the resource scheduling system schedules the service access request of the target client in the target area to the target node server and the server to be added.
  • the method further includes:
  • the resource scheduling system dispatches the target customer’s service access request in the target area to all All node servers corresponding to the target client in the target area.
  • the resource scheduling system determines the target node server among all the node servers according to the real-time bandwidth demand, the bandwidth cost of all the node servers, and the available bandwidth of the target client, including:
  • the resource scheduling system periodically obtains the latest change time of the bandwidth bearing server of the target client in the target area
  • the resource scheduling system determines the target node server among all the node servers according to the real-time bandwidth demand, the bandwidth cost of all the node servers, and the available bandwidth of the target client.
  • the resource scheduling system determines the target node server among all the node servers according to the real-time bandwidth demand, the bandwidth cost of all the node servers, and the available bandwidth of the target client, including:
  • the resource scheduling system determines that the current moment belongs to the low bandwidth demand period according to the real-time bandwidth demand, then according to the real-time bandwidth demand, the bandwidth cost of all the node servers, and the available bandwidth of the target customer, the current time is Determine the target node server in the server.
  • the method further includes:
  • the resource scheduling system determines that the current moment does not belong to the low bandwidth demand period according to the real-time bandwidth demand, it will gradually add other node servers other than the target node server corresponding to the target area as the target node server.
  • the bandwidth bearer server of the target client in the target area is not limited to the bandwidth bearer server of the target client in the target area.
  • a resource scheduling system in a second aspect, includes a data collection layer, an analysis decision layer, and a request scheduling layer, wherein:
  • the data collection layer is used to periodically collect the available bandwidth of the target customer of all node servers corresponding to the target customer in the target area, and the real-time bandwidth demand of the target customer in the target area;
  • the analysis and decision-making layer is configured to determine a target node server among all the node servers according to the real-time bandwidth demand, the bandwidth cost of all the node servers, and the available bandwidth of the target client;
  • the request scheduling layer is used to schedule the service access request of the target client in the target area to the target node server.
  • analysis and decision-making layer is specifically used for:
  • all the bandwidth bearing nodes are determined as target node servers.
  • the request scheduling layer is also used for:
  • the request scheduling layer is also used for:
  • the request scheduling layer is also used for:
  • the business access request of the target customer in the target area is dispatched to the target customer in the target area. All node servers corresponding to the target area.
  • analysis and decision-making layer is specifically used for:
  • the target node server is determined among all the node servers according to the real-time bandwidth demand, the bandwidth cost of all the node servers and the available bandwidth of the target client.
  • analysis and decision-making layer is specifically used for:
  • the target node is determined among all the node servers according to the real-time bandwidth demand, the bandwidth cost of all the node servers and the available bandwidth of the target client server.
  • the request scheduling layer is also used for:
  • the resource scheduling system determines that the current moment does not belong to the low bandwidth demand period according to the real-time bandwidth demand, it will gradually add other node servers other than the target node server corresponding to the target area as the target node server.
  • the bandwidth bearer server of the target client in the target area is not limited to the bandwidth bearer server of the target client in the target area.
  • a network device including:
  • Memory used to store program instructions:
  • the processor is configured to call the program instructions stored in the memory, and execute the resource scheduling method described in the first aspect of the foregoing embodiments according to the obtained program.
  • a computer storage medium stores computer-executable instructions for causing the computer to execute the method steps performed by the network device in the third aspect.
  • the resource scheduling system periodically collects the available bandwidth of the target customer of all node servers corresponding to the target customer in the target area, and the real-time bandwidth demand of the target customer in the target area; the resource scheduling system is based on the real-time bandwidth demand and all nodes
  • the bandwidth cost of the server and the available bandwidth of the target customer are determined among all the node servers.
  • the resource scheduling system dispatches the target customer’s business access requests in the target area to the target node server. In this way, the resource scheduling system can dispatch the customer's service access requests in each area to the node server with lower bandwidth cost in the corresponding area, and the node server with lower bandwidth cost provides the customer with the required services in each area.
  • the bandwidth resource can not only effectively reduce the bandwidth cost of the CDN acceleration service, but also can effectively increase the bandwidth resource utilization rate of the node server with lower bandwidth cost.
  • FIG. 1 is a schematic structural diagram of a resource scheduling system provided by an embodiment of the present application
  • FIG. 2 is a flowchart of a resource scheduling method provided by an embodiment of the present application
  • FIG. 3 is a network scenario diagram of a resource scheduling method provided by an embodiment of the present application.
  • Figure 4 is a network scenario diagram of a resource scheduling method provided by an embodiment of the present application.
  • Fig. 5 is a schematic structural diagram of a network device provided by an embodiment of the present application.
  • the embodiment of the present application provides a resource scheduling method, which can be applied to the resource scheduling system shown in FIG. 1, and the resource scheduling system may include a data collection layer, an analysis decision layer, and a request scheduling layer.
  • the above-mentioned data collection layer, analysis and decision layer, and request scheduling layer may all be independent servers, or may be different components on the same server, or may be a server cluster composed of multiple servers.
  • the resource scheduling system can correspond to the customer’s service access request in each area according to the target customer’s available bandwidth, bandwidth cost and the customer’s real-time bandwidth demand in each area of all node servers corresponding to the customer in each area. Scheduling between node servers.
  • the data collection layer can collect the data needed in the above process; the analysis and decision layer can analyze and process the data collected by the data collection layer; the request scheduling layer can perform corresponding resource scheduling based on the analysis and processing results of the analysis and decision layer.
  • Step 201 The resource scheduling system periodically collects the available bandwidth of the target client of all the node servers corresponding to the target client in the target area, and the real-time bandwidth demand of the target client in the target area.
  • CDN service providers can usually provide customers with network acceleration services through node servers deployed in various regions, and can store customer information (such as customer names, node servers corresponding to customers in various regions, etc.), and node server billing Information (such as billing methods, billing coefficients, etc.) and configuration data such as the rated bandwidth of each node server.
  • customer information such as customer names, node servers corresponding to customers in various regions, etc.
  • node server billing Information such as billing methods, billing coefficients, etc.
  • configuration data such as the rated bandwidth of each node server.
  • the technical personnel of the CDN service provider can schedule the service access requests of customers in each area through the resource scheduling system among the node servers.
  • the resource scheduling system can periodically obtain configuration data such as the above-mentioned customer information, the billing information of the node server, and the rated bandwidth of each node server, and collect various customers (may be called target customers) in various regions (may be called target Area) real-time bandwidth data and real-time bandwidth carrying value of each node server and other real-time data. Then, the resource scheduling system can analyze and process the above configuration data and real-time data to obtain the target customer available bandwidth of all node servers corresponding to the target area of the target customer, and the total amount generated by the target customer on all corresponding node servers in the target area. Real-time bandwidth data, the total real-time bandwidth data is determined as the real-time bandwidth demand of the target customer in the target area.
  • the available bandwidth of the target client consists of the remaining bandwidth of the node server and the real-time occupied bandwidth of the target client on the node server.
  • the remaining bandwidth of the node server is 500Mb/s
  • the real-time occupied bandwidth of the target client on the node server is 200Mb/s
  • the available bandwidth of the target customer is 700Mb/s.
  • the resource scheduling system can also store the real-time data after the above analysis and processing on the big data platform to provide a data basis for the CDN service provider to perform other data analysis. After that, the resource scheduling system can schedule the service access requests of the target customers in the target area according to the available bandwidth of the target customers of all the node servers and the real-time bandwidth requirements of the target customers in the target area.
  • Step 202 The resource scheduling system determines the target node server among all the node servers according to the real-time bandwidth demand, the bandwidth cost of all node servers and the available bandwidth of the target customer.
