WO2021017161A1 - 一种资源获取方法及边缘计算调度服务器 - Google Patents

一种资源获取方法及边缘计算调度服务器 Download PDF

Info

Publication number
WO2021017161A1
WO2021017161A1 PCT/CN2019/110151 CN2019110151W WO2021017161A1 WO 2021017161 A1 WO2021017161 A1 WO 2021017161A1 CN 2019110151 W CN2019110151 W CN 2019110151W WO 2021017161 A1 WO2021017161 A1 WO 2021017161A1
Authority
WO
WIPO (PCT)
Prior art keywords
address
domain name
edge
node
server
Prior art date
Application number
PCT/CN2019/110151
Other languages
English (en)
French (fr)
Inventor
王文庭
Original Assignee
网宿科技股份有限公司
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 网宿科技股份有限公司 filed Critical 网宿科技股份有限公司
Priority to US17/630,154 priority Critical patent/US20220286526A1/en
Priority to EP19939974.2A priority patent/EP4002796A4/en
Publication of WO2021017161A1 publication Critical patent/WO2021017161A1/zh

Links

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
    • H04L67/56—Provisioning of proxy services
    • H04L67/568—Storing data temporarily at an intermediate stage, e.g. caching
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L61/00—Network arrangements, protocols or services for addressing or naming
    • H04L61/45—Network directories; Name-to-address mapping
    • H04L61/4505—Network directories; Name-to-address mapping using standardised directories; using standardised directory access protocols
    • H04L61/4511—Network directories; Name-to-address mapping using standardised directories; using standardised directory access protocols using domain name system [DNS]
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L61/00—Network arrangements, protocols or services for addressing or naming
    • H04L61/58—Caching of addresses or names
    • 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
    • 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
    • H04L67/63—Routing a service request depending on the request content or context