  • the resource scheduling system can determine the bandwidth cost of each node server according to the charging information such as the charging method and charging coefficient of each node server, and obtain the bandwidth cost of all node servers corresponding to the target customer in the target area. Then, the resource scheduling system can determine the target customers from all the node servers corresponding to the target customers in the target area based on the bandwidth cost of all the node servers mentioned above, the available bandwidth of the target customer, and the target customer’s real-time bandwidth demand in the target area.
  • the bandwidth of the node server may be called the target node server).
  • the resource scheduling system may preferentially select a node server with a lower bandwidth cost to carry client bandwidth requirements.
  • the processing of the above step 202 may be as follows: the resource scheduling system in the order of bandwidth cost from low to high, in all nodes The bandwidth bearing nodes are selected sequentially from the server; when the total available bandwidth of the target client of all the selected bandwidth bearing nodes is greater than the real-time bandwidth demand, the resource scheduling system determines all the bandwidth bearing nodes as the target node server.
  • the resource scheduling system collects the available bandwidth of the target customer of all the node servers corresponding to the target customer in the target area, and the real-time bandwidth demand of the target customer in the target area, and can sort all the node servers mentioned above in descending order according to the bandwidth cost Arrange and select bandwidth bearer nodes from the node servers corresponding to the target customers in the target area in the order of bandwidth cost from lowest to highest, and determine whether the total available bandwidth of the target customers of the bandwidth bearer nodes is greater than the target customer’s in the target area Real-time bandwidth requirements.
  • the resource scheduling system can determine the bandwidth bearing node as the target node server.
  • bandwidth bearer nodes when selecting bandwidth bearer nodes according to the bandwidth cost, if multiple unselected node servers have the same bandwidth cost and have the lowest bandwidth cost among all unselected node servers, the multiple nodes can be The server is also selected as the bandwidth bearer node.
  • the CDN service provider usually pays according to the bandwidth usage of each node server and the corresponding billing methods (such as 95 billing, peak billing, bandwidth buyout billing, and actual traffic billing, etc.)
  • the operator pays the fee, so the higher the billing coefficient for the CDN service provider, the higher the bandwidth cost, and the bandwidth cost of the node server charged according to the actual traffic is higher than 95 billing, peak billing, and bandwidth buyout
  • the node server for billing methods such as billing, and the node server for 95 billing, peak billing, and bandwidth buyout billing are not billed according to the actual traffic, and their bandwidth costs can be considered equal.
  • node 1 billing coefficient 8.5
  • billing coefficient 4.8 billing based on actual traffic
  • node 3 95 billing
  • node 4 bandwidth buy-out billing
  • the real-time bandwidth demand of the target customer in the target area is 2000Mb/s as an example, suppose the available bandwidth of the target customer of node 1 is 1000Mb/s, and the available bandwidth of the target customer of node 2 is 800Mb/s, the available bandwidth of the target customer of node 3 is 600Mb/s, and the available bandwidth of the target customer of node 4 is 900Mb/s.
  • the resource scheduling system can arrange the above nodes in descending order of bandwidth cost as follows:
  • Node 1 The billing coefficient is 8.5, the available bandwidth of the target customer is 1000Mb/s, and the actual traffic is charged;
  • Node 2 The billing coefficient is 4.8, the available bandwidth of the target customer is 800Mb/s, and the actual traffic is charged;
  • Node 3 The available bandwidth of the target customer is 600Mb/s, 95 billing;
  • Node 4 Available bandwidth of the target customer is 900Mb/s, and bandwidth buy-out is charged.
  • the resource scheduling system can first select the node server with the lowest bandwidth cost, namely node 3 and node 4, as the bandwidth bearer nodes. Since the total available bandwidth of the target customers of node 3 and node 4 is 1500Mb/s, which is less than the real-time target customer in the target area The bandwidth requirement is 2000Mb/s, and the resource scheduling system can further select node 2 as the bandwidth bearer node.
  • the resource scheduling system can determine the aforementioned node 2, node 3, and node 4 as the target node server, that is, node 2, node 3, and node 4 carry service access requests of target customers in the target area.
  • the resource scheduling system sets the minimum number of bandwidth-carrying nodes (may be called the minimum reserved number) of the target client in the target area. Still take all the node servers corresponding to the above target customers in the target area as node 1: billing coefficient 8.5, billing according to actual traffic, node 2: billing coefficient 4.8, billing according to actual traffic, node 3: 95 billing, Node 4: Bandwidth buyout billing.
  • the target customer’s real-time bandwidth demand in the target area is 1000Mb/s.
  • the resource scheduling system can first assign the three node servers with the lowest bandwidth cost, namely, node 3 and Node 4 is selected as the bandwidth bearer node. At this time, node 2, node 3, and node 4 can carry the real-time bandwidth requirements of the target customer in the target area, and the resource scheduling system can determine node 2, node 3, and node 4 as target node servers.
  • Step 203 The resource scheduling system schedules the service access request of the target client in the target area to the target node server.
  • the resource scheduling system can dispatch all service access requests of the target client in the target area to all target node servers. Still taking all the node servers corresponding to the target customer in the target area as node 1, node 2, node 3, node 4, and the target node server as node 1, node 2, the resource scheduling system can only allow node 3 and node 4 to be shared Provide the bandwidth resources required by the real-time bandwidth requirements of the target customers in the target area, as shown in Figure 4. That is, the resource scheduling system can set the bandwidth bearer nodes of the target customers in the target area to node 3 and node 4 on the load balancing device. , So that the load balancing device can distribute the service access request of the target client in the target area to the aforementioned node 3 and node 4.
  • the resource scheduling system can temporarily stop scheduling service access requests to the node server, and the corresponding processing can be as follows: When the target client is detected in the target area When the node available bandwidth of a node server is less than the preset bandwidth warning value, the resource scheduling system dispatches the service access request in the target area of the target customer to all the node servers except the first node server.
  • the technical staff of the CDN service provider can set the early warning value of the available bandwidth of each node server through the resource scheduling system (which can be called the preset bandwidth warning value).
  • the resource scheduling system can temporarily stop dispatching the target client's business access request in the target area to the first node server , Dispatch the target client's service access request in the target area to the other node servers corresponding to the target client in the target area except the first node server.
  • the billing method is that the bandwidth cost of the node server charged according to the actual traffic increases with the increase of the actual traffic, so the above-mentioned preset bandwidth warning value may not be set for this type of node server.
  • the resource scheduling system can determine whether to add a bandwidth bearer server according to the available bandwidth of the node except the target node server.
  • the following processing can be performed: when the target client is detected in the bandwidth bearer server in the target area
  • the resource scheduling system selects the server to be added with the lowest bandwidth cost among all node servers except the target node server; the resource scheduling system will target customers
  • the service access request in the target area is dispatched to the target node server and the server to be added.
  • the resource scheduling system can detect the available bandwidth of the node of the bandwidth bearer server of the target client in the target area. When detecting that there is a node server whose node available bandwidth is less than the preset bandwidth warning value in the bandwidth bearer server, the resource scheduling system can select the lowest bandwidth cost among the remaining node servers corresponding to the target customer in the target area except the target node server Then, the target client's service access request in the target area is dispatched to the target node server and the server to be added, that is, the target node server and the server to be added together serve as the target client's bandwidth bearing server in the target area.
  • the selected server to be added can be the node server with the lowest bandwidth cost, or multiple node servers with the same bandwidth cost and the lowest. Further, if after selecting the server to be added once, it is detected again that there is a node server whose available bandwidth of the node is less than the preset bandwidth warning value in the bandwidth bearer server, you can continue to select the server to be added as the bandwidth again from the remaining node servers Host server.
  • the resource scheduling system can determine whether to resume scheduling according to the rate of change of the real-time bandwidth demand of the target customer in the target area.
  • the following processing can be performed: When the real-time bandwidth requirement of the target customer in the target area is detected The growth rate of bandwidth demand is greater than the preset increase threshold, and the resource scheduling system dispatches the service access request of the target customer in the target area to all the node servers corresponding to the target customer in the target area.
  • the resource scheduling system can detect and record the real-time bandwidth bearing value of the target client in the target area corresponding to the other node servers except the target node server according to the preset detection period.
  • the resource scheduling system can obtain the real-time bandwidth demand of the target customer in the target area in the previous preset detection period and the real-time bandwidth demand of the target customer in the target area at the current moment, and calculate the growth rate of the real-time bandwidth demand of the target customer in the target area ,