Definitions

  • the present invention relates to the field of Internet technology, in particular to a resource acquisition method and an edge computing scheduling server.
  • the node server when the node server receives the resource acquisition request sent by the user client, it can query from the local cache whether the resource acquisition request points to the target resource, if it exists , The target resource can be provided to the user client. If it does not exist, you can obtain the target resource from the cache server by returning to the source, and provide the obtained target resource to the user client.
  • CDN Content Delivery Network
  • the purpose of this application is to provide a resource acquisition method and an edge computing scheduling server, which can improve the efficiency of resource acquisition.
  • one aspect of this application provides a resource acquisition method, the method includes: receiving a resource acquisition request sent by a terminal device, and determining whether the target resource pointed to by the resource acquisition request is stored in another terminal device; If the target resource is stored in another terminal device, the target terminal device is determined from the other terminal device, and the device identification of the target terminal device is fed back to the terminal device, so that the terminal device is Obtain the target resource from the target terminal device.
  • the edge computing scheduling server includes a processor and a memory.
  • the memory is used to store a computer program that is executed by the processor.
  • the terminal device After receiving the resource prefetch instruction issued by the equipment management center, the terminal device may send a resource acquisition request to the edge computing scheduling system in response to the resource prefetch instruction. After receiving the resource acquisition request, the edge computing scheduling system can determine whether the target resource pointed to by the resource acquisition request is stored in other terminal devices. If the target resource is stored in other terminal devices, the edge computing scheduling system can determine a target terminal device from the other terminal devices obtained by the query, and the edge computing scheduling system can feed back the device identification of the target terminal device to the terminal Device, so that the terminal device obtains the target resource from the target terminal device. In this way, multiple terminal devices can provide each other with the resources they need. On the one hand, the burden of the CDN is reduced, and on the other hand, the flexibility of resource acquisition is also improved, thereby improving the efficiency of resource acquisition.
  • FIG. 1 is a schematic diagram of the structure of a CDN system in an embodiment of the present invention
  • FIG. 2 is a schematic diagram of the steps of a node scheduling method in an embodiment of the present invention.
  • 3 is a schematic diagram of the IP address of the same POP in the embodiment of the present invention.
  • FIG. 4 is a schematic diagram of the structure of an edge computing scheduling server in an embodiment of the present invention.
  • FIG. 5 is a schematic diagram of a domain name resolution method in an embodiment of the present invention.
  • Fig. 6 is a schematic diagram of an application scenario of a resource acquisition method in an embodiment of the present invention.
  • This application provides a node scheduling method and a domain name resolution method.
  • the method can be applied to the system architecture shown in FIG. 1.
  • it may include multiple components such as equipment management center, terminal equipment, edge nodes, origin server, public network DNS server, edge computing scheduling system, and CDN scheduling system, wherein the edge node It can include a cache server, a node DNS server inside the node, and multiple node servers.
  • the aforementioned terminal device may be an electronic device sold by a manufacturer to a user, and the electronic device may have a network communication function and a resource storage function.
  • the terminal device may be a router with a data storage function.
  • the device management center may be a central server pre-deployed by a manufacturer, and the device management center may communicate with the terminal device, so that it may issue a resource prefetch instruction to the terminal device.
  • the terminal device may subsequently perform domain name resolution through a public network DNS server, so as to access a node server in a certain edge node in the CDN nearby.
  • the node server may obtain the corresponding resource from the cache server. If the cache server does not store the corresponding resource locally, it may download it from the origin server back to the source, and provide the downloaded resource to the node server.
  • the node scheduling method provided by the present application can be applied to the foregoing edge computing scheduling system, and the method can include the following steps.
  • S11 Receive a domain name resolution request from a public network DNS server, where the domain name resolution request includes an edge computing domain name to be resolved.
  • the terminal device after receiving the resource prefetch instruction of the hot resource issued by the device management center, the terminal device can construct an edge computing domain name for accessing the hot resource according to the resource identifier carried in the resource prefetch instruction.
  • the constructed edge computing domain name may include the top-level domain name of the website storing the hot resource and the resource identifier of the hot resource. In this way, the constructed edge computing domain name may be a URL (Uniform Resource Locator) carrying the resource identifier.
  • the terminal device can obtain the corresponding hotspot resources by accessing the aforementioned edge computing domain name.
  • the terminal device may send a domain name resolution request to a public network DNS server, and the domain name resolution request may carry the aforementioned edge computing domain name.
  • the public network DNS server can resolve the edge computing domain name into the corresponding IP address according to the pre-configured mapping relationship between the domain name and the IP address.
  • the edge computing scheduling system can be used to change the domain name resolution result of the public network DNS server. Therefore, after the public network DNS server receives the domain name resolution request sent by the terminal device, it can forward the domain name resolution request to the edge computing scheduling system, so that the edge computing scheduling system decides which edge computing domain name should be resolved to The IP address of the node server.
  • S13 Determine an accelerated domain name that has a mapping relationship with the edge computing domain name, and respectively determine an edge IP address corresponding to the edge computing domain name and a cache IP address corresponding to the accelerated domain name; wherein, the edge IP address and the The cached IP address belongs to the same access point in the current area.
  • the edge computing domain name for the terminal device to access the hotspot resource usually has an accelerated domain name that is a mutual mapping relationship in the CDN.
  • the mapping relationship between the edge computing domain name and the accelerated domain name may be pre-deployed in the CDN.
  • the edge computing domain name accessed by the terminal device is www.***.com
  • the accelerated domain name of the edge computing domain name in the CDN may be ***.wangsu.com.
  • the edge computing domain name can be used to determine the node server accessed by the terminal device, and the accelerated domain name can be used to determine the cache server that stores the hotspot resource.
  • the edge computing dispatch system after the edge computing dispatch system receives the domain name resolution request forwarded by the public network DNS server, in order to ensure that the subsequently determined node server and the cache server belong to the same POP, it can be queried according to the pre-configured mapping table An accelerated domain name that has a mapping relationship with the edge computing domain name.
  • the edge IP address of each node server in the CDN and the cache IP address of each cache server can be collected, and the POP information corresponding to each IP address can be summarized.
  • the CDN scheduling system collects the edge IP address, the cached IP address, and the POP information, it can provide these IP addresses and POP information to the edge computing scheduling system.
  • the edge computing scheduling system can determine the node server and cache server of the same POP based on the information provided by the CDN scheduling system. In this way, in the edge computing scheduling system, the edge IP address set of the edge computing domain name may be pre-configured, and one or more edge IP addresses may be included in the edge IP address set.
  • edge IP addresses can point to different node servers in the CDN. In this way, by resolving the edge computing domain name into one of the edge IP addresses, the terminal device can be dispatched to the node server pointed to by the edge IP address .
  • different edge IP addresses may have different POP information. Therefore, according to the POP information of each IP address provided by the CDN scheduling system, each edge IP address in the edge IP address set can be classified into the same category. The edge IP address can belong to the same POP. Refer to Figure 3, assuming that the edge IP address set includes 5 edge IP addresses, and these 5 edge IP addresses can be divided into two categories, IP1 to IP3 belong to POP1, and IP4 and IP5 belong to POP2.
  • the edge computing scheduling system there may also be a cache IP address set of the accelerated domain name, and each cache IP address in the cache IP address set may also be classified according to POP information.
  • the cache IP address set of the accelerated domain name includes three cache IP addresses, among which, IP6 belongs to POP1, and IP7 and IP8 belong to POP3.
  • the foregoing edge IP address set and cache IP address set are specific to a certain area. As the area expands or shrinks, the number of IP addresses contained in the edge IP address set and the cached IP address set may also increase or decrease. Specifically, the edge IP address set and the cached IP address set of the corresponding area may be selected according to the geographic location of the terminal device sending the domain name resolution request. For example, if the terminal device is located in Xiamen, the edge computing scheduling system will call the edge IP address set and the cached IP address set in the Xiamen area when a domain name resolution request is sent by the terminal device.
  • the edge IP address set and/or cache IP address set of the upper-level area to which the current area belongs can be selected. For example, for a domain name resolution request issued by a terminal device located in Xiamen, the edge IP address set and/or cached IP address set in Fujian or South China can be selected.
  • the edge computing scheduling system can respectively determine the edge IP address set corresponding to the edge computing domain name and the accelerated domain name in the current area The corresponding cache IP address set, and then, in order to ensure that the node server and the cache server belong to the same POP, the edge computing scheduling system may determine the edge IP address belonging to the same access network point from the edge IP address set and the cache IP address set And cache the IP address. For example, in FIG. 3, IP1 to IP3 belong to POP1, and IP6 also belongs to POP1. Therefore, IP1 to IP3 can be used as certain edge IP addresses, and IP6 can be used as certain cache IP addresses.
  • the edge IP address corresponding to the edge computing domain name and the cache IP address corresponding to the acceleration domain name can be determined respectively, and the edge IP address and the cache IP address belong to the current area. The same access point.
  • S15 Feed back the edge IP address to the public network DNS server to schedule the edge computing domain name to the node server pointed to by the edge IP address.
  • the edge computing scheduling system can feed back the edge IP address to the public network DNS server, so that the public network
  • the DNS server resolves the edge computing domain name of the terminal device to the received edge IP address. If there are multiple edge IP addresses received, you can choose one of them arbitrarily, or choose one of them according to the pre-configured load balancing policy.
  • the terminal device can access the corresponding node server according to the edge IP address fed back by the public network DNS server. In this way, the edge computing domain name accessed by the terminal device is scheduled to the node server pointed to by the edge IP address.
  • the node DNS server of the node can be deployed, and the node DNS server can be used to resolve the accelerated domain name accessed by the node server into the corresponding cache IP address.
  • the edge computing domain name accessed by the terminal device can be converted into the accelerated domain name with the mapping relationship in the edge node.
  • the node server can send a domain name resolution request including the accelerated domain name to the node DNS server of the local node.
  • the node DNS server After the node DNS server receives the domain name resolution request, it can also forward the domain name resolution request to the edge computing scheduling system to determine which cache IP address the accelerated domain name should be resolved to through the edge computing scheduling system.
  • the edge computing scheduling system Since the edge computing scheduling system has determined the edge IP address and the cache IP address belonging to the same POP in the current area, it can feed back and report to the node DNS server after receiving the domain name resolution request sent by the node DNS server
  • the edge IP address belongs to the cache IP address of the same access network point.
  • the node DNS server can resolve the accelerated domain name to the cache IP address, thereby dispatching the accelerated domain name to the cache server pointed to by the cache IP address. Subsequently, the node server can access the cache server pointed to by the cache IP address. Since the edge IP address of the node server and the cache IP address of the cache server belong to the same POP in the current area, there is no cross-POP communication problem, thereby avoiding the problems of extra charges and high delay caused by cross-POP communication.
  • the edge computing scheduling system can expand the coverage of the area, and from a larger coverage area Query whether there are edge IP addresses and cached IP addresses belonging to the same access network point. Specifically, if there are no edge IP addresses and cached IP addresses belonging to the same access network point in the current area, the edge computing scheduling system can determine the upper-level area to which the current area belongs, and determine in the upper-level area The edge IP address and cache IP address belonging to the same access network point.
  • the edge computing scheduling system can determine that they belong to the same access point in the area of Fujian or even the entire South China area.
  • the edge IP address and cache IP address of the entry point In this way, by continuously expanding the area coverage, it is possible to gradually find out the edge IP addresses and cached IP addresses belonging to the same access network point.
  • edge IP addresses and cache IP addresses belonging to the same access network point in the upper-level area then in order to ensure that the terminal device and the node server, as well as the node server and the cache server, Communication quality, it is not advisable to continue to expand the coverage of the area. Otherwise, although the edge IP address and the cached IP address obtained by the query belong to the same POP, they may be far apart due to the geographical location, resulting in the terminal device and the node server, as well as the node server and the cache server The quality of communication between them is poor.
  • the edge computing scheduling system can obtain the edge computing domain name in the current area. Pre-configured edge IP address set, and feed back the edge IP address set to the public network DNS server, so that the public network DNS server resolves the edge computing domain name into a target in the edge IP address set IP address. Subsequently, after the node DNS server in the edge node where the node server pointed to by the target IP address sends a domain name resolution request containing the accelerated domain name to the edge computing scheduling system, the edge computing scheduling system can obtain the accelerated domain name where the accelerated domain name is located.
  • the pre-configured cache IP address set in the current zone and feed back the cache IP address set to the node DNS server, so that the node DNS server resolves the accelerated domain name into a cache IP address in the cache IP address set .
  • the purpose of such processing is that in this scenario, the node server and the cache server are no longer required to belong to the same POP, but the network communication quality is the premise, so that the terminal device can enjoy better data transmission effects.
  • the node server and cache server in the CDN may fail or change, which may cause the edge IP address corresponding to the edge computing domain name and the cache IP address corresponding to the cache server to change.
  • POP information of each IP address may also be changed. Therefore, the CDN scheduling system can periodically collect IP addresses corresponding to edge computing domain names and accelerated domain names, and POP information of each IP address. Once the IP address or POP information changes from the previous version, the node information change notification can be sent to the edge computing scheduling system. The node information change notification may indicate the changed IP address and/or the POP information after the IP address change.
  • the edge computing scheduling system can update the locally stored edge IP address set and the IP address information and access network point information in the cached IP address set according to the node information change notification, so as to ensure subsequent processing of domain name resolution
  • the node server and cache server belonging to the same POP in the current area can be determined.
  • This application also provides an edge computing scheduling system, which includes:
  • the domain name resolution request receiving unit is configured to receive a domain name resolution request sent by a public network DNS server, where the domain name resolution request includes the edge computing domain name to be resolved;
  • the same access network point determining unit is used to determine the accelerated domain name that has a mapping relationship with the edge computing domain name, and respectively determine the edge IP address corresponding to the edge computing domain name and the cache IP address corresponding to the accelerated domain name; wherein, The edge IP address and the cached IP address belong to the same access network point in the current area;
  • An edge scheduling unit configured to feed back the edge IP address to the public network DNS server, so as to schedule the edge computing domain name to the node server pointed to by the edge IP address;
  • the cache scheduling unit is configured to feed back the edge IP address to the node DNS server after receiving the domain name resolution request including the accelerated domain name sent by the node DNS server in the edge node where the node server is located
  • the cache IP addresses belonging to the same access network point are used to schedule the accelerated domain name to the cache server pointed to by the cache IP address.
  • the same access network point determining unit includes:
  • the IP address set determining module is configured to determine the edge IP address set corresponding to the edge computing domain name and the cache IP address set corresponding to the acceleration domain name in the current area;
  • the same access network point screening module is configured to determine the edge IP address and the cache IP address belonging to the same access network point from the edge IP address set and the cache IP address set.
  • system further includes:
  • the upper-level area determining unit is configured to determine the upper-level area to which the current area belongs if there are no edge IP addresses and cached IP addresses belonging to the same access network point in the current area, and to determine that the current area belongs to Edge IP address and cache IP address of the same access point;
  • the pre-configuration unit is configured to, if there are no edge IP addresses and cached IP addresses belonging to the same access network point in the upper-level area, obtain a set of edge IP addresses pre-configured by the edge computing domain name in the current area, and The edge IP address set is fed back to the public network DNS server, so that the public network DNS server resolves the edge computing domain name into a target IP address in the edge IP address set.
  • the cache scheduling unit is further configured to obtain the accelerated domain name after receiving the domain name resolution request including the accelerated domain name sent by the node DNS server in the edge node pointed to by the target IP address Domain name is a set of cached IP addresses pre-configured in the current zone, and feeds back the set of cached IP addresses to the node DNS server, so that the node DNS server resolves the accelerated domain name into the set of cached IP addresses IP address in.
  • system further includes:
  • the information update unit is used to receive the node information change notification sent by the CDN scheduling system, and update the IP address information and access network points in the edge IP address set and the cached IP address set according to the node information change notice information.
  • this application also provides an edge computing scheduling server.
  • the edge computing scheduling server includes a processor and a memory.
  • the memory is used to store a computer program. When the computer program is executed by the processor, The node scheduling method described above.
  • This application also provides a domain name resolution method, which can be applied to a node DNS server in an edge node. Please refer to FIG. 5.
  • the method includes the following steps.
  • S21 Receive a domain name resolution request sent by a node server in the edge node, where the domain name resolution request includes an accelerated domain name.
  • S23 Obtain a cache IP address of a cache server that belongs to the same access network point as the node server in the current area from the edge computing scheduling system.
  • S25 Feed back the cache IP address to the node server, so that the node server obtains the resource corresponding to the acceleration domain name from the cache server pointed to by the cache IP address.
  • the node DNS server of the node can be deployed, and the node DNS server can be used to resolve the accelerated domain name accessed by the node server into the corresponding cache IP address.
  • the edge computing domain name accessed by the terminal device can be converted into the accelerated domain name with the mapping relationship in the edge node.
  • the node server can send a domain name resolution request including the accelerated domain name to the node DNS server of the local node.
  • the node DNS server may forward the domain name resolution request to the edge computing scheduling system to determine which cache IP address the accelerated domain name should be resolved to through the edge computing scheduling system.
  • the edge computing scheduling system Since the edge computing scheduling system has determined the edge IP address and the cache IP address belonging to the same POP in the current area, it can feed back and report to the node DNS server after receiving the domain name resolution request sent by the node DNS server
  • the node server belongs to the cache IP address of the cache server of the same access network point. In this way, the node DNS server can resolve the accelerated domain name to the cache IP address, thereby dispatching the accelerated domain name to the cache server pointed to by the cache IP address. Subsequently, the node server can access the cache server pointed to by the cache IP address. Because the edge IP address of the node server and the cache IP address of the cache server belong to the same POP in the current area, there is no cross-POP communication problem, thereby avoiding the problem of extra charges and high delay caused by cross-POP communication.
  • the method further includes:
  • the method further includes:
  • This application also provides a node DNS server, the node DNS server is located in an edge node, and the node DNS server includes:
  • a request receiving unit configured to receive a domain name resolution request sent by a node server in the edge node, where the domain name resolution request includes an accelerated domain name;
  • An IP address information obtaining unit configured to obtain a cache IP address of a cache server that belongs to the same access network point as the node server in the current area from the edge computing scheduling system;
  • the domain name resolution unit is configured to feed back the cache IP address to the node server, so that the node server obtains the resource corresponding to the accelerated domain name from the cache server pointed to by the cache IP address.
  • the IP address information obtaining unit is further configured to, if there is no node server and cache server belonging to the same access network point in the current area, obtain the information from the edge computing scheduling system and the node server The cache IP address of the cache server belonging to the same access network point in the upper-level area to which the server belongs;
  • the domain name resolution unit is further configured to obtain the node DNS from the edge computing scheduling system if there is no node server and cache server belonging to the same access network point in the upper-level area
  • the server configures a set of cache IP addresses in advance, and feeds back the set of cache IP addresses to the node server, so that the node server obtains all information from the cache server pointed to by any IP address in the set of cache IP addresses Describe the resources corresponding to the accelerated domain name.
  • the present application also provides a node DNS server.
  • the node DNS server includes a processor and a memory.
  • the memory is used to store a computer program.
  • the computer program is executed by the processor, the above-mentioned domain name resolution method is implemented.
  • This application also provides a resource acquisition method. Please refer to FIG. 6.
  • the method can be applied to the system shown in FIG. 6. Specifically, the method may include the following steps.
  • S31 Receive a resource acquisition request sent by a terminal device, and determine whether the target resource pointed to by the resource acquisition request is stored in another terminal device.
  • the terminal device may respond to the resource prefetch instruction and send a resource acquisition request to the edge computing scheduling system.
  • the edge computing scheduling system can determine whether the target resource pointed to by the resource acquisition request is stored in other terminal devices.
  • mapping relationship table between resources and terminal devices can be maintained.
  • the resource identifier and the terminal device storing the resource can be associated and stored. In this way, through the resource identification, it is possible to quickly query whether there is an associated terminal device.
  • the edge computing scheduling system can extract the resource identifier of the target resource in the resource acquisition request, and query the mapping table for whether there are other terminal devices associated with the resource identifier of the target resource, thereby It is determined whether the target resource pointed to by the resource acquisition request is stored in other terminal devices.
  • the target terminal device is determined from the other terminal device, and the device identifier of the target terminal device is fed back to the terminal device, so that the terminal device is Acquiring the target resource from the target terminal device.
  • the edge computing scheduling system can determine a target terminal device from the other terminal devices obtained by the query.
  • the target terminal device can be determined in various ways. For example, one terminal device may be randomly selected from other terminal devices obtained by the query as the target terminal device. It is also possible to use a terminal device with a lower current load as the target terminal device according to a load balancing algorithm.
  • the load of the terminal device may refer to the number of P2P (Point to Point) communication links currently established by the terminal device. The higher the number, the higher the load.
  • P2P Point to Point
  • the edge computing scheduling system may feed back the device identification of the target terminal device to the terminal device, so that the terminal device obtains all information from the target terminal device.
  • the target resource may refer to the network communication address of the target terminal device, or may refer to a character string used to characterize the identity of the target terminal device that complies with the P2P communication protocol.
  • the terminal device sending the resource acquisition request receives the device identifier of the target terminal device, it can establish P2P communication with the target terminal device, and can acquire the target resource from the target terminal device.
  • the edge computing scheduling system may choose to obtain the target resource by returning to the source. Specifically, the edge computing scheduling system may identify the edge computing domain name in the resource acquisition request, and determine the accelerated domain name that has a mapping relationship with the edge computing domain name, and then may obtain the target pointed to by the accelerated domain name from the cache server Resource, and provide the acquired target resource to the terminal device.
  • the process of determining the accelerated domain name is similar to the description in step S13, and will not be repeated here.
  • the manufacturer may be allowed to perform operations such as encryption and transcoding on the cached content.
  • the manufacturer can deploy a custom program on the node server in the edge node of the CDN, so as to perform operations such as encryption and transcoding on the cached content through the custom program.
  • a custom program is also deployed in the edge node.
  • the edge node may use the custom program to process the target resource, and then provide the processed resource to the terminal device.
  • the edge computing scheduling server includes a processor and a memory.
  • the memory is used to store a computer program.
  • the computer program is executed by the processor, the aforementioned resource acquisition can be achieved. method.
  • This application also provides a node scheduling method, which can be applied to an edge computing scheduling system, and the method includes the following steps.
  • S41 Receive a resource acquisition request sent by a terminal device, and query a cache server that stores the target resource pointed to by the resource acquisition request.
  • S42 Feed back the IP address of the edge node in the same computer room as the cache server to the terminal device, so that after the terminal device accesses the edge node at the IP address, the terminal device can access the edge node from the same computer room through the edge node. Acquiring the target resource from the cache server.
  • the edge computing scheduling system can receive a resource acquisition request sent by a terminal device, and can query the CDN for a cache server that stores the target resource pointed to by the resource acquisition request.
  • the edge computing scheduling system can query the edge node in the same computer room as the cache server. In this way, since the edge node and the cache server are in the same computer room, it can be ensured that the edge node accessed by the terminal device and the cache server storing the target resource are located in the same POP.
  • the edge computing scheduling system may feed back to the terminal device the IP address of the edge node in the same computer room as the cache server, so that after the terminal device accesses the edge node at the IP address, Obtain the target resource from the cache server in the same computer room through the edge node.
  • the terminal device can obtain the encrypted original link issued by the device management center, and the target resource can be accessed through the original link.
  • the original link can be carried in the resource acquisition request sent by the terminal device to the edge computing scheduling system.
  • the edge computing scheduling system can generate a redirect link containing the IP address after querying the IP address of the edge node, and The redirection link is fed back to the terminal device.
  • the edge computing scheduling system can also feed back the original link of the target resource to the terminal device. In this way, the terminal device can access the above-mentioned edge node according to the redirection link, and can provide the original link to the edge node at the same time .
  • the edge node can then obtain the target resource from the cache server in the same computer room according to the original link, and feed the obtained target resource back to the terminal device. In this way, it can be ensured that the edge node accessed by the terminal device and the cache server are in the same POP.
  • the edge computing scheduling server includes a processor and a memory.
  • the memory is used to store a computer program.
  • the above-mentioned node scheduling can be implemented. method.
  • the edge computing scheduling system can execute the content of steps S31 and S33, and prioritize whether the target resource exists in other terminal devices. If so, it can pass between the terminal device and the target terminal device. P2P method for resource acquisition. If the target resource does not exist in other terminal devices, the resource can be obtained through two different scheduling methods of steps S11 to S17 or steps S41 and S42. Through the above two different scheduling methods, cross-POP can be avoided. Data transmission, thereby saving billing costs, and can have a lower network delay, thereby improving the efficiency of data transmission.
  • the terminal device After receiving the resource prefetch instruction issued by the equipment management center, the terminal device may send a resource acquisition request to the edge computing scheduling system in response to the resource prefetch instruction. After receiving the resource acquisition request, the edge computing scheduling system can determine whether the target resource pointed to by the resource acquisition request is stored in other terminal devices. If the target resource is stored in other terminal devices, the edge computing scheduling system can determine a target terminal device from the other terminal devices obtained by the query, and the edge computing scheduling system can feed back the device identification of the target terminal device to the terminal Device, so that the terminal device obtains the target resource from the target terminal device. In this way, multiple terminal devices can provide each other with the resources they need.
  • the preferred implementation of the present application can add an edge computing scheduling system to the CDN, and add a node DNS server in each edge node.
  • the public network DNS server After the public network DNS server receives the domain name resolution request sent by the terminal device, it can forward the domain name resolution request to the edge computing scheduling system, and the domain name resolution request can carry the edge computing domain name to be resolved.
  • the edge computing domain name can be mapped to an accelerated domain name in advance.
  • the above public network DNS server can resolve the IP address of the corresponding node server through the edge computing domain name, and the accelerated domain name can be resolved by the node DNS server to obtain The IP address of the corresponding cache server.
  • the edge computing scheduling system firstly, the accelerated domain name with the mapping relationship can be determined according to the above edge computing domain name. Then, the edge IP addresses and cached IP addresses of the same POP belonging to the current area can be filtered out, and the filtered edge IP addresses can be fed back to the public network DNS server.
  • the public network DNS server can resolve the edge computing domain name to the edge IP address, so that the domain name resolution request of the terminal device can be dispatched to the node server pointed to by the edge IP address.
  • the edge computing scheduling system may feed back the above-mentioned cached IP address to the node DNS server. In this way, the node server can obtain the corresponding resource from the cache server pointed to by the cache IP address.
  • the edge IP address and the cache IP address belong to the same POP in the current area, the data transmission between the corresponding node server and the cache server belongs to the data transmission within the same POP, which will not cause cross-POP data transmission, thereby saving billing Cost, and can have a lower network delay, thereby improving the efficiency of data transmission.
  • each implementation manner can be implemented by software plus a necessary general hardware platform, and of course, it can also be implemented by hardware.
  • the above technical solutions can be embodied in the form of software products, which can be stored in computer-readable storage media, such as ROM/RAM, magnetic A disc, an optical disc, etc., include a number of instructions to make a computer device (which may be a personal computer, a server, or a network device, etc.) execute the methods described in each embodiment or some parts of the embodiment.