  • the calculation method can be as follows:
  • the technical staff of the CDN service provider can set the adjacent preset detection for the target customer in the target area through the resource scheduling system
  • the highest value of the growth rate of the periodic real-time bandwidth carrying value may be referred to as the preset increase threshold.
  • the resource scheduling system detects that the above growth rate is greater than the preset increase threshold, the resource scheduling system can schedule the target client's service access request in the target area to all node servers corresponding to the target client in the target area.
  • the resource scheduling system can set the shortest time interval between two changes of the bandwidth bearer server for the same client in the same area.
  • the processing of step 202 above can be as follows: the resource scheduling system periodically obtains the target client's The bandwidth of the target area bears the latest change time of the server; the resource scheduling system determines whether the interval between the latest change and the current time is greater than the preset interval; if so, the resource scheduling system will base on the real-time bandwidth demand and the bandwidth cost of all node servers And the available bandwidth of the target customer, determine the target node server among all the node servers.
  • the technical staff of the CDN service provider can record the change time of the bandwidth bearing server of the target customer in the target area through the resource scheduling system, and set it as the target customer
  • the minimum interval between two adjacent changes of the bandwidth bearer server in the target area may be referred to as the preset interval. Therefore, the resource scheduling system can obtain the latest change time of the bandwidth bearing server of the target client in the target area, calculate the interval between the latest change and the current time, and determine whether the interval is greater than the preset interval.
  • the resource scheduling system can determine the target node server among all the node servers according to the real-time bandwidth requirements, the bandwidth cost of all node servers and the available bandwidth of the target customer; otherwise, the resource scheduling system may not re-determine the target node server.
  • the foregoing preset interval duration must be greater than the period duration in step 201. Therefore, using the foregoing method can effectively prevent the bandwidth bearing server from being removed in a short time after the bandwidth bearing server is added, effectively avoiding repetition. Resource waste caused by scheduling.
  • the resource scheduling system may only perform the deletion processing of the bandwidth bearer server during the low bandwidth demand period.
  • the processing of step 202 above may be as follows: if the resource scheduling system determines that the current moment belongs to the low bandwidth demand period according to the real-time bandwidth demand , According to the real-time bandwidth requirements, the bandwidth cost of all node servers and the available bandwidth of the target customer, the target node server is determined among all the node servers.
  • the resource scheduling system can first determine whether the current moment belongs to the low bandwidth demand period according to the real-time bandwidth demand. If the current moment is in the low period of bandwidth demand, the resource scheduling system can determine the target node server among all the node servers according to the real-time bandwidth demand, the bandwidth cost of all node servers and the available bandwidth of the target customer, and schedule the target customer’s business access request To all node servers corresponding to the target customer in the target area. If the current moment is not in the low bandwidth demand period, the bandwidth bearing server of the target customer in the target area can be kept unchanged, or the bandwidth bearing server that was removed before can be restored.
  • the technical staff of the CDN service provider can also input the resource scheduling recovery instructions of the target customer in the target area through the UI interface of the resource scheduling system according to the actual bandwidth demand of the target customer, and then the resource scheduling system can also send the target customer The service access request is dispatched to all the node servers pointed to by the resource dispatch recovery instruction.
  • the resource scheduling system can perform the processing of adding the bandwidth bearer server during the non-bandwidth demand low period, and the corresponding processing can be as follows: if the resource scheduling system determines that the current moment does not belong to the low bandwidth demand period according to the real-time bandwidth demand, it will target customers Node servers other than the target node server corresponding to the target area are gradually added as bandwidth bearer servers for target customers in the target area.
  • the resource scheduling system can restore the previously eliminated bandwidth bearer server of the target customer in the target area That is, the resource scheduling system can gradually add other node servers other than the target node server corresponding to the target customer in the target area as the bandwidth bearing server of the target customer in the target area. It is worth mentioning that the step-by-step process here can be to add a specified number of bandwidth bearing servers at a fixed time interval.
  • the resource scheduling system periodically collects the available bandwidth of the target customer of all node servers corresponding to the target customer in the target area, and the real-time bandwidth demand of the target customer in the target area; the resource scheduling system is based on the real-time bandwidth demand and all nodes
  • the bandwidth cost of the server and the available bandwidth of the target customer are determined among all the node servers.
  • the resource scheduling system dispatches the target customer’s business access requests in the target area to the target node server. In this way, the resource scheduling system can dispatch the customer's service access requests in each area to the node server with lower bandwidth cost in the corresponding area, and the node server with lower bandwidth cost provides the customer with the required services in each area.
  • the bandwidth resource can not only effectively reduce the bandwidth cost of the CDN acceleration service, but also can effectively increase the bandwidth resource utilization rate of the node server with lower bandwidth cost.
  • an embodiment of the present application also provides a resource scheduling system.
  • the system includes a data collection layer 101, an analysis decision layer 102, and a request scheduling layer 103, where:
  • the data collection layer 101 is configured to periodically collect the available bandwidth of the target customer of all node servers corresponding to the target customer in the target area, and the real-time bandwidth demand of the target customer in the target area;
  • the analysis and decision-making layer 102 is configured to determine a target node server among all the node servers according to the real-time bandwidth demand, the bandwidth cost of all the node servers, and the available bandwidth of the target client;
  • the request scheduling layer 103 is configured to schedule the service access request of the target client in the target area to the target node server.
  • analysis and decision-making layer 102 is specifically used for:
  • all the bandwidth bearing nodes are determined as target node servers.
  • the request scheduling layer 103 is also used for:
  • the request scheduling layer 103 is also used for:
  • the request scheduling layer 103 is also used for:
  • the business access request of the target customer in the target area is dispatched to the target customer in the target area. All node servers corresponding to the target area.
  • analysis and decision-making layer 102 is specifically used for:
  • the target node server is determined among all the node servers according to the real-time bandwidth demand, the bandwidth cost of all the node servers and the available bandwidth of the target client.
  • analysis and decision-making layer 102 is specifically used for:
  • the target node is determined among all the node servers according to the real-time bandwidth demand, the bandwidth cost of all the node servers and the available bandwidth of the target client server.
  • the request scheduling layer 103 is also used for:
  • the resource scheduling system determines that the current moment does not belong to the low bandwidth demand period according to the real-time bandwidth demand, it will gradually add other node servers other than the target node server corresponding to the target area as the target node server.
  • the bandwidth bearer server of the target client in the target area is not limited to the bandwidth bearer server of the target client in the target area.
  • the resource scheduling system periodically collects the available bandwidth of the target customer of all node servers corresponding to the target customer in the target area, and the real-time bandwidth demand of the target customer in the target area; the resource scheduling system is based on the real-time bandwidth demand and all nodes
  • the bandwidth cost of the server and the available bandwidth of the target customer are determined among all the node servers.
  • the resource scheduling system dispatches the target customer’s business access requests in the target area to the target node server. In this way, the resource scheduling system can dispatch the customer's service access requests in each area to the node server with lower bandwidth cost in the corresponding area, and the node server with lower bandwidth cost provides the customer with the required services in each area.
  • the bandwidth resource can not only effectively reduce the bandwidth cost of the CDN acceleration service, but also can effectively increase the bandwidth resource utilization rate of the node server with lower bandwidth cost.
  • the embodiment of the present application also provides a network device, as shown in FIG. 5, including:
  • the memory 501 is used to store program instructions:
  • the processor 502 is configured to call program instructions stored in the memory, and execute the resource scheduling method described in steps 201-203 in the foregoing embodiment according to the obtained program.
  • the embodiments of the present application also provide a computer storage medium, and the computer storage medium stores computer executable instructions, and the computer executable instructions are used to make the computer execute the method steps executed by the network device in the foregoing embodiment.

Landscapes

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

Abstract

本申请提供了一种资源调度方法及系统，所述方法包括：资源调度系统周期性采集目标客户在目标区域对应的所有节点服务器的目标客户可用带宽，和所述目标客户在所述目标区域内的实时带宽需求；所述资源调度系统根据所述实时带宽需求、所述所有节点服务器的带宽成本和所述目标客户可用带宽，在所述所有节点服务器中确定目标节点服务器；所述资源调度系统将所述目标客户在所述目标区域内的业务访问请求，调度至所述目标节点服务器。采用本申请可以有效降低CDN加速服务的带宽成本，还可以提高带宽成本较低的节点服务器的带宽资源的使用率。

Description

一种资源调度方法及系统
交叉引用
本申请引用于2019年05月05日递交的名称为“一种资源调度方法及系统”的第201910368450.9号中国专利申请，其通过引用被全部并入本申请。
技术领域
本申请涉及互联网技术领域，特别涉及一种资源调度方法及系统。
背景技术
随着互联网技术的不断发展，越来越多的业务提供方(即下文中的客户)使用CDN(Content Delivery Network，内容分发网络)服务对其业务进行加速。CDN服务商可以根据客户的业务带宽需求为客户分配带宽资源。
CDN服务商通常可以根据客户的实际业务带宽为客户在一个或者多个区域分配多台节点服务器以满足客户的业务带宽需求。具体来说，由于客户的业务可能会覆盖一个或者多个区域，且客户在各区域的业务带宽需求通常均存在明显的高峰期和低谷期，为满足客户在各区域高峰期的业务带宽需求，CDN服务商通常可以根据各区域高峰期内客户的实际业务带宽为客户在各区域分配多台节点服务器，并基于负载均衡策略将客户的业务访问请求分发至各区域内的多台节点服务器上，由各区域内的多台节点服务器共同为客户提供其业务所需的带宽资源。
在实现本申请的过程中，申请人发现现有技术至少存在以下问题：
在各区域的业务带宽需求低谷期，由于客户的实际业务带宽较低，所以给客户分配的各节点服务器的带宽资源的使用率较低，受采用的流量计费方式不同的影响，不同节点服务器的带宽成本不同，在带宽成本较高的节点服务器为客户提供带宽资源时，若带宽成本较低的节点服务器的存在大量闲置的带宽资源，则会一定程度上增加CDN加速服务的整体带宽成本。
发明内容
为了解决现有技术的问题，本申请实施例提供了一种资源调度方法及系统。所述技术方案如下：
第一方面，提供了一种资源调度方法，所述方法包括：
资源调度系统周期性采集目标客户在目标区域对应的所有节点服务器的目标客户可用带宽，和所述目标客户在所述目标区域内的实时带宽需求；
所述资源调度系统根据所述实时带宽需求、所述所有节点服务器的带宽成本和所述目标客户可用带宽，在所述所有节点服务器中确定目标节点服务器；
所述资源调度系统将所述目标客户在所述目标区域内的业务访问请求，调度至所述目标节点服务器。
可选的，所述资源调度系统根据所述实时带宽需求、所述所有节点服务器的带宽成本和所述目标客户可用带宽，在所述所有节点服务器中确定目标节点服务器，包括：
所述资源调度系统按照所述带宽成本从低到高的顺序，在所述所有节点服务器中依次选取带宽承载节点；
当选取到的所有带宽承载节点的目标客户可用带宽之和大于所述实时带宽需求时，所述资源调度系统将所述所有带宽承载节点确定为目标节点服务器。
可选的，所述方法还包括：
当检测到所述目标客户在所述目标区域内的第一节点服务器的节点可用带宽小于预设带宽预警值时，所述资源调度系统将所述目标客户所述目标区域内的业务访问请求，调度至所述所有节点服务器中除所述第一节点服务器之外的其余节点服务器。
可选的，所述资源调度系统将所述目标客户在所述目标区域内的业务访问请求，调度至所述目标节点服务器之后，还包括：
当检测到所述目标客户在所述目标区域内的带宽承载服务器中存在节点可用带宽小于预设带宽预警值的节点服务器时，所述资源调度系统则在所述所有节点服务器除所述目标节点服务器之外的其余节点服务器中选取带宽成本最低的待添加服务器；
所述资源调度系统将所述目标客户在所述目标区域内的业务访问请求调度至所述目标节点服务器和所述待添加服务器。
可选的，所述资源调度系统将所述目标客户在所述目标区域内的业务访问请求，调度至所述目标节点服务器之后，还包括：
当检测到所述目标客户在所述目标区域内的实时带宽需求的增长率大于预设增幅阈值，所述资源调度系统则将所述目标客户在所述目标区域内的业务访问请求调度至所述目标客户在所述目标区域对应的所有节点服务器。
可选的，所述资源调度系统根据所述实时带宽需求、所述所有节点服务器的带宽成本和所述目标客户可用带宽，在所述所有节点服务器中确定目标节点服务器，包括：
所述资源调度系统周期性获取所述目标客户在所述目标区域内的带宽承载服务器的最近变更时刻；
所述资源调度系统确定所述最近变更时刻与当前时刻的间隔时长是否大于预设间隔时长；
如果是，所述资源调度系统则根据所述实时带宽需求、所述所有节点服务器的带宽成本和所述目标客户可用带宽，在所述所有节点服务器中确定目标节点服务器。
可选的，所述资源调度系统根据所述实时带宽需求、所述所有节点服务器的带宽成本和所述目标客户可用带宽，在所述所有节点服务器中确定目标节点服务器，包括：
所述资源调度系统如果根据所述实时带宽需求确定当前时刻属于带宽需求低谷期，则根据所述实时带宽需求、所述所有节点服务器的带宽成本和所述目标客户可用带宽，在所述所有节点服务器中确定目标节点服务器。
可选的，所述方法还包括：
所述资源调度系统如果根据所述实时带宽需求确定当前时刻不属于带宽需求低谷期，则将所述目标客户在所述目标区域对应的所述目标节点服务器之外的其它节点服务器逐步添加为所述目标客户在所述目标区域内的带宽承载服务器。
第二方面，提供了一种资源调度系统，所述系统包括数据采集层、分析决策层和请求调度层，其中：
所述数据采集层，用于周期性采集目标客户在目标区域对应的所有节点服 务器的目标客户可用带宽，和所述目标客户在所述目标区域内的实时带宽需求；
所述分析决策层，用于根据所述实时带宽需求、所述所有节点服务器的带宽成本和所述目标客户可用带宽，在所述所有节点服务器中确定目标节点服务器；
所述请求调度层，用于将所述目标客户在所述目标区域内的业务访问请求，调度至所述目标节点服务器。
可选的，所述分析决策层，具体用于：
按照所述带宽成本从低到高的顺序，在所述所有节点服务器中依次选取带宽承载节点；
当选取到的所有带宽承载节点的目标客户可用带宽之和大于所述实时带宽需求时，将所有所述带宽承载节点确定为目标节点服务器。
可选的，所述请求调度层，还用于：
当检测到所述目标客户在所述目标区域内的第一节点服务器的节点可用带宽小于预设带宽预警值时，将所述目标客户所述目标区域内的业务访问请求，调度至所述所有节点服务器中除所述第一节点服务器之外的其余节点服务器。
可选的，所述请求调度层，还用于：
当检测到所述目标客户在所述目标区域内的带宽承载服务器中存在节点可用带宽小于预设带宽预警值的节点服务器时，则在所述所有节点服务器除所述目标节点服务器之外的其余节点服务器中选取带宽成本最低的待添加服务器；
将所述目标客户在所述目标区域内的业务访问请求调度至所述目标节点服务器和所述待添加服务器。
可选的，所述请求调度层，还用于：
当检测到所述目标客户在所述目标区域内的实时带宽需求的增长率大于预设增幅阈值，将所述目标客户在所述目标区域内的业务访问请求调度至所述目标客户在所述目标区域对应的所有节点服务器。
可选的，所述分析决策层，具体用于：
周期性获取所述目标客户在所述目标区域内的带宽承载服务器的最近变更时刻；
确定所述最近变更时刻与当前时刻的间隔时长是否大于预设间隔时长；
如果是，则根据所述实时带宽需求、所述所有节点服务器的带宽成本和所 述目标客户可用带宽，在所述所有节点服务器中确定目标节点服务器。
可选的，所述分析决策层，具体用于：
如果根据所述实时带宽需求确定当前时刻属于带宽需求低谷期，则根据所述实时带宽需求、所述所有节点服务器的带宽成本和所述目标客户可用带宽，在所述所有节点服务器中确定目标节点服务器。
可选的，所述请求调度层，还用于：
所述资源调度系统如果根据所述实时带宽需求确定当前时刻不属于带宽需求低谷期，则将所述目标客户在所述目标区域对应的所述目标节点服务器之外的其它节点服务器逐步添加为所述目标客户在所述目标区域内的带宽承载服务器。
第三方面，提供了一种网络设备，包括：
存储器，用于存储程序指令:
处理器，用于调用所述存储器中存储的程序指令，按照获得的程序执行前述实施例中的第一方面所述的资源调度方法。
第四方面，还提供了一种计算机存储介质，计算机存储介质存储有计算机可执行指令，所述计算机可执行指令用于使所述计算机执行第三方面中网络设备所执行的方法步骤。
本申请实施例提供的技术方案带来的有益效果是：
在本实施例中，资源调度系统周期性采集目标客户在目标区域对应的所有节点服务器的目标客户可用带宽，和目标客户在目标区域内的实时带宽需求；资源调度系统根据实时带宽需求、所有节点服务器的带宽成本和目标客户可用带宽，在所有节点服务器中确定目标节点服务器；资源调度系统将目标客户在目标区域内的业务访问请求，调度至目标节点服务器。这样，资源调度系统可以将客户在各区域内的业务访问请求调度至客户在相应区域内带宽成本较低的节点服务器，由带宽成本较低的节点服务器为客户在各区域提供其业务所需的带宽资源，进而不仅可以有效降低CDN加速服务的带宽成本，还可以有效提高上述带宽成本较低的节点服务器的带宽资源的使用率。
附图说明
为了更清楚地说明本申请实施例中的技术方案，下面将对实施例描述中所需要使用的附图作简单地介绍，显而易见地，下面描述中的附图仅仅是本申请的一些实施例，对于本领域普通技术人员来讲，在不付出创造性劳动的前提下，还可以根据这些附图获得其它的附图。
图1是本申请实施例提供的一种资源调度系统的结构示意图；
图2是本申请实施例提供的一种资源调度方法的流程图；
图3是本申请实施例提供的一种资源调度方法的网络场景图；
图4是本申请实施例提供的一种资源调度方法的网络场景图；