Landscapes

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

Abstract

本发明公开了一种资源获取方法及边缘计算调度服务器，其中，所述方法包括：接收终端设备发来的资源获取请求，并判断所述资源获取请求指向的目标资源是否存储于其它终端设备中；若所述目标资源存储于其它终端设备中，从所述其它终端设备中确定目标终端设备，并将所述目标终端设备的设备标识反馈给所述终端设备，以使得所述终端设备从所述目标终端设备处获取所述目标资源。本申请提供的技术方案，能够提高资源获取的效率。

Description

一种资源获取方法及边缘计算调度服务器 技术领域
本发明涉及互联网技术领域，特别涉及一种资源获取方法及边缘计算调度服务器。
背景技术
在当前的内容分发网络(Content Delivery Network，CDN)中，当节点服务器接收到用户客户端发来的资源获取请求后，可以从本地缓存中查询是否存在该资源获取请求指向的目标资源，如果存在，则可以将目标资源提供给用户客户端。如果不存在，则可以通过回源的方式，从缓存服务器中获取目标资源，并将获取的目标资源提供给用户客户端。
然而，上述现有技术中的这种数据传输方式，各个用户客户端的资源获取过程都是需要CDN来支持，这样一方面会增加CDN的负担，另一方面如果CDN出现网络波动，会影响大量用户客户端获取资源的效率。
发明内容
本申请的目的在于提供一种资源获取方法及边缘计算调度服务器，能够提高资源获取的效率。
为实现上述目的，本申请一方面提供一种资源获取方法，所述方法包括：接收终端设备发来的资源获取请求，并判断所述资源获取请求指向的目标资源是否存储于其它终端设备中；若所述目标资源存储于其它终端设备中，从所述其它终端设备中确定目标终端设备，并将所述目标终端设备的设备标识反馈给所述终端设备，以使得所述终端设备从所述目标终端设备处获取所述目标资源。
为实现上述目的，本申请另一方面还提供一种边缘计算调度服务器，所述边缘计算调度服务器包括处理器和存储器，所述存储器用于存储计算机程序，所述计算机程序被所述处理器执行时，实现上述的资源获取方法。
由上可见，本申请提供的技术方案，可以维护资源与终端设备的映射关系。 终端设备在接收到设备管理中心下达的资源预取指令后，可以响应于该资源预取指令，向边缘计算调度***发送资源获取请求。边缘计算调度***接收到该资源获取请求后，可以判断该资源获取请求指向的目标资源是否存储于其它的终端设备中。如果目标资源存储于其它终端设备中，那么边缘计算调度***可以从查询得到的其它终端设备中，确定一个目标终端设备，边缘计算调度***可以将所述目标终端设备的设备标识反馈给所述终端设备，以使得所述终端设备从所述目标终端设备处获取所述目标资源。这样，多个终端设备之间可以相互提供各自所需的资源，一方面降低了CDN的负担，另一方面也提高了资源获取的灵活度，进而提高了资源获取的效率。
附图说明
为了更清楚地说明本发明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发明的一些实施例，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他的附图。
图1是本发明实施方式中CDN***的结构示意图；
图2是本发明实施方式中节点调度方法的步骤示意图；
图3是本发明实施方式中同POP的IP地址的示意图；
图4是本发明实施方式中边缘计算调度服务器的结构示意图；
图5是本发明实施方式中域名解析方法的示意图；
图6是本发明实施方式中资源获取方法的应用场景示意图。
具体实施方式
为使本发明的目的、技术方案和优点更加清楚，下面将结合附图对本发明实施方式作进一步地详细描述。
本申请提供一种节点调度方法和域名解析方法，所述方法可以应用于如图1所示的***架构中。在图1所示的***中，可以包括设备管理中心、终端设备、边缘节点、源站服务器、公网DNS服务器、边缘计算调度***以及CDN调度***等多个组成部分，其中，所述边缘节点中可以包括缓存服务器、节点内部的节点DNS服务器以及多个节点服务器。
在本实施方式中，上述的终端设备，可以是厂商出售给用户的电子设备，该电子设备可以具备网络通信功能以及资源存储功能。例如，所述终端设备可以是具备数据存储功能的路由器。所述设备管理中心可以是厂商预先部署的中心服务器，该设备管理中心可以与所述终端设备进行通信连接，从而可以向该终端设备下达资源预取指令。所述终端设备后续可以通过公网DNS服务器进行域名解析，从而就近接入至CDN中的某个边缘节点中的节点服务器处。所述节点服务器可以从所述缓存服务器处获取对应的资源，如果所述缓存服务器本地未存储对应的资源，则可以从源站服务器回源下载，并将下载的资源提供给节点服务器。
请参阅图2，本申请提供的节点调度方法，可以应用于上述的边缘计算调度***中，所述方法可以包括以下步骤。
S11：接收公网DNS服务器发来的域名解析请求，所述域名解析请求中包括待解析的边缘计算域名。
在本实施方式中，终端设备在接收到设备管理中心下达的热点资源的资源预取指令后，可以根据该资源预取指令中携带的资源标识，构建用于访问热点资源的边缘计算域名。构建的所述边缘计算域名中，可以包括存储所述热点资源的网站的顶级域名和该热点资源的资源标识。这样，构建得到的所述边缘计算域名可以是一个携带所述资源标识的URL(Uniform Resource Locator，统一资源定位符)。
在本实施方式中，终端设备可以通过访问上述的边缘计算域名来获取对应的热点资源。具体地，终端设备可以向公网DNS服务器发送域名解析请求，该域名解析请求中可以携带上述的边缘计算域名。在常规的域名解析技术中，公网DNS服务器可以根据预先配置的域名与IP地址之间的映射关系，将该边缘计算域名解析为对应的IP地址。然而，为了保证终端设备接入的节点服务器与缓存服务器处于同一POP(Point of Presence，接入网点)，在本实施方式中可以利用边缘计算调度***来改变公网DNS服务器的域名解析结果。因此，公网DNS服务器在接收到终端设备发来的域名解析请求后，可以将该域名解析请求转发至边缘计算调度***，从而由边缘计算调度***决定，应当将上述的边缘计算域名解析为哪个节点服务器的IP地址。
S13：确定与所述边缘计算域名具备映射关系的加速域名，并分别确定所述 边缘计算域名对应的边缘IP地址和所述加速域名对应的缓存IP地址；其中，所述边缘IP地址和所述缓存IP地址属于当前区域的同一接入网点。
在本实施方式中，终端设备访问热点资源的边缘计算域名，在CDN中通常会具备互为映射关系的加速域名。该边缘计算域名与加速域名之间的映射关系，可以是在CDN中预先部署的。例如，终端设备访问的边缘计算域名为www.***.com，而该边缘计算域名在CDN内部的加速域名可以是***.wangsu.com。其中，所述边缘计算域名可以用于确定终端设备接入的节点服务器，而所述加速域名则可以用于确定存储上述热点资源的缓存服务器。
在本实施方式中，边缘计算调度***接收到公网DNS服务器转发来的域名解析请求后，为了保证后续确定的节点服务器与缓存服务器属于同POP，因此可以根据预先配置的映射关系表，查询到与所述边缘计算域名具备映射关系的加速域名。
在上述的CDN调度***中，可以收集CDN中各个节点服务器的边缘IP地址以及各个缓存服务器的缓存IP地址，并且可以汇总各个IP地址对应的POP信息。在本实施方式中，CDN调度***在汇总得到边缘IP地址、缓存IP地址以及POP信息后，可以将这些IP地址和POP信息都提供给所述边缘计算调度***。后续，边缘计算调度***便可以根据CDN调度***提供的信息，确定出同POP的节点服务器和缓存服务器。这样，在所述边缘计算调度***中，可以预先配置所述边缘计算域名的边缘IP地址集合，在所述边缘IP地址集合中可以包括一个或者多个边缘IP地址。其中，不同的边缘IP地址可以指向CDN中不同的节点服务器，这样，通过将所述边缘计算域名解析为其中的一个边缘IP地址，便可以将终端设备调度至该边缘IP地址指向的节点服务器处。在实际应用中，不同的边缘IP地址可能会具备不同的POP信息，因此，根据CDN调度***提供的各个IP地址的POP信息，可将边缘IP地址集合中的各个边缘IP地址归类，同一类的边缘IP地址可以属于同一个POP。请参阅图3，假设边缘IP地址集合中包括5个边缘IP地址，而这5个边缘IP地址可以划分为两类，IP1至IP3属于POP1，而IP4和IP5属于POP2。类似地，在所述边缘计算调度***中，也可以具备所述加速域名的缓存IP地址集合，并且该缓存IP地址集合中的各个缓存IP地址也可以按照POP信息归类。例如，在图3中，加速域名的缓存IP地址集合中包括3个缓存IP地址，其中，IP6属于POP1，IP7和IP8属于POP3。
需要说明的是，在实际应用中，由于CDN***通常会覆盖不同的区域，因此上述的边缘IP地址集合和缓存IP地址集合是针对于某一个区域而言的。随着区域范围的扩大或者缩小，边缘IP地址集合和缓存IP地址集合内包含的IP地址的数量也可能随之增多或者减少。具体地，可以根据发送所述域名解析请求的终端设备所处的地理位置，选用对应区域的边缘IP地址集合和缓存IP地址集合。例如，所述终端设备位于厦门，那么边缘计算调度***在针对该终端设备发出的域名解析请求时，会调用厦门区域的边缘IP地址集合和缓存IP地址集合。当然，在一些特殊情况中，如果当前区域不具备对应的边缘IP地址集合和/或缓存IP地址集合，则可以选用该当前区域归属的上级区域的边缘IP地址集合和缓/或缓存IP地址集合。例如，针对位于厦门的终端设备发出的域名解析请求，可以选用福建或者华南地区的边缘IP地址集合和缓/或缓存IP地址集合。
在本实施方式中，在确定出与所述边缘计算域名具备映射关系的加速域名后，边缘计算调度***可以在当前区域中分别确定所述边缘计算域名对应的边缘IP地址集合和所述加速域名对应的缓存IP地址集合，然后，为了保证节点服务器与缓存服务器属于同POP，边缘计算调度***可以在所述边缘IP地址集合和所述缓存IP地址集合中确定属于同一接入网点的边缘IP地址和缓存IP地址。例如，在图3中，IP1至IP3属于POP1，而IP6也属于POP1，因此，可以将IP1至IP3作为确定的边缘IP地址，以及将IP6作为确定的缓存IP地址。
这样，通过边缘计算调度***，可以分别确定出所述边缘计算域名对应的边缘IP地址和所述加速域名对应的缓存IP地址，并且，所述边缘IP地址和所述缓存IP地址属于当前区域的同一接入网点。
S15：向所述公网DNS服务器反馈所述边缘IP地址，以将所述边缘计算域名调度至所述边缘IP地址指向的节点服务器。