图5是本申请实施例提供的一种网络设备的结构示意图。
具体实施例
为使本申请的目的、技术方案和优点更加清楚，下面将结合附图对本申请实施例作进一步地详细描述。
本申请实施例提供了一种资源调度方法，该方法可以应用于如图1所示的资源调度系统，该资源调度系统可以包括数据采集层、分析决策层、请求调度层。其中，上述数据采集层、分析决策层、请求调度层均可以是独立的服务器，也可以是同一服务器上的不同组件，还可以是由多台服务器构成的服务器集群。资源调度系统可以根据客户在各区域内对应的所有节点服务器的目标客户可用带宽、带宽成本以及客户在各区域内的实时带宽需求，将客户在各区域内的业务访问请求在相应区域内对应的节点服务器间进行调度。其中，数据采集层可以采集上述过程中需要的数据；分析决策层可以对数据采集层采集的数据进行分析处理；请求调度层可以根据分析决策层的分析处理结果进行相应的资源调度。
下面将结合具体实施例，对图2所示的一种资源调度方法的处理流程进行详细的说明，内容可以如下：
步骤201：资源调度系统周期性采集目标客户在目标区域对应的所有节点服务器的目标客户可用带宽，和目标客户在目标区域内的实时带宽需求。
在实施中，CDN服务商通常可以通过部署在各个区域的节点服务器为客 户提供网络加速服务，并可以存储客户信息(如客户名称、客户在各地区对应的节点服务器等)、节点服务器的计费信息(如计费方式、计费系数等)及各节点服务器的额定带宽等配置数据。考虑到各节点服务器的带宽资源在不同时间可能会出现不同程度的空置现象，CDN服务商的技术人员可以通过资源调度系统对各区域内的客户的业务访问请求在节点服务器间进行调度。具体的，资源调度系统可以周期性获取上述客户信息、节点服务器的计费信息及各节点服务器的额定带宽等配置数据，并采集各个客户(可称为目标客户)在各个区域(可称为目标区域)的实时带宽数据以及各个节点服务器的实时带宽承载值等实时数据。然后，资源调度系统可以对上述配置数据和实时数据进行分析处理，得到目标客户目标区域对应的所有节点服务器的目标客户可用带宽，和目标客户在目标区域内的对应的所有节点服务器上产生的总实时带宽数据，将该总实时带宽数据确定为目标客户在目标区域内的实时带宽需求。其中，目标客户可用带宽由节点服务器的剩余带宽和节点服务器上目标客户的实时占用带宽组成，例如，节点服务器的剩余带宽为500Mb/s，且节点服务器上目标客户的实时占用带宽为200Mb/s，则目标客户可用带宽为700Mb/s。此处，资源调度系统还可以将上述分析处理后的实时数据存储在大数据平台，以为CDN服务商进行其他数据分析时提供数据依据。之后，资源调度系统可以根据上述所有节点服务器的目标客户可用带宽和目标客户在目标区域内的实时带宽需求，对目标客户在目标区域内的业务访问请求进行调度。
步骤202：资源调度系统根据实时带宽需求、所有节点服务器的带宽成本和目标客户可用带宽，在所有节点服务器中确定目标节点服务器。
在实施中，资源调度系统可以根据各节点服务器的计费方式和计费系数等计费信息确定各节点服务器的带宽成本，并获取目标客户在目标区域内对应的所有节点服务器的带宽成本。然后，资源调度系统可以根据上述所有节点服务器的带宽成本、目标客户可用带宽及目标客户在目标区域的实时带宽需求，在目标客户在目标区域内对应的所有节点服务器中确定出用于承载目标客户的带宽的节点服务器(可称为目标节点服务器)。
可选的，资源调度系统可以优先选择带宽成本较低的节点服务器来承载客户带宽需求，相应的，上述步骤202的处理可以如下：资源调度系统按照带宽成本从低到高的顺序，在所有节点服务器中依次选取带宽承载节点；当选取 到的所有带宽承载节点的目标客户可用带宽之和大于实时带宽需求时，资源调度系统将所有带宽承载节点确定为目标节点服务器。
在实施中，资源调度系统在采集到目标客户在目标区域对应的所有节点服务器的目标客户可用带宽，和目标客户在目标区域内的实时带宽需求之后，可以将上述所有节点服务器按照带宽成本进行降序排列，并按照带宽成本从低到高的顺序依次在目标客户在目标区域对应的节点服务器中选取带宽承载节点，并确定带宽承载节点的目标客户可用带宽之和是否大于目标客户在目标区域内的实时带宽需求。当带宽承载节点的目标客户可用带宽之和大于目标客户在目标区域内的实时带宽需求时，资源调度系统可以确定上述带宽承载节点为目标节点服务器。值得一提的是，在按照带宽成本选取带宽承载节点时，如果多台未被选取的节点服务器的带宽成本相同且在所有未被选取的节点服务器中带宽成本最低，则可以将该多台节点服务器同时选取为带宽承载节点。
具体来讲，由于CDN服务商通常根据各节点服务器的带宽使用情况及相应的计费方式(如95计费、峰值计费、带宽买断计费和按实际流量计费等计费方式)向运营商支付费用，所以对于CDN服务商来说计费系数越高，则带宽成本越高，且按实际流量计费的节点服务器的带宽成本要高于95计费、峰值计费、带宽买断计费等计费方式的节点服务器，而95计费、峰值计费、带宽买断计费的节点服务器均不按照实际流量计费，其带宽成本可以认为相等。如图3所示，以目标客户在目标区域内对应的所有节点服务器为节点1：计费系数8.5、按实际流量计费，节点2：计费系数4.8、按实际流量计费，节点3：95计费，节点4：带宽买断计费，目标客户在目标区域的实时带宽需求为2000Mb/s为例，假设节点1的目标客户可用带宽为1000Mb/s，节点2的目标客户可用带宽为800Mb/s，节点3的目标客户可用带宽为600Mb/s，节点4的目标客户可用带宽为900Mb/s，资源调度系统可以将上述节点按照带宽成本进行降序排列如下：
节点1：计费系数8.5、目标客户可用带宽为1000Mb/s、按实际流量计费；
节点2：计费系数4.8、目标客户可用带宽为800Mb/s、按实际流量计费；
节点3：目标客户可用带宽为600Mb/s、95计费；
节点4：目标客户可用带宽为900Mb/s、带宽买断计费。
资源调度系统可以先将带宽成本最低的节点服务器即节点3和节点4选取为带宽承载节点，由于节点3和节点4的目标客户可用带宽之和为1500Mb/s， 小于目标客户在目标区域的实时带宽需求2000Mb/s，资源调度系统则可以进一步将节点2选取为带宽承载节点，此时，节点2、节点3和节点4的目标客户可用带宽之和为2300Mb/s，大于目标客户在目标区域的实时带宽需求2000Mb/s，资源调度系统可以确定上述节点2、节点3和节点4为目标节点服务器，即由节点2、节点3和节点4承载目标客户在目标区域内的业务访问请求。
考虑到若将目标客户在目标区域的所有业务访问请求调度到少数几个带宽承载节点上可能会影响客户的服务质量，或出现带宽承载节点负载过大的情况，CDN服务商的技术人员可以通过资源调度系统对目标客户在目标区域的带宽承载节点的最少个数(可称为最少保留个数)进行设置。仍以上述目标客户在目标区域内对应的所有节点服务器为节点1：计费系数8.5、按实际流量计费，节点2：计费系数4.8、按实际流量计费，节点3：95计费，节点4：带宽买断计费，目标客户在目标区域的实时带宽需求为1000Mb/s为例，假设节点1的目标客户可用带宽为1000Mb/s，节点2的目标客户可用带宽为800Mb/s，节点3的目标客户可用带宽为600Mb/s，节点4的目标客户可用带宽为900Mb/s，最少保留个数为3个，资源调度系统可以先将带宽成本最低的3个节点服务器即节点3和节点4选取为带宽承载节点。此时，节点2、节点3和节点4可以承载目标客户在目标区域的实时带宽需求，资源调度系统可以确定节点2、节点3和节点4为目标节点服务器。
步骤203：资源调度系统将目标客户在目标区域内的业务访问请求，调度至目标节点服务器。
在实施中，资源调度系统在确定目标节点服务器之后，可以将目标客户在目标区域内所有的业务访问请求，调度至所有目标节点服务器上。仍以目标客户在目标区域内对应的所有节点服务器为节点1、节点2、节点3、节点4，目标节点服务器为节点1、节点2为例，资源调度系统可以只允许节点3、节点4共同提供上述目标客户在目标区域的实时带宽需求所需的带宽资源，如图4所示，即资源调度系统可以在负载均衡设备上将目标客户在目标区域的带宽承载节点设置为节点3和节点4，从而可以使得负载均衡设备将目标客户在目标区域的业务访问请求分发至上述节点3、节点4上。
可选的，当某个节点服务器承载的带宽过高时，资源调度系统可以将暂时停止将业务访问请求调度至该节点服务器，相应的处理可以如下：当检测到 目标客户在目标区域内的第一节点服务器的节点可用带宽小于预设带宽预警值时，资源调度系统将目标客户目标区域内的业务访问请求，调度至所有节点服务器中除第一节点服务器之外的其余节点服务器。
在实施中，考虑到节点服务器的实时带宽承载值太高可能增加带宽成本，CDN服务商的技术人员可以通过资源调度系统对各节点服务器的节点可用带宽设置预警值(可称为预设带宽预警值)。当检测到有节点可用带宽小于预设带宽预警值的节点服务器(可称为第一节点服务器)时，资源调度系统可以暂时停止将目标客户在目标区域的业务访问请求调度至第一节点服务器上，将目标客户在目标区域的业务访问请求均调度至目标客户在目标区域中对应的除第一节点服务器之外的其余节点服务器。可以理解，计费方式为按实际流量计费的节点服务器的带宽成本随实际流量的增加而增加，故而可以不对该类型的节点服务器设置上述预设带宽预警值。
可选的，资源调度系统可以根据除目标节点服务器的节点可用带宽确定是否添加带宽承载服务器，相应的，上述步骤203之后可以进行如下处理：当检测到目标客户在目标区域内的带宽承载服务器中存在节点可用带宽小于预设带宽预警值的节点服务器时，资源调度系统则在所有节点服务器除目标节点服务器之外的其余节点服务器中选取带宽成本最低的待添加服务器；资源调度系统将目标客户在目标区域内的业务访问请求调度至目标节点服务器和待添加服务器。
在实施中，在对目标客户在目标区域内的业务访问请求进行调度之后，资源调度系统可以检测目标客户在目标区域内的带宽承载服务器的节点可用带宽。当检测到带宽承载服务器中存在节点可用带宽小于预设带宽预警值的节点服务器时，资源调度系统可以在目标客户在目标区域对应的除目标节点服务器之外的其余节点服务器中，选取带宽成本最低的待添加服务器，然后将目标客户在目标区域内的业务访问请求调度至目标节点服务器和待添加服务器，即由目标节点服务器和待添加服务器共同作为目标客户在目标区域的带宽承载服务器。值得一提的是，选取出的待添加服务器可以是带宽成本最低的一台节点服务器，也可以是带宽成本相等且最低的多台节点服务器。进一步的，如果在选取了一次待添加服务器后，再次检测到带宽承载服务器中存在节点可用带宽小于预设带宽预警值的节点服务器，则可以将继续在其余节点服务器中再次选取 待添加服务器作为带宽承载服务器。
可选的，资源调度系统可以根据目标客户在目标区域内的实时带宽需求的变化率确定是否恢复调度，相应的，上述步骤203之后可以进行如下处理：当检测到目标客户在目标区域内的实时带宽需求的增长率大于预设增幅阈值，资源调度系统则将目标客户在目标区域内的业务访问请求调度至目标客户在目标区域对应的所有节点服务器。
在实施中，资源调度系统可以按照预设检测周期检测并记录目标客户在目标区域对应的除目标节点服务器之外的其余节点服务器的实时带宽承载值。从而，资源调度系统可以获取到上一预设检测周期目标客户在目标区域的实时带宽需求和当前时刻目标客户在目标区域的实时带宽需求，并计算目标客户在目标区域的实时带宽需求的增长率，计算方式可以如下：
Figure PCTCN2019090318-appb-000001
考虑到可能出现由于目标客户在目标区域的实时带宽承载值的增长率过大而发生频繁调度的情况，CDN服务商的技术人员可以通过资源调度系统为目标客户在目标区域设置相邻预设检测周期的实时带宽承载值的增长率最高值(可称为预设增幅阈值)。当资源调度系统检测到上述增长率大于预设增幅阈值时，资源调度系统可以将目标客户在目标区域内的业务访问请求调度至目标客户在目标区域对应的所有节点服务器。
可选的，资源调度系统可以对同一客户在同一区域内带宽承载服务器两次变更间的最短时间间隔进行设定，相应的，上述步骤202的处理可以如下：资源调度系统周期性获取目标客户在目标区域内的带宽承载服务器的最近变更时刻；资源调度系统确定最近变更时刻与当前时刻的间隔时长是否大于预设间隔时长；如果是，资源调度系统则根据实时带宽需求、所有节点服务器的带宽成本和目标客户可用带宽，在所有节点服务器中确定目标节点服务器。
在实施中，为防止目标客户在目标区域的带宽承载服务器频繁变更的现象，CDN服务商的技术人员可以通过资源调度系统记录目标客户在目标区域内的带宽承载服务器的变更时刻，并为目标客户在目标区域内带宽承载服务器相邻的两次变更设置最小间隔时长(可称为预设间隔时长)。从而，资源调度系统可以获取目标客户在目标区域内的带宽承载服务器的最近变更时刻，计算最近 变更时刻与当前时刻的间隔时长，并判断该间隔时长是否大于预设间隔时长。如果是，资源调度系统则可以根据实时带宽需求、所有节点服务器的带宽成本和目标客户可用带宽，在所有节点服务器中确定目标节点服务器；否则，资源调度系统可以不重新确定目标节点服务器。可以理解，上述预设间隔时长必然大于步骤201中的周期时长，故而，采用上述方法，可以有效防止在增添带宽承载服务器后，短时间内又将带宽承载服务器剔除的现象，有效避免了因反复调度带来的资源浪费。
可选的，资源调度系统可以仅在带宽需求低谷期执行带宽承载服务器的删减处理，相应的，上述步骤202的处理可以如下：资源调度系统如果根据实时带宽需求确定当前时刻属于带宽需求低谷期，则根据实时带宽需求、所有节点服务器的带宽成本和目标客户可用带宽，在所有节点服务器中确定目标节点服务器。
在实施中，资源调度系统在采集到目标客户在目标区域内的实时带宽需求之后，可以先根据该实时带宽需求判断当前时刻是否属于带宽需求低谷期。如果当前时刻属于带宽需求低谷期，资源调度系统则可以根据实时带宽需求、所有节点服务器的带宽成本和目标客户可用带宽，在所有节点服务器中确定目标节点服务器，并将目标客户的业务访问请求调度至目标客户在目标区域对应的所有节点服务器。如果当前时刻不属于带宽需求低谷期，则可以保持目标客户在目标区域内的带宽承载服务器不变，或者还原之前剔除的带宽承载服务器。需要说明的是，CDN服务商的技术人员也可以根据目标客户的实际带宽需求情况通过资源调度系统的UI界面输入目标客户在目标区域的资源调度恢复指令，然后，资源调度系统也可以将目标客户的业务访问请求调度至资源调度恢复指令所指向的所有节点服务器。
可选的，资源调度系统可以在非带宽需求低谷期执行带宽承载服务器的添加处理，相应的处理可以如下：资源调度系统如果根据实时带宽需求确定当前时刻不属于带宽需求低谷期，则将目标客户在目标区域对应的目标节点服务器之外的其它节点服务器逐步添加为目标客户在目标区域内的带宽承载服务器。
在实施中，资源调度系统在采集到目标客户在目标区域内的实时带宽需求之后，如果判断出当前时刻不属于带宽需求低谷期，则可以还原之前剔除的目标客户在目标区域内的带宽承载服务器，即资源调度系统可以将目标客户在 目标区域对应的目标节点服务器之外的其它节点服务器，逐步添加为目标客户在目标区域内的带宽承载服务器。值得一提的是，此处的逐步可以是以固定的时间间隔添加指定个数的带宽承载服务器。
在本实施例中，资源调度系统周期性采集目标客户在目标区域对应的所有节点服务器的目标客户可用带宽，和目标客户在目标区域内的实时带宽需求；资源调度系统根据实时带宽需求、所有节点服务器的带宽成本和目标客户可用带宽，在所有节点服务器中确定目标节点服务器；资源调度系统将目标客户在目标区域内的业务访问请求，调度至目标节点服务器。这样，资源调度系统可以将客户在各区域内的业务访问请求调度至客户在相应区域内带宽成本较低的节点服务器，由带宽成本较低的节点服务器为客户在各区域提供其业务所需的带宽资源，进而不仅可以有效降低CDN加速服务的带宽成本，还可以有效提高上述带宽成本较低的节点服务器的带宽资源的使用率。
基于相同的技术构思，本申请实施例还提供了一种资源调度系统，如图1所示，所述系统包括数据采集层101、分析决策层102和请求调度层103，其中：
所述数据采集层101，用于周期性采集目标客户在目标区域对应的所有节点服务器的目标客户可用带宽，和所述目标客户在所述目标区域内的实时带宽需求；
所述分析决策层102，用于根据所述实时带宽需求、所述所有节点服务器的带宽成本和所述目标客户可用带宽，在所述所有节点服务器中确定目标节点服务器；
所述请求调度层103，用于将所述目标客户在所述目标区域内的业务访问请求，调度至所述目标节点服务器。
可选的，所述分析决策层102，具体用于：
按照所述带宽成本从低到高的顺序，在所述所有节点服务器中依次选取带宽承载节点；
当选取到的所有带宽承载节点的目标客户可用带宽之和大于所述实时带宽需求时，将所有所述带宽承载节点确定为目标节点服务器。
可选的，所述请求调度层103，还用于：
当检测到所述目标客户在所述目标区域内的第一节点服务器的节点可用带 宽小于预设带宽预警值时，将所述目标客户所述目标区域内的业务访问请求，调度至所述所有节点服务器中除所述第一节点服务器之外的其余节点服务器。
可选的，所述请求调度层103，还用于：
当检测到所述目标客户在所述目标区域内的带宽承载服务器中存在节点可用带宽小于预设带宽预警值的节点服务器时，则在所述所有节点服务器除所述目标节点服务器之外的其余节点服务器中选取带宽成本最低的待添加服务器；
将所述目标客户在所述目标区域内的业务访问请求调度至所述目标节点服务器和所述待添加服务器。
可选的，所述请求调度层103，还用于：
当检测到所述目标客户在所述目标区域内的实时带宽需求的增长率大于预设增幅阈值，将所述目标客户在所述目标区域内的业务访问请求调度至所述目标客户在所述目标区域对应的所有节点服务器。
可选的，所述分析决策层102，具体用于：
周期性获取所述目标客户在所述目标区域内的带宽承载服务器的最近变更时刻；
确定所述最近变更时刻与当前时刻的间隔时长是否大于预设间隔时长；
如果是，则根据所述实时带宽需求、所述所有节点服务器的带宽成本和所述目标客户可用带宽，在所述所有节点服务器中确定目标节点服务器。
可选的，所述分析决策层102，具体用于：
如果根据所述实时带宽需求确定当前时刻属于带宽需求低谷期，则根据所述实时带宽需求、所述所有节点服务器的带宽成本和所述目标客户可用带宽，在所述所有节点服务器中确定目标节点服务器。
可选的，所述请求调度层103，还用于：
所述资源调度系统如果根据所述实时带宽需求确定当前时刻不属于带宽需求低谷期，则将所述目标客户在所述目标区域对应的所述目标节点服务器之外的其它节点服务器逐步添加为所述目标客户在所述目标区域内的带宽承载服务器。
在本实施例中，资源调度系统周期性采集目标客户在目标区域对应的所有节点服务器的目标客户可用带宽，和目标客户在目标区域内的实时带宽需求；资源调度系统根据实时带宽需求、所有节点服务器的带宽成本和目标客户可用 带宽，在所有节点服务器中确定目标节点服务器；资源调度系统将目标客户在目标区域内的业务访问请求，调度至目标节点服务器。这样，资源调度系统可以将客户在各区域内的业务访问请求调度至客户在相应区域内带宽成本较低的节点服务器，由带宽成本较低的节点服务器为客户在各区域提供其业务所需的带宽资源，进而不仅可以有效降低CDN加速服务的带宽成本，还可以有效提高上述带宽成本较低的节点服务器的带宽资源的使用率。
本申请实施例还提供了一种网络设备，如图5所示，包括：
存储器501，用于存储程序指令:
处理器502，用于调用所述存储器中存储的程序指令，按照获得的程序执行前述实施例中的步骤201-203所述的资源调度方法。
本申请实施例还提供了一种计算机存储介质，计算机存储介质存储有计算机可执行指令，所述计算机可执行指令用于使所述计算机执行前述实施例中的网络设备所执行的方法步骤。
本领域普通技术人员可以理解实现上述实施例的全部或部分步骤可以通过硬件来完成，也可以通过程序来指令相关的硬件完成，所述的程序可以存储于一种计算机可读存储介质中，上述提到的存储介质可以是只读存储器，磁盘或光盘等。
以上所述仅为本申请的较佳实施例，并不用以限制本申请，凡在本申请的精神和原则之内，所作的任何修改、等同替换、改进等，均应包含在本申请的保护范围之内。