在本实施方式中，边缘计算调度***在确定出属于当前区域的同POP的边缘IP地址和缓存IP地址后，可以将该边缘IP地址反馈给所述公网DNS服务器，以使得所述公网DNS服务器将终端设备的边缘计算域名解析为接收到的边缘IP地址。如果接收到的边缘IP地址有多个，则可以任意选择其中的一个，或者按照预先配置的负载均衡策略选择其中的一个。
在本实施方式中，终端设备可以根据公网DNS服务器反馈的边缘IP地址， 访问对应的节点服务器。这样，终端设备访问的边缘计算域名便被调度至该边缘IP地址指向的节点服务器处。
S17：在接收到所述节点服务器所处的边缘节点中的节点DNS服务器发来的包含所述加速域名的域名解析请求后，向所述节点DNS服务器反馈与所述边缘IP地址属于同一接入网点的所述缓存IP地址，以将所述加速域名调度至所述缓存IP地址指向的缓存服务器。
在本实施方式中，节点服务器所处的边缘节点中，可以部署本节点的节点DNS服务器，该节点DNS服务器可以用于将节点服务器访问的加速域名解析为对应的缓存IP地址。具体地，终端设备访问的边缘计算域名，在边缘节点中可以转换为具备映射关系的所述加速域名。这样，节点服务器可以向本节点的节点DNS服务器发送包含所述加速域名的域名解析请求。所述节点DNS服务器接收到该域名解析请求后，同样可以将该域名解析请求转发至所述边缘计算调度***，以通过边缘计算调度***来决定应当将该加速域名解析为哪个缓存IP地址。
由于边缘计算调度***已经确定出当前区域中属于同一POP的边缘IP地址和缓存IP地址，因此，在接收到所述节点DNS服务器发来的域名解析请求后，可以向所述节点DNS服务器反馈与所述边缘IP地址属于同一接入网点的所述缓存IP地址。这样，所述节点DNS服务器便可以将所述加速域名解析为该缓存IP地址，从而将该加速域名调度至所述缓存IP地址指向的缓存服务器处。后续，节点服务器便可以访问该缓存IP地址指向的缓存服务器。由于该节点服务器的边缘IP地址与该缓存服务器的缓存IP地址属于当前区域的同POP，因此不存在跨POP通信的问题，从而避免了跨POP通信导致的额外收费和延时较高的问题。
在一个实施方式中，如果在所述当前区域中不存在属于同一接入网点的边缘IP地址和缓存IP地址，那么边缘计算调度***可以扩大区域的覆盖范围，并从更大覆盖范围的区域中查询是否存在属于同一接入网点的边缘IP地址和缓存IP地址。具体地，若在所述当前区域中不存在属于同一接入网点的边缘IP地址和缓存IP地址，边缘计算调度***可以确定所述当前区域归属的上级区域，并在所述上级区域中，确定属于同一接入网点的边缘IP地址和缓存IP地址。举例来说，如果在厦门的区域内，不存在属于同一接入网点的边缘IP地址和缓存IP地址，那么边缘计算调度***则可以在福建的区域内，甚至整个华南区域内，确定属于同一接入网点的边缘IP地址和缓存IP地址。这样，通过不断扩 大区域覆盖范围的方法，可以逐步找出属于同一接入网点的边缘IP地址和缓存IP地址。
在一个实施方式中，如果在所述上级区域中仍然不存在属于同一接入网点的边缘IP地址和缓存IP地址，那么此时为了保证终端设备与节点服务器，以及节点服务器与缓存服务器之间的通信质量，不宜再继续扩大区域的覆盖范围，否则查询得到的边缘IP地址和缓存IP地址尽管属于同POP，但可能由于地理位置相隔较远，导致终端设备与节点服务器，以及节点服务器与缓存服务器之间的通信质量较差。鉴于此，在本实施方式中，若在所述上级区域中不存在属于同一接入网点的边缘IP地址和缓存IP地址，那么边缘计算调度***可以获取所述边缘计算域名在所述当前区域中预先配置的边缘IP地址集合，并将所述边缘IP地址集合反馈给所述公网DNS服务器，以使得所述公网DNS服务器将所述边缘计算域名解析为所述边缘IP地址集合中的目标IP地址。后续，当该目标IP地址指向的节点服务器所处的边缘节点中的节点DNS服务器向边缘计算调度***发送包含所述加速域名的域名解析请求后，边缘计算调度***可以获取所述加速域名在所述当前区域中预先配置的缓存IP地址集合，并向该节点DNS服务器反馈所述缓存IP地址集合，从而使得节点DNS服务器将所述加速域名解析为所述缓存IP地址集合中的一个缓存IP地址。这样处理的目的在于，在这种场景下，就不再刻意要求节点服务器和缓存服务器需要属于同POP，而是以网络通信质量为前提，让终端设备能够享受到较好的数据传输效果。
在一个实施方式中，CDN中的节点服务器和缓存服务器有可能会出现故障，或者出现变更，从而导致边缘计算域名对应的边缘IP地址以及缓存服务器对应的缓存IP地址发生变更，此外，随着网络线路调整，各个IP地址的POP信息也可能发生变更。因此，所述CDN调度***可以定期地收集边缘计算域名和加速域名对应的IP地址，以及各个IP地址的POP信息。一旦IP地址或者POP信息与之前的版本发生变化，可以向边缘计算调度***发送节点信息变更通知。该节点信息变更通知中，可以注明存在变更的IP地址和/或IP地址变更后的POP信息。这样，边缘计算调度***可以根据所述节点信息变更通知，更新本地存储的所述边缘IP地址集合和所述缓存IP地址集合中的IP地址信息和接入网点信息，以保证后续在处理域名解析请求时，能够确定出当前区域内属于同POP的节点服务器和缓存服务器。
本申请还提供一种边缘计算调度***，所述***包括：
域名解析请求接收单元，用于接收公网DNS服务器发来的域名解析请求，所述域名解析请求中包括待解析的边缘计算域名；
同接入网点确定单元，用于确定与所述边缘计算域名具备映射关系的加速域名，并分别确定所述边缘计算域名对应的边缘IP地址和所述加速域名对应的缓存IP地址；其中，所述边缘IP地址和所述缓存IP地址属于当前区域的同一接入网点；
边缘调度单元，用于向所述公网DNS服务器反馈所述边缘IP地址，以将所述边缘计算域名调度至所述边缘IP地址指向的节点服务器；
缓存调度单元，用于在接收到所述节点服务器所处的边缘节点中的节点DNS服务器发来的包含所述加速域名的域名解析请求后，向所述节点DNS服务器反馈与所述边缘IP地址属于同一接入网点的所述缓存IP地址，以将所述加速域名调度至所述缓存IP地址指向的缓存服务器。
在一个实施方式中，所述同接入网点确定单元包括：
IP地址集合确定模块，用于在当前区域中分别确定所述边缘计算域名对应的边缘IP地址集合和所述加速域名对应的缓存IP地址集合；
同接入网点筛选模块，用于在所述边缘IP地址集合和所述缓存IP地址集合中确定属于同一接入网点的边缘IP地址和缓存IP地址。
在一个实施方式中，所述***还包括：
上级区域确定单元，用于若在所述当前区域中不存在属于同一接入网点的边缘IP地址和缓存IP地址，确定所述当前区域归属的上级区域，并在所述上级区域中，确定属于同一接入网点的边缘IP地址和缓存IP地址；
预先配置单元，用于若在所述上级区域中不存在属于同一接入网点的边缘IP地址和缓存IP地址，获取所述边缘计算域名在所述当前区域中预先配置的边缘IP地址集合，并将所述边缘IP地址集合反馈给所述公网DNS服务器，以使得所述公网DNS服务器将所述边缘计算域名解析为所述边缘IP地址集合中的目标IP地址。
在一个实施方式中，所述缓存调度单元，还用于在接收到所述目标IP地址指向的边缘节点中的节点DNS服务器发来的包含所述加速域名的域名解析请求后，获取所述加速域名在所述当前区域中预先配置的缓存IP地址集合，并向所 述节点DNS服务器反馈所述缓存IP地址集合，以使得所述节点DNS服务器将所述加速域名解析为所述缓存IP地址集合中的IP地址。
在一个实施方式中，所述***还包括：
信息更新单元，用于接收CDN调度***发来的节点信息变更通知，并根据所述节点信息变更通知，更新所述边缘IP地址集合和所述缓存IP地址集合中的IP地址信息和接入网点信息。
请参阅图4，本申请还提供一种边缘计算调度服务器，所述边缘计算调度服务器包括处理器和存储器，所述存储器用于存储计算机程序，所述计算机程序被所述处理器执行时，实现上述的节点调度方法。
本申请还提供一种域名解析方法，所述方法可以应用于边缘节点内的节点DNS服务器中，请参阅图5，所述方法包括以下步骤。
S21：接收所述边缘节点内的节点服务器发来的域名解析请求，所述域名解析请求中包括加速域名。
S23：从边缘计算调度***处获取与所述节点服务器在当前区域中属于同一接入网点的缓存服务器的缓存IP地址。
S25：将所述缓存IP地址反馈给所述节点服务器，以使得所述节点服务器从所述缓存IP地址指向的缓存服务器处获取所述加速域名对应的资源。
在本实施方式中，节点服务器所处的边缘节点中，可以部署本节点的节点DNS服务器，该节点DNS服务器可以用于将节点服务器访问的加速域名解析为对应的缓存IP地址。具体地，终端设备访问的边缘计算域名，在边缘节点中可以转换为具备映射关系的所述加速域名。这样，节点服务器可以向本节点的节点DNS服务器发送包含所述加速域名的域名解析请求。所述节点DNS服务器接收到该域名解析请求后，可以将该域名解析请求转发至所述边缘计算调度***，以通过边缘计算调度***来决定应当将该加速域名解析为哪个缓存IP地址。
由于边缘计算调度***已经确定出当前区域中属于同一POP的边缘IP地址和缓存IP地址，因此，在接收到所述节点DNS服务器发来的域名解析请求后，可以向所述节点DNS服务器反馈与所述节点服务器属于同一接入网点的缓存服务器的缓存IP地址。这样，所述节点DNS服务器便可以将所述加速域名解析为该缓存IP地址，从而将该加速域名调度至所述缓存IP地址指向的缓存服务器处。后续，节点服务器便可以访问该缓存IP地址指向的缓存服务器。由于该节 点服务器的边缘IP地址与该缓存服务器的缓存IP地址属于当前区域的同POP，因此不存在跨POP通信的问题，从而避免了跨POP通信导致的额外收费和延时较高的问题。
在一个实施方式中，所述方法还包括：
若在所述当前区域中不存在属于同一接入网点的节点服务器和缓存服务器，从所述边缘计算调度***处获取与所述节点服务器在所述当前区域归属的上级区域中属于同一接入网点的缓存服务器的缓存IP地址。
在一个实施方式中，所述方法还包括：
若在所述上级区域中不存在属于同一接入网点的节点服务器和缓存服务器，从所述边缘计算调度***处获取为所述节点DNS服务器预先配置的缓存IP地址集合，并将所述缓存IP地址集合反馈给所述节点服务器，以使得所述节点服务器从所述缓存IP地址集合中的任一IP地址指向的缓存服务器处获取所述加速域名对应的资源。
本申请还提供一种节点DNS服务器，所述节点DNS服务器处于边缘节点内，所述节点DNS服务器包括：
请求接收单元，用于接收所述边缘节点内的节点服务器发来的域名解析请求，所述域名解析请求中包括加速域名；