Claims (14)

  1. 一种资源调度方法，包括：
    资源调度系统周期性采集目标客户在目标区域对应的所有节点服务器的目标客户可用带宽，和所述目标客户在所述目标区域内的实时带宽需求；
    所述资源调度系统根据所述实时带宽需求、所述所有节点服务器的带宽成本和所述目标客户可用带宽，在所述所有节点服务器中确定目标节点服务器；
    所述资源调度系统将所述目标客户在所述目标区域内的业务访问请求，调度至所述目标节点服务器。
  2. 根据权利要求1所述的方法，其中，所述资源调度系统根据所述实时带宽需求、所述所有节点服务器的带宽成本和所述目标客户可用带宽，在所述所有节点服务器中确定目标节点服务器，包括：
    所述资源调度系统按照所述带宽成本从低到高的顺序，在所述所有节点服务器中依次选取带宽承载节点；
    当选取到的所有带宽承载节点的目标客户可用带宽之和大于所述实时带宽需求时，所述资源调度系统将所述所有带宽承载节点确定为目标节点服务器。
  3. 根据权利要求1所述的方法，其中，所述方法还包括：
    当检测到所述目标客户在所述目标区域内的第一节点服务器的节点可用带宽小于预设带宽预警值时，所述资源调度系统将所述目标客户所述目标区域内的业务访问请求，调度至所述所有节点服务器中除所述第一节点服务器之外的其余节点服务器。
  4. 根据权利要求1所述的方法，其中，所述资源调度系统将所述目标客户在所述目标区域内的业务访问请求，调度至所述目标节点服务器之后，还包括：
    当检测到所述目标客户在所述目标区域内的带宽承载服务器中存在节点可用带宽小于预设带宽预警值的节点服务器时，所述资源调度系统则在所述所有节点服务器除所述目标节点服务器之外的其余节点服务器中选取带宽成本最低的待添加服务器；
    所述资源调度系统将所述目标客户在所述目标区域内的业务访问请求调度至所述目标节点服务器和所述待添加服务器。
  5. 根据权利要求1所述的方法，其中，所述资源调度系统将所述目标客户在所述目标区域内的业务访问请求，调度至所述目标节点服务器之后，还包括：
    当检测到所述目标客户在所述目标区域内的实时带宽需求的增长率大于预设增幅阈值，所述资源调度系统则将所述目标客户在所述目标区域内的业务访问请求调度至所述目标客户在所述目标区域对应的所有节点服务器。
  6. 根据权利要求3-5任一项所述的方法，其中，所述资源调度系统根据所述实时带宽需求、所述所有节点服务器的带宽成本和所述目标客户可用带宽，在所述所有节点服务器中确定目标节点服务器，包括：
    所述资源调度系统周期性获取所述目标客户在所述目标区域内的带宽承载服务器的最近变更时刻；
    所述资源调度系统确定所述最近变更时刻与当前时刻的间隔时长是否大于预设间隔时长；
    如果是，所述资源调度系统则根据所述实时带宽需求、所述所有节点服务器的带宽成本和所述目标客户可用带宽，在所述所有节点服务器中确定目标节点服务器。
  7. 根据权利要求1所述的方法，其中，所述资源调度系统根据所述实时带宽需求、所述所有节点服务器的带宽成本和所述目标客户可用带宽，在所述所有节点服务器中确定目标节点服务器，包括：
    所述资源调度系统如果根据所述实时带宽需求确定当前时刻属于带宽需求低谷期，则根据所述实时带宽需求、所述所有节点服务器的带宽成本和所述目标客户可用带宽，在所述所有节点服务器中确定目标节点服务器。
  8. 根据权利要求7所述的方法，其中，所述方法还包括：
    所述资源调度系统如果根据所述实时带宽需求确定当前时刻不属于带宽需求低谷期，则将所述目标客户在所述目标区域对应的所述目标节点服务器之外 的其它节点服务器逐步添加为所述目标客户在所述目标区域内的带宽承载服务器。
  9. 一种资源调度系统，包括数据采集层、分析决策层和请求调度层，其中：
    所述数据采集层，用于周期性采集目标客户在目标区域对应的所有节点服务器的目标客户可用带宽，和所述目标客户在所述目标区域内的实时带宽需求；
    所述分析决策层，用于根据所述实时带宽需求、所述所有节点服务器的带宽成本和所述目标客户可用带宽，在所述所有节点服务器中确定目标节点服务器；
    所述请求调度层，用于将所述目标客户在所述目标区域内的业务访问请求，调度至所述目标节点服务器。
  10. 根据权利要求9所述的系统，其中，所述分析决策层，具体用于：
    按照所述带宽成本从低到高的顺序，在所述所有节点服务器中依次选取带宽承载节点；
    当选取到的所有带宽承载节点的目标客户可用带宽之和大于所述实时带宽需求时，将所有所述带宽承载节点确定为目标节点服务器。
  11. 根据权利要求9所述的系统，其中，所述请求调度层，还用于：
    当检测到所述目标客户在所述目标区域内的第一节点服务器的节点可用带宽小于预设带宽预警值时，将所述目标客户所述目标区域内的业务访问请求，调度至所述所有节点服务器中除所述第一节点服务器之外的其余节点服务器。
  12. 根据权利要求9所述的系统，其中，所述请求调度层，还用于：
    当检测到所述目标客户在所述目标区域内的带宽承载服务器中存在节点可用带宽小于预设带宽预警值的节点服务器时，则在所述所有节点服务器除所述目标节点服务器之外的其余节点服务器中选取带宽成本最低的待添加服务器；
    将所述目标客户在所述目标区域内的业务访问请求调度至所述目标节点服务器和所述待添加服务器。
  13. 一种网络设备，包括：
    至少一个处理器；以及，
    与所述至少一个处理器通信连接的存储器；其中，
    所述存储器存储有可被所述至少一个处理器执行的指令，所述指令被所述至少一个处理器执行，以使所述至少一个处理器能够执行如权利要求1至8中任一所述的资源调度方法。
  14. 一种计算机可读存储介质，存储有计算机程序，其中，所述计算机程序被处理器执行时实现权利要求1至8中任一项所述的资源调度方法。
PCT/CN2019/090318 2019-05-05 2019-06-06 一种资源调度方法及系统 Ceased WO2020224022A1 (zh)

Priority Applications (2)

Application Number Priority Date Filing Date Title
EP19874760.2A EP3754944B1 (en) 2019-05-05 2019-06-06 Resource scheduling method and system
US16/867,433 US11153370B2 (en) 2019-05-05 2020-05-05 Resource scheduling method and system

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201910368450.9A CN110198344A (zh) 2019-05-05 2019-05-05 一种资源调度方法及系统
CN201910368450.9 2019-05-05

Related Child Applications (1)

Application Number Title Priority Date Filing Date
US16/867,433 Continuation US11153370B2 (en) 2019-05-05 2020-05-05 Resource scheduling method and system

Publications (1)

Publication Number Publication Date
WO2020224022A1 true WO2020224022A1 (zh) 2020-11-12

Family

ID=67752397

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2019/090318 Ceased WO2020224022A1 (zh) 2019-05-05 2019-06-06 一种资源调度方法及系统

Country Status (4)

Country Link
US (1) US11153370B2 (zh)
EP (1) EP3754944B1 (zh)
CN (1) CN110198344A (zh)
WO (1) WO2020224022A1 (zh)

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113391985A (zh) * 2021-06-09 2021-09-14 北京猿力未来科技有限公司 资源分配方法及装置
CN114500285A (zh) * 2021-12-28 2022-05-13 天翼云科技有限公司 一种带宽调度方法及装置
CN115129750A (zh) * 2021-03-24 2022-09-30 腾讯云计算(长沙)有限责任公司 一种数据处理方法和相关装置