IP地址信息获取单元，用于从边缘计算调度***处获取与所述节点服务器在当前区域中属于同一接入网点的缓存服务器的缓存IP地址；
域名解析单元，用于将所述缓存IP地址反馈给所述节点服务器，以使得所述节点服务器从所述缓存IP地址指向的缓存服务器处获取所述加速域名对应的资源。
在一个实施方式中，所述IP地址信息获取单元还用于若在所述当前区域中不存在属于同一接入网点的节点服务器和缓存服务器，从所述边缘计算调度***处获取与所述节点服务器在所述当前区域归属的上级区域中属于同一接入网点的缓存服务器的缓存IP地址；
在一个实施方式中，所述域名解析单元，还用于若在所述上级区域中不存在属于同一接入网点的节点服务器和缓存服务器，从所述边缘计算调度***处获取为所述节点DNS服务器预先配置的缓存IP地址集合，并将所述缓存IP地址集合反馈给所述节点服务器，以使得所述节点服务器从所述缓存IP地址集合 中的任一IP地址指向的缓存服务器处获取所述加速域名对应的资源。
本申请还提供一种节点DNS服务器，所述节点DNS服务器包括处理器和存储器，所述存储器用于存储计算机程序，所述计算机程序被所述处理器执行时，实现上述的域名解析方法。
本申请还提供一种资源获取方法，请参阅图6，所述方法可以应用于图6所示的***中。具体地，所述方法可以包括以下步骤。
S31：接收终端设备发来的资源获取请求，并判断所述资源获取请求指向的目标资源是否存储于其它终端设备中。
在本实施方式中，终端设备在接收到设备管理中心下达的资源预取指令后，可以响应于该资源预取指令，向边缘计算调度***发送资源获取请求。边缘计算调度***接收到该资源获取请求后，可以判断该资源获取请求指向的目标资源是否存储于其它的终端设备中。
在实际应用中，可以维护资源与终端设备的映射关系表。在该映射关系表中，可以将资源标识与存储资源的终端设备进行关联存储。这样，通过资源标识，可以快速查询到是否存在关联的终端设备。
在本实施方式中，边缘计算调度***可以提取资源获取请求中的目标资源的资源标识，并在所述映射关系表中查询是否存在与所述目标资源的资源标识相关联的其它终端设备，从而判断所述资源获取请求指向的目标资源是否存储于其它终端设备中。
S33：若所述目标资源存储于其它终端设备中，从所述其它终端设备中确定目标终端设备，并将所述目标终端设备的设备标识反馈给所述终端设备，以使得所述终端设备从所述目标终端设备处获取所述目标资源。
在本实施方式中，如果目标资源存储于其它终端设备中，那么边缘计算调度***可以从查询得到的其它终端设备中，确定一个目标终端设备。在实际应用中，可以通过多种方式确定所述目标终端设备。例如，可以从查询得到的其它终端设备中随机选择一个终端设备，作为所述目标终端设备。也可以根据负载均衡算法，将当前负载较低的终端设备作为所述目标终端设备。其中，终端设备的负载，可以指该终端设备目前建立的P2P(Point to Point，点对点)通信链接的数量。数量越多，可以表示负载越高。当然，在实际应用中，还存在其它更多的方式从多个其它终端设备中选择目标终端设备，这里便不一一例举。
在本实施方式中，在确定出目标终端设备后，边缘计算调度***可以将所述目标终端设备的设备标识反馈给所述终端设备，以使得所述终端设备从所述目标终端设备处获取所述目标资源。其中，所述目标终端设备的设备标识可以指目标终端设备的网络通信地址，也可以指符合P2P通信协议的用于表征目标终端设备身份的字符串。这样，当发送资源获取请求的终端设备接收到目标终端设备的设备标识后，便可以与目标终端设备建立P2P通信，并可以从目标终端设备处获取所述目标资源。
在一个实施方式中，若所述目标资源未存储于其它终端设备中，那么边缘计算调度***可以选择通过回源的方式获取目标资源。具体地，边缘计算调度***可以识别所述资源获取请求中的边缘计算域名，并确定与所述边缘计算域名具备映射关系的加速域名，然后，可以从缓存服务器中获取所述加速域名指向的目标资源，并将获取的所述目标资源提供给所述终端设备。其中，确定加速域名的过程与步骤S13中的描述类似，这里便不再赘述。
在一个实施方式中，为了满足厂商的自定制需求，可以允许厂商对缓存内容进行加密、转码等操作。具体地，厂商可以在CDN的边缘节点中的节点服务器上部署自定义程序，从而通过该自定义程序，对缓存内容进行加密、转码等操作。这样，所述边缘节点中还部署有自定义程序。在从缓存服务器中获取所述加速域名指向的目标资源之后，边缘节点可以利用所述自定义程序对所述目标资源进行处理后，将处理得到的资源提供给所述终端设备。
本申请还提供一种边缘计算调度服务器，所述边缘计算调度服务器包括处理器和存储器，所述存储器用于存储计算机程序，所述计算机程序被所述处理器执行时，可以实现上述的资源获取方法。
本申请还提供一种节点调度方法，所述方法可以应用于边缘计算调度***中，所述方法包括以下步骤。
S41：接收终端设备发来的资源获取请求，并查询存储所述资源获取请求指向的目标资源的缓存服务器。
S42：向所述终端设备反馈与所述缓存服务器处于同机房的边缘节点的IP地址，以使得所述终端设备接入所述IP地址处的边缘节点后，通过所述边缘节点从同机房的所述缓存服务器处获取所述目标资源。
在本实施方式中，边缘计算调度***可以接收终端设备发来的资源获取请 求，并可以在CDN中查询存储有该资源获取请求指向的目标资源的缓存服务器。为了使得终端设备接入的边缘节点与存储目标资源的缓存服务器处于同POP中，边缘计算调度***可以查询与所述缓存服务器处于同机房的边缘节点。这样，由于边缘节点与缓存服务器处于同机房中，从而可以保证终端设备接入的边缘节点与存储目标资源的缓存服务器位于同POP中。
在本实施方式中，边缘计算调度***可以向所述终端设备反馈与所述缓存服务器处于同机房的边缘节点的IP地址，以使得所述终端设备接入所述IP地址处的边缘节点后，通过所述边缘节点从同机房的所述缓存服务器处获取所述目标资源。
在实际应用中，终端设备可以获取到设备管理中心下发的经过加密的原始链接，通过该原始链接可以访问目标资源。在终端设备发送给边缘计算调度***的资源获取请求中可以携带该原始链接，后续，边缘计算调度***在查询得到边缘节点的IP地址后，可以生成包含所述IP地址的重定向链接，并将所述重定向链接反馈给所述终端设备。同时，边缘计算调度***可以一并向所述终端设备反馈所述目标资源的原始链接，这样，终端设备根据重定向链接，可以接入上述的边缘节点中，同时可以将原始链接提供给边缘节点。边缘节点便可以按照原始链接，从同机房的缓存服务器处获取目标资源，并将获取的目标资源反馈给终端设备。这样便可以保证终端设备接入的边缘节点与缓存服务器处于同POP中。
本申请还提供一种边缘计算调度服务器，所述边缘计算调度服务器包括处理器和存储器，所述存储器用于存储计算机程序，所述计算机程序被所述处理器执行时，可以实现上述的节点调度方法。
需要说明的是，在实际应用中，边缘计算调度***可以执行步骤S31和S33的内容，优先判断目标资源是否存在于其它终端设备中，如果存在，那么可以在终端设备和目标终端设备之间通过P2P的方式进行资源获取。如果目标资源不存在于其它终端设备中，那么可以分别通过步骤S11至S17或者步骤S41和S42的两种不同的调度方式来进行资源获取，通过上述两种不同的调度方式，可以避免跨POP的数据传输，从而节省了计费成本，并且能够具备较低的网络延时，进而提高了数据传输的效率。
由上可见，本申请提供的技术方案，可以维护资源与终端设备的映射关系。 终端设备在接收到设备管理中心下达的资源预取指令后，可以响应于该资源预取指令，向边缘计算调度***发送资源获取请求。边缘计算调度***接收到该资源获取请求后，可以判断该资源获取请求指向的目标资源是否存储于其它的终端设备中。如果目标资源存储于其它终端设备中，那么边缘计算调度***可以从查询得到的其它终端设备中，确定一个目标终端设备，边缘计算调度***可以将所述目标终端设备的设备标识反馈给所述终端设备，以使得所述终端设备从所述目标终端设备处获取所述目标资源。这样，多个终端设备之间可以相互提供各自所需的资源，一方面降低了CDN的负担，另一方面也提高了资源获取的灵活度，进而提高了资源获取的效率。此外，本申请的优选实施方式可以在CDN中加入边缘计算调度***，并且在各个边缘节点内加入节点DNS服务器。当公网DNS服务器接收到终端设备发来的域名解析请求后，可以将该域名解析请求转发至边缘计算调度***，该域名解析请求中可以携带待解析的边缘计算域名。在CDN内部，边缘计算域名可以预先与一个加速域名互相映射，上述的公网DNS服务器可以通过边缘计算域名，解析得到对应的节点服务器的IP地址，而加速域名可以由节点DNS服务器进行解析，得到对应的缓存服务器的IP地址。为了让节点服务器与缓存服务器属于同一个POP，在边缘计算调度***中，首先可以根据上述的边缘计算域名，确定出具备映射关系的加速域名。然后，可以筛选出属于当前区域的同POP的边缘IP地址和缓存IP地址，并可以将筛选出的边缘IP地址反馈给公网DNS服务器。这样，公网DNS服务器便可以将边缘计算域名解析为该边缘IP地址，从而可以将终端设备的域名解析请求调度至该边缘IP地址指向的节点服务器处。后续，在接收到边缘节点内的节点DNS服务器发来的加速域名解析请求后，边缘计算调度***可以向该节点DNS服务器反馈上述的缓存IP地址。这样，节点服务器便可以从该缓存IP地址指向的缓存服务器处获取对应的资源。由于边缘IP地址和缓存IP地址属于当前区域的同POP，因此对应的节点服务器和缓存服务器之间的数据传输属于同POP内的数据传输，不会造成跨POP的数据传输，从而节省了计费成本，并且能够具备较低的网络延时，进而提高了数据传输的效率。
本说明书中的各个实施方式均采用递进的方式描述，各个实施方式之间相同相似的部分互相参见即可，每个实施方式重点说明的都是与其他实施方式的不同之处。尤其，针对边缘节点的实施方式来说，均可以参照前述方法的实施 方式的介绍对照解释。
通过以上的实施方式的描述，本领域的技术人员可以清楚地了解到各实施方式可借助软件加必需的通用硬件平台的方式来实现，当然也可以通过硬件。基于这样的理解，上述技术方案本质上或者说对现有技术做出贡献的部分可以以软件产品的形式体现出来，该计算机软件产品可以存储在计算机可读存储介质中，如ROM/RAM、磁碟、光盘等，包括若干指令用以使得一台计算机设备(可以是个人计算机，服务器，或者网络设备等)执行各个实施例或者实施例的某些部分所述的方法。
以上所述仅为本发明的较佳实施例，并不用以限制本发明，凡在本发明的精神和原则之内，所作的任何修改、等同替换、改进等，均应包含在本发明的保护范围之内。