Families Citing this family (25)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110830604B (zh) * 2019-10-28 2021-09-10 腾讯科技（深圳）有限公司 Dns调度方法、装置
CN110830564B (zh) * 2019-10-30 2022-11-01 北京金山云网络技术有限公司 Cdn调度方法、装置、系统及计算机可读存储介质
CN110896365B (zh) * 2019-12-20 2022-09-02 网宿科技股份有限公司 网络节点内的流量调度方法、服务器及存储介质
CN111901425B (zh) * 2020-07-28 2021-05-28 平安科技（深圳）有限公司 基于Pareto算法的CDN调度方法、装置、计算机设备及存储介质
CN112243025B (zh) * 2020-09-22 2023-10-17 网宿科技股份有限公司 节点成本的调度方法、电子设备及存储介质
CN114844949B (zh) * 2021-02-02 2023-10-27 北京金山云网络技术有限公司 业务调度方法、装置、设备及存储介质
CN112866985B (zh) * 2021-02-20 2023-06-23 百度在线网络技术（北京）有限公司 流量控制方法、资源下载方法、装置、设备和存储介质
CN115134369B (zh) * 2021-03-26 2024-01-19 北京金山云网络技术有限公司 一种cdn节点分配方法、装置、电子设备及存储介质
CN113840330B (zh) * 2021-09-16 2023-10-13 阿里巴巴达摩院(杭州)科技有限公司 建立连接的方法、网关设备、网络系统及调度中心
CN114338705B (zh) * 2021-11-24 2023-12-01 阿里巴巴（中国）有限公司 内容分发网络cdn节点的资源水位控制方法、设备及介质
CN114071770B (zh) * 2021-12-20 2023-09-05 广东新电电力科技有限公司 基于智能通信管理网关的通信控制方法和装置
CN114338531B (zh) * 2021-12-22 2024-04-05 广东明创软件科技有限公司 一种流量控制方法、装置及存储介质
CN114430374B (zh) * 2021-12-22 2024-09-27 天翼云科技有限公司 一种网络资源分配方法、装置及计算机设备
CN116708186B (zh) * 2022-02-25 2025-10-03 武汉金山云信息技术有限公司 带宽使用效率确定方法和装置、电子设备和存储介质
CN114726731B (zh) * 2022-02-28 2024-11-26 网宿科技股份有限公司 带宽成本的确定方法、电子设备及存储介质
CN114726732B (zh) * 2022-02-28 2024-10-22 网宿科技股份有限公司 Cdn带宽资源的规划方法、装置、电子设备及存储介质
CN114615097B (zh) * 2022-03-02 2023-09-19 北京云享智胜科技有限公司 客户付费带宽占比的确定方法、装置以及存储介质
CN115225507B (zh) * 2022-07-21 2024-03-08 天翼云科技有限公司 一种服务器组资源分配方法、装置、设备及介质
CN115460090B (zh) * 2022-08-31 2024-03-19 上海哔哩哔哩科技有限公司 边缘记录上传速度动态设置方法及装置
CN116032673B (zh) * 2022-12-22 2024-12-10 上海哔哩哔哩科技有限公司 访问调度方法及装置
CN116455756B (zh) * 2023-06-15 2023-08-29 深圳华锐分布式技术股份有限公司 基于行情推送及查询的带宽均衡方法、设备及介质
CN117112236B (zh) * 2023-10-23 2024-02-20 山东曙光照信息技术股份有限公司 基于数据涌流及波动性预测的辖区服务器配置方法及系统
CN118509326B (zh) * 2024-04-29 2025-11-07 天翼云科技有限公司 一种基于ip百分比调度的cdn客户带宽精准牵引方法
CN119788522B (zh) * 2024-11-15 2025-11-04 天翼云科技有限公司 一种带宽调度方法及装置
CN119603156B (zh) * 2024-11-28 2025-10-31 天翼云科技有限公司 带宽调度方法、装置、计算机设备、存储介质和程序产品

Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101127700A (zh) * 2007-08-16 2008-02-20 武汉理工大学 基于网络带宽和用户费用下的网格资源调度方法
US20150244769A1 (en) * 2014-02-27 2015-08-27 Verizon Patent And Licensing Inc. Content delivery system
CN107465708A (zh) * 2016-06-02 2017-12-12 腾讯科技（深圳）有限公司 一种cdn带宽调度系统及方法

Family Cites Families (15)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101272479B (zh) * 2008-03-31 2011-11-30 华为软件技术有限公司 资源调度装置及方法,点播节目的方法及系统和机顶盒
CN101346008B (zh) * 2008-08-26 2010-12-01 北京邮电大学 一种面向业务的光网络资源和应用资源的联合调度方法
US7822856B2 (en) * 2008-10-15 2010-10-26 Patentvc Ltd. Obtaining erasure-coded fragments using push and pull protocols
CN102780759B (zh) * 2012-06-13 2016-05-18 合肥工业大学 基于调度目标空间的云计算资源调度方法
CN104348647B (zh) * 2013-07-31 2019-04-12 腾讯科技（深圳）有限公司 多源带宽调度方法、装置及系统
CN104580012B (zh) * 2013-10-14 2018-08-07 深圳市腾讯计算机系统有限公司 一种内容分发网络节点的选择方法及装置
CN104902340A (zh) * 2014-03-04 2015-09-09 中辉世纪传媒发展有限公司 一种对等节点全局负载均衡媒体内容分发方法及系统
CN106034083B (zh) * 2015-03-12 2019-10-08 阿里巴巴集团控股有限公司 Cdn网络系统及其限速控制方法、cdn节点和服务器
CN107124375B (zh) * 2017-03-27 2020-02-18 网宿科技股份有限公司 Cdn网络带宽资源的错峰调度方法、系统以及服务器
CN109547517B (zh) * 2017-09-22 2021-11-09 贵州白山云科技股份有限公司 一种带宽资源调度方法和装置
CN108616401B (zh) * 2018-05-09 2020-07-28 华南理工大学 一种智能化的视频内容服务器部署方法及系统
CN108845874B (zh) * 2018-06-25 2023-03-21 腾讯科技（深圳）有限公司 资源的动态分配方法及服务器
CN109063978A (zh) * 2018-07-12 2018-12-21 南京邮电大学 能源互联网资源调度方法及系统、可读存储介质和终端
CN108897626A (zh) * 2018-07-20 2018-11-27 浪潮电子信息产业股份有限公司 一种资源调度方法及服务器
CN109067670B (zh) * 2018-09-28 2022-07-12 杭州领智云画科技有限公司 一种cdn调度方法及cdn调度设备

Patent Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101127700A (zh) * 2007-08-16 2008-02-20 武汉理工大学 基于网络带宽和用户费用下的网格资源调度方法
US20150244769A1 (en) * 2014-02-27 2015-08-27 Verizon Patent And Licensing Inc. Content delivery system
CN107465708A (zh) * 2016-06-02 2017-12-12 腾讯科技（深圳）有限公司 一种cdn带宽调度系统及方法

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
See also references of EP3754944A4 *

Cited By (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN115129750A (zh) * 2021-03-24 2022-09-30 腾讯云计算(长沙)有限责任公司 一种数据处理方法和相关装置
CN113391985A (zh) * 2021-06-09 2021-09-14 北京猿力未来科技有限公司 资源分配方法及装置
CN114500285A (zh) * 2021-12-28 2022-05-13 天翼云科技有限公司 一种带宽调度方法及装置
CN114500285B (zh) * 2021-12-28 2024-01-02 天翼云科技有限公司 一种带宽调度方法及装置

Also Published As

Publication number Publication date
EP3754944B1 (en) 2021-12-15
CN110198344A (zh) 2019-09-03
US20200382582A1 (en) 2020-12-03
EP3754944A4 (en) 2020-12-23
US11153370B2 (en) 2021-10-19
EP3754944A1 (en) 2020-12-23

Similar Documents

Publication Publication Date Title
WO2020224022A1 (zh) 一种资源调度方法及系统
US7174379B2 (en) Managing server resources for hosted applications
CN107707382B (zh) 一种智能调度方法和装置
US9497139B2 (en) Client-allocatable bandwidth pools
US20110078303A1 (en) Dynamic load balancing and scaling of allocated cloud resources in an enterprise network
CN109324900B (zh) 用于按需服务环境中的消息队列的基于竞价的资源共享
US9306870B1 (en) Emulating circuit switching in cloud networking environments
US20160321112A1 (en) Management system, virtual communication-function management node, and management method
JP6881575B2 (ja) 資源割当システム、管理装置、方法およびプログラム
US9971683B1 (en) Automatic computer memory management coordination across a group of servers
JP2004021982A (ja) コンピュータ・システム・リソースを動的に割り当てる方法およびシステム
US20190377596A1 (en) Flexible batch job scheduling in virtualization environments
CN105007287A (zh) 公共云资源动态配置方法及系统
CN103401938A (zh) 分布式云架构下基于业务特性的资源分配系统及其方法
WO2017107456A1 (zh) 确定任务消耗资源的方法及装置
CN104580012A (zh) 一种内容分发网络节点的选择方法及装置
JP5596716B2 (ja) リソース管理装置、リソース管理システム、リソース管理方法およびリソース管理プログラム
CN105119751B (zh) 一种基于环境实时感知的服务评估及选取方法
WO2024094104A1 (zh) 一种动态反馈加权云存储资源调度方法、装置及设备
CN116233027B (zh) Cdn资源调度方法、装置、电子设备及存储介质
US20240020168A1 (en) Resource Scheduling Method and Apparatus, Computer Device, System, and Storage Medium
CN107479947A (zh) 一种虚拟机能耗优化方法和系统
WO2016206441A1 (zh) 一种虚拟资源分配方法及装置、计算机存储介质
WO2020082518A1 (zh) 一种识别带宽需求突发的方法和装置
CN116032673A (zh) 访问调度方法及装置

Legal Events

Date Code Title Description
ENP Entry into the national phase

Ref document number: 2019874760

Country of ref document: EP

Effective date: 20200428

NENP Non-entry into the national phase

Ref country code: DE