Claims (16)

  1. 一种资源获取方法，其特征在于，所述方法包括：
    接收终端设备发来的资源获取请求，并判断所述资源获取请求指向的目标资源是否存储于其它终端设备中；
    若所述目标资源存储于其它终端设备中，从所述其它终端设备中确定目标终端设备，并将所述目标终端设备的设备标识反馈给所述终端设备，以使得所述终端设备从所述目标终端设备处获取所述目标资源。
  2. 根据权利要求1所述的方法，其特征在于，所述方法还包括：
    若所述目标资源未存储于其它终端设备中，识别所述资源获取请求中的边缘计算域名，并确定与所述边缘计算域名具备映射关系的加速域名；
    通过边缘节点从缓存服务器中获取所述加速域名指向的目标资源，并将获取的所述目标资源提供给所述终端设备。
  3. 根据权利要求2所述的方法，其特征在于，识别所述资源获取请求中的边缘计算域名，并确定与所述边缘计算域名具备映射关系的加速域名包括：
    接收公网DNS服务器发来的域名解析请求，所述域名解析请求中包括待解析的边缘计算域名；
    确定与所述边缘计算域名具备映射关系的加速域名，并分别确定所述边缘计算域名对应的边缘IP地址和所述加速域名对应的缓存IP地址；其中，所述边缘IP地址和所述缓存IP地址属于当前区域的同一接入网点；
    向所述公网DNS服务器反馈所述边缘IP地址，以将所述边缘计算域名调度至所述边缘IP地址指向的节点服务器；
    在接收到所述节点服务器所处的边缘节点中的节点DNS服务器发来的包含所述加速域名的域名解析请求后，向所述节点DNS服务器反馈与所述边缘IP地址属于同一接入网点的所述缓存IP地址，以将所述加速域名调度至所述缓存IP地址指向的缓存服务器。
  4. 根据权利要求3所述的方法，其特征在于，分别确定所述边缘计算域名 对应的边缘IP地址和所述加速域名对应的缓存IP地址包括：
    在当前区域中分别确定所述边缘计算域名对应的边缘IP地址集合和所述加速域名对应的缓存IP地址集合；
    在所述边缘IP地址集合和所述缓存IP地址集合中确定属于同一接入网点的边缘IP地址和缓存IP地址。
  5. 根据权利要求3所述的方法，其特征在于，所述方法还包括：
    若在所述当前区域中不存在属于同一接入网点的边缘IP地址和缓存IP地址，确定所述当前区域归属的上级区域，并在所述上级区域中，确定属于同一接入网点的边缘IP地址和缓存IP地址。
  6. 根据权利要求5所述的方法，其特征在于，所述方法还包括：
    若在所述上级区域中不存在属于同一接入网点的边缘IP地址和缓存IP地址，获取所述边缘计算域名在所述当前区域中预先配置的边缘IP地址集合，并将所述边缘IP地址集合反馈给所述公网DNS服务器，以使得所述公网DNS服务器将所述边缘计算域名解析为所述边缘IP地址集合中的目标IP地址。
  7. 根据权利要求6所述的方法，其特征在于，所述方法还包括：
    在接收到所述目标IP地址指向的边缘节点中的节点DNS服务器发来的包含所述加速域名的域名解析请求后，获取所述加速域名在所述当前区域中预先配置的缓存IP地址集合，并向所述节点DNS服务器反馈所述缓存IP地址集合，以使得所述节点DNS服务器将所述加速域名解析为所述缓存IP地址集合中的IP地址。
  8. 根据权利要求3所述的方法，其特征在于，所述方法还包括：
    接收CDN调度***发来的节点信息变更通知，并根据所述节点信息变更通知，更新所述边缘IP地址集合和所述缓存IP地址集合中的IP地址信息和接入网点信息。
  9. 根据权利要求3所述的方法，其特征在于，所述方法还包括：
    所述节点DNS服务器接收所述边缘节点内的节点服务器发来的域名解析请求，所述域名解析请求中包括加速域名；
    所述节点DNS服务器从边缘计算调度***处获取与所述节点服务器在当前区域中属于同一接入网点的缓存服务器的缓存IP地址；
    所述节点DNS服务器将所述缓存IP地址反馈给所述节点服务器，以使得所述节点服务器从所述缓存IP地址指向的缓存服务器处获取所述加速域名对应的资源。
  10. 根据权利要求9所述的方法，其特征在于，所述方法还包括：
    若在所述当前区域中不存在属于同一接入网点的节点服务器和缓存服务器，所述节点DNS服务器从所述边缘计算调度***处获取与所述节点服务器在所述当前区域归属的上级区域中属于同一接入网点的缓存服务器的缓存IP地址。
  11. 根据权利要求10所述的方法，其特征在于，所述方法还包括：
    若在所述上级区域中不存在属于同一接入网点的节点服务器和缓存服务器，所述节点DNS服务器从所述边缘计算调度***处获取为所述节点DNS服务器预先配置的缓存IP地址集合，并将所述缓存IP地址集合反馈给所述节点服务器，以使得所述节点服务器从所述缓存IP地址集合中的任一IP地址指向的缓存服务器处获取所述加速域名对应的资源。
  12. 根据权利要求2所述的方法，其特征在于，从缓存服务器中获取所述加速域名指向的目标资源，并将获取的所述目标资源提供给所述终端设备包括：
    接收所述终端设备发来的资源获取请求，并查询存储所述资源获取请求指向的目标资源的缓存服务器；
    向所述终端设备反馈与所述缓存服务器处于同机房的边缘节点的IP地址，以使得所述终端设备接入所述IP地址处的边缘节点后，通过所述边缘节点从同机房的所述缓存服务器处获取所述目标资源。
  13. 根据权利要求12所述的方法，其特征在于，向所述终端设备反馈与所 述缓存服务器处于同机房的边缘节点的IP地址包括：
    在确定出与所述缓存服务器处于同机房的边缘节点的IP地址后，生成包含所述IP地址的重定向链接，并将所述重定向链接反馈给所述终端设备。
  14. 根据权利要求12所述的方法，其特征在于，所述方法还包括：
    在向所述终端设备反馈与所述缓存服务器处于同机房的边缘节点的IP地址时，还向所述终端设备反馈所述目标资源的原始链接，以使得所述边缘节点按照所述原始链接从所述缓存服务器处获取所述目标资源。
  15. 根据权利要求2所述的方法，其特征在于，所述边缘节点中还部署有自定义程序；在从缓存服务器中获取所述加速域名指向的目标资源之后，所述方法还包括：
    利用所述自定义程序对所述目标资源进行处理后，将处理得到的资源提供给所述终端设备。
  16. 一种边缘计算调度服务器，其特征在于，所述边缘计算调度服务器包括处理器和存储器，所述存储器用于存储计算机程序，所述计算机程序被所述处理器执行时，实现如权利要求1至15中任一所述的方法。
PCT/CN2019/110151 2019-07-26 2019-10-09 一种资源获取方法及边缘计算调度服务器 WO2021017161A1 (zh)

Priority Applications (2)

Application Number Priority Date Filing Date Title
US17/630,154 US20220286526A1 (en) 2019-07-26 2019-10-09 Method for obtaining resource and scheduling server for edge computing
EP19939974.2A EP4002796A4 (en) 2019-07-26 2019-10-09 RESOURCE PROCUREMENT PROCEDURES AND EDGE CALCULATION PLANNING SERVER

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201910681392.5 2019-07-26
CN201910681392.5A CN110460652B (zh) 2019-07-26 2019-07-26 一种资源获取方法及边缘计算调度服务器

Publications (1)

Publication Number Publication Date
WO2021017161A1 true WO2021017161A1 (zh) 2021-02-04

Family

ID=68483636

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2019/110151 WO2021017161A1 (zh) 2019-07-26 2019-10-09 一种资源获取方法及边缘计算调度服务器

Country Status (4)

Country Link
US (1) US20220286526A1 (zh)
EP (1) EP4002796A4 (zh)
CN (1) CN110460652B (zh)
WO (1) WO2021017161A1 (zh)

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113259359A (zh) * 2021-05-21 2021-08-13 重庆紫光华山智安科技有限公司 一种边缘节点能力补充方法、***、介质及电子终端
CN114666277A (zh) * 2022-05-05 2022-06-24 中国互联网络信息中心 一种基于域名的数据处理方法及装置
WO2022218304A1 (zh) * 2021-04-15 2022-10-20 华为技术有限公司 报文转发方法、装置及***、计算机可读存储介质

Families Citing this family (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110944061A (zh) * 2019-12-06 2020-03-31 北京博睿宏远数据科技股份有限公司 内容分发网络调度方法、装置、计算机设备及存储介质
CN113824750B (zh) * 2020-06-19 2023-04-18 网宿科技股份有限公司 一种传输数据的方法和数据传输***
CN112910997B (zh) * 2021-02-01 2023-11-24 福建升腾资讯有限公司 一种局域网的资源获取方法
CN113301184B (zh) * 2021-07-08 2021-10-26 凌锐蓝信科技(北京)有限公司 远程访问方法、装置、计算机设备和存储介质
CN114448990B (zh) * 2021-12-23 2023-06-23 天翼云科技有限公司 一种基于融合cdn的资源调度方法、装置及设备
CN114629911A (zh) * 2022-04-18 2022-06-14 北京字节跳动网络技术有限公司 域名解析请求的处理方法、装置、设备、介质和程序产品
CN114979146B (zh) * 2022-05-25 2024-06-11 中移（杭州）信息技术有限公司 回源方法、cdn设备、存储介质以及设备
CN115914374A (zh) * 2022-12-02 2023-04-04 中国工商银行股份有限公司 数据传输方法、装置、计算机设备、存储介质和程序产品

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101159580A (zh) * 2007-11-29 2008-04-09 中国电信股份有限公司 内容分发网络中内容p2p化方法及***
CN101193294A (zh) * 2006-11-29 2008-06-04 中兴通讯股份有限公司 一种视频内容服务***和服务方法
US20100036954A1 (en) * 2008-08-06 2010-02-11 Edgecast Networks, Inc. Global load balancing on a content delivery network
US20100042681A1 (en) * 2008-08-13 2010-02-18 Sk Telecom Co., Ltd. Contents delivery system and method using object redirection, and gslb switch thereof
CN103916328A (zh) * 2012-12-28 2014-07-09 财团法人工业技术研究院 内容分发网络及对等网络的流量控制方法及***

Family Cites Families (19)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US7991910B2 (en) * 2008-11-17 2011-08-02 Amazon Technologies, Inc. Updating routing information based on client location
US9137300B1 (en) * 2009-06-30 2015-09-15 Amazon Technologies, Inc. Opportunistic pipe switching
US8688775B2 (en) * 2010-05-28 2014-04-01 Juniper Network, Inc. Application-layer traffic optimization service spanning multiple networks
CN101984637B (zh) * 2010-11-02 2014-06-11 中兴通讯股份有限公司 内容分发实现方法及***
CA2825393C (en) * 2011-01-28 2019-03-12 Level 3 Communications, Llc Content delivery network with deep caching infrastructure
CN102694831B (zh) * 2011-03-25 2015-09-16 中国电信股份有限公司 移动终端流媒体数据补偿方法与***、内容分发网络
CN102510395B (zh) * 2011-10-21 2014-11-12 合一网络技术(北京)有限公司 一种基于P2SP的flash视频调度方法
US9160809B2 (en) * 2012-11-26 2015-10-13 Go Daddy Operating Company, LLC DNS overriding-based methods of accelerating content delivery
CN104320487B (zh) * 2014-11-11 2018-03-20 网宿科技股份有限公司 内容分发网络的http调度***和方法
US10097448B1 (en) * 2014-12-18 2018-10-09 Amazon Technologies, Inc. Routing mode and point-of-presence selection service
CN105991787A (zh) * 2015-02-16 2016-10-05 中兴通讯股份有限公司 分组数据业务的处理***、方法及装置
US10097602B2 (en) * 2016-03-29 2018-10-09 Nokia Of America Corporation Channel-aware caching-aided coded multicast
US10880395B2 (en) * 2016-09-27 2020-12-29 Level 3 Communications, Llc System and method for improvements to a content delivery network
CN106790704B (zh) * 2017-02-27 2019-09-13 网宿科技股份有限公司 一种访问云存储文件的方法及***
CN107483630A (zh) * 2017-09-19 2017-12-15 北京工业大学 一种基于边缘缓存的isp与cp联合内容分发机制的构建方法
CN109379426B (zh) * 2018-10-19 2021-08-31 中国联合网络通信集团有限公司 基于x-dns的x-cdn调度方法、设备及***
CN110012106B (zh) * 2019-04-15 2021-01-29 北京邮电大学 一种基于边缘计算的协调缓存方法、装置及***
CN110198307B (zh) * 2019-05-10 2021-05-18 深圳市腾讯计算机***有限公司 一种移动边缘计算节点的选择方法、装置及***
JP7508204B2 (ja) * 2019-06-21 2024-07-01 エヌ・ティ・ティ・コミュニケーションズ株式会社 誘導先評価装置、誘導先評価方法およびプログラム

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101193294A (zh) * 2006-11-29 2008-06-04 中兴通讯股份有限公司 一种视频内容服务***和服务方法
CN101159580A (zh) * 2007-11-29 2008-04-09 中国电信股份有限公司 内容分发网络中内容p2p化方法及***
US20100036954A1 (en) * 2008-08-06 2010-02-11 Edgecast Networks, Inc. Global load balancing on a content delivery network
US20100042681A1 (en) * 2008-08-13 2010-02-18 Sk Telecom Co., Ltd. Contents delivery system and method using object redirection, and gslb switch thereof
CN103916328A (zh) * 2012-12-28 2014-07-09 财团法人工业技术研究院 内容分发网络及对等网络的流量控制方法及***

Non-Patent Citations (3)

* Cited by examiner, † Cited by third party
Title
GONG MAOBIN; DONG SHIMENG: "Research on Relieving Network Pressure due to IPTV Development", COMMUNICATIONS MANAGEMENT AND TECHNOLOGY, no. 03, 30 June 2016 (2016-06-30), pages 67 - 69, XP009533731, ISSN: 1672-6200 *
QIAO, ZHI: "The Application of the Combination Technology of CDN and P2P to Streaming Media Distribution and Delivery System", DESIGNING TECHNIQUES OF POSTS AND TELECOMMUNICATIONS, no. 2, 29 February 2008 (2008-02-29), pages 44 - 48, XP055893630 *
See also references of EP4002796A4 *

Cited By (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2022218304A1 (zh) * 2021-04-15 2022-10-20 华为技术有限公司 报文转发方法、装置及***、计算机可读存储介质
CN113259359A (zh) * 2021-05-21 2021-08-13 重庆紫光华山智安科技有限公司 一种边缘节点能力补充方法、***、介质及电子终端
CN114666277A (zh) * 2022-05-05 2022-06-24 中国互联网络信息中心 一种基于域名的数据处理方法及装置
CN114666277B (zh) * 2022-05-05 2023-10-24 中国互联网络信息中心 一种基于域名的数据处理方法及装置

Also Published As

Publication number Publication date
CN110460652A (zh) 2019-11-15
EP4002796A4 (en) 2022-07-20
US20220286526A1 (en) 2022-09-08
CN110460652B (zh) 2021-09-14
EP4002796A1 (en) 2022-05-25

Similar Documents

Publication Publication Date Title
WO2021017161A1 (zh) 一种资源获取方法及边缘计算调度服务器
US20210021692A1 (en) Translation of resource identifiers using popularity information upon client request
US10264062B2 (en) Request routing using a popularity identifier to identify a cache component
EP2695358B1 (en) Selection of service nodes for provision of services
US7725596B2 (en) System and method for resolving network layer anycast addresses to network layer unicast addresses
US7577754B2 (en) System and method for controlling access to content carried in a caching architecture
US7343422B2 (en) System and method for using uniform resource locators to map application layer content names to network layer anycast addresses
US8577992B1 (en) Request routing management based on network components
US8938526B1 (en) Request routing management based on network components
US20020004846A1 (en) System and method for using network layer uniform resource locator routing to locate the closest server carrying specific content
JP2014501958A (ja) ネットワークにおけるコンテンツにアクセスする方法および対応するシステム
WO2021057369A1 (zh) HttpDNS调度方法、装置、介质及设备
Wang et al. Evolution and challenges of DNS-based CDNs
US10536429B2 (en) Conveying information in hostname in a content delivery network (CDN)
US11095605B1 (en) Request routing utilizing encoded DNS-based messaging parameters
EP1277327B1 (en) System and method for using network layer uniform resource locator routing to locate the closest server carrying specific content
JP6487870B2 (ja) 名前解決装置、名前解決方法及び名前解決プログラム
CN114268605A (zh) 一种智能dns实现方法、装置及计算机存储介质
CN115701067A (zh) 切换用户端的边缘节点的方法、装置和***
CN117527809A (zh) 资源获取方法、装置、设备及存储介质
WO2001084802A2 (en) System and method for using uniform resource locators to map application layer content names to network layer anycast addresses
WO2001084803A2 (en) System and method for resolving network layer anycast addresses to network layer unicast addresses

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 19939974

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

ENP Entry into the national phase

Ref document number: 2019939974

Country of ref document: EP

Effective date: 20220216