(19)
(11)EP 3 547 649 B1

(12)EUROPEAN PATENT SPECIFICATION

(45)Mention of the grant of the patent:
25.03.2020 Bulletin 2020/13

(21)Application number: 18796356.6

(22)Date of filing:  28.02.2018
(51)International Patent Classification (IPC): 
H04L 29/08(2006.01)
(86)International application number:
PCT/CN2018/077554
(87)International publication number:
WO 2019/148568 (08.08.2019 Gazette  2019/32)

(54)

METHOD AND SYSTEM FOR SENDING REQUEST FOR ACQUIRING DATA RESOURCE

VERFAHREN UND SYSTEM ZUR ÜBERTRAGUNG VON DATENRESSOURCENERFASSUNGSANFRAGEN

PROCÉDÉ ET SYSTÈME POUR ENVOYER UNE DEMANDE D'ACQUISITION DE RESSOURCE DE DONNÉES


(84)Designated Contracting States:
AL AT BE BG CH CY CZ DE DK EE ES FI FR GB GR HR HU IE IS IT LI LT LU LV MC MK MT NL NO PL PT RO RS SE SI SK SM TR

(30)Priority: 02.02.2018 CN 201810106492

(43)Date of publication of application:
02.10.2019 Bulletin 2019/40

(73)Proprietor: Wangsu Science & Technology Co., Ltd.
Shanghai 200030 (CN)

(72)Inventors:
  • ZHANG, Yu
    Shanghai 200030 (CN)
  • DONG, Shujia
    Shanghai 200030 (CN)

(74)Representative: Hanna Moore + Curley 
Garryard House 25/26 Earlsfort Terrace
Dublin 2, D02 PX51
Dublin 2, D02 PX51 (IE)


(56)References cited: : 
EP-A1- 3 232 710
CN-A- 106 656 674
CN-A- 107 222 560
US-A1- 2017 366 448
CN-A- 104 980 416
CN-A- 106 685 762
US-A1- 2017 366 448
US-A1- 2018 026 938
  
      
    Note: Within nine months from the publication of the mention of the grant of the European patent, any person may give notice to the European Patent Office of opposition to the European patent granted. Notice of opposition shall be filed in a written reasoned statement. It shall not be deemed to have been filed until the opposition fee has been paid. (Art. 99(1) European Patent Convention).


    Description

    FIELD OF THE DISCLOSURE



    [0001] The present disclosure generally relates to the field of data transmission technology and, more particularly, to a method and a system for transmitting a data resource acquisition request.

    BACKGROUND



    [0002] As the Internet technology develops continuously, Content Delivery Network (CDN) services have also developed rapidly. The CDN service cluster includes a large number of node servers for storing data resources and accelerating network services. The node servers may include edge node servers (hereinafter referred to as "edge nodes") and parent node servers (hereinafter referred to as a "parent nodes", which include dynamic parent nodes and static parent nodes).

    [0003] When a user wants to acquire a specific data resource, the user may, through a terminal, transmit an acquisition request for the data resource to the CDN service cluster, such that an edge node in the CDN service cluster can receive the acquisition request. If the data resource is not stored locally, the edge node may select an optimal path for back-to-source processing to acquire the data resource. That is, the acquisition request for the data resource may be transmitted to a corresponding resource server through dynamic parent node(s) on the optimal path.

    [0004] When implementing the present disclosure, the inventors found at least following issues in the existing technologies:

    [0005] If at a certain moment, there exists a large number of back-to-source demands for data resources in the edge nodes and the corresponding optimal paths pass through the same dynamic parent node, such dynamic parent node may need to forward a large number of data resource acquisition requests at the same time. As a result, the back-to-source efficiency of the dynamic parent node may be reduced due to the continuous overload of the traffic, and the quality of the back-to-source acceleration service of the CDN service cluster becomes poorer. The document EP 3232710 A1 discloses a method, a device, and a system for content delivery network (CDN)-based mobile terminal traffic processing. The method includes: sending, by a mobile terminal, data information to a scheduling mechanism, wherein the scheduling mechanism is configured to determine addresses of first target service nodes based on the data information; receiving, by the mobile terminal, the addresses of the first target service nodes; determining, by the mobile terminal, the first target service nodes, the first target service nodes being service nodes corresponding to the addresses of the first target service nodes; determining, by the mobile terminal, service types of application traffic; and directing, by the mobile terminal, the application traffic to the first target service nodes based on the addresses of service nodes and the service types of the application traffic.

    BRIEF SUMMARY OF THE DISCLOSURE



    [0006] To solve the issues in the existing technologies, embodiments of the present disclosure provide a method and a system for transmitting a data resource acquisition request. The technical solutions are as follows:

    [0007] In a first aspect, a method for transmitting a data resource acquisition request is provided. The method includes:

    when there is a need to transmit an acquisition request for a first data resource, acquiring, by a first node, traffic scheduling information of a plurality of secondary nodes stored locally corresponding to a resource server to which the first data resource belongs, where the traffic scheduling information includes traffic-to-be-split domain name level and traffic-to-be-split ratio;

    selecting, by the first node, a target node among the plurality of secondary nodes based on the traffic scheduling information of the plurality of secondary nodes; and

    transmitting, by the first node, the acquisition request for the first data resource to the target node.



    [0008] Optionally, the method further includes:

    periodically acquiring, by a second node, historic traffic corresponding to each domain name level in a historic cycle, to predict latest traffic corresponding to each domain name level in a current cycle;

    determining, by the second node, traffic scheduling information of the second node based on a preset traffic-splitting threshold and the latest traffic corresponding to each domain name level.



    [0009] Optionally, the determining, by the second node, traffic scheduling information of the second node based on a preset traffic-splitting threshold and the latest traffic corresponding to each domain name level includes:

    adding, by the second node, the latest traffic corresponding to each domain name level starting from a highest domain name level;

    when an adding result is greater than the preset traffic-splitting threshold, stopping adding and determining, by the second node, a domain name level corresponding to lastly added latest traffic as the traffic-to-be-split domain name level;

    determining, by the second node, a ratio of a difference between the adding result and the preset traffic-splitting threshold to latest traffic corresponding to the traffic-to-be-split domain name level as the traffic-to-be-split ratio.



    [0010] Optionally, the method further includes:

    periodically transmitting, by the first node, a probing message to each of the plurality of secondary nodes;

    receiving, by the first node, a probing response fed back by each of the plurality of secondary nodes, and determining, by the first node, a network delay between the first node and each of the plurality of secondary nodes; and

    configuring, by the first node, a priority level for each of the plurality of secondary nodes based on the network delay, where a secondary node with a higher priority level corresponds to a smaller network delay.



    [0011] Optionally, the selecting, by the first node, the target node among the plurality of secondary nodes based on the traffic scheduling information of the plurality of secondary nodes includes:

    sequentially selecting, by the first node, a candidate node from the plurality of secondary nodes based on a descending order of priority levels; and

    when a domain name level of the resource server to which the first data resource belongs is equal to traffic-to-be-split domain name level included in the traffic scheduling information of the candidate node and the acquisition request of the first data resource does not satisfy preset traffic-splitting criteria, or when the domain name level of the resource server to which the first data resource belongs is greater than the traffic-to-be-split domain name level included in the traffic scheduling information of the candidate node, determining, by the first node, the candidate node to be the target node.



    [0012] Optionally, the probing response fed back by each of the plurality of secondary nodes includes traffic scheduling information carried thereby respectively, and
    after receiving, by the first node, the probing response fed back by each of the plurality of secondary nodes, the method further includes:
    updating, by the first node, traffic scheduling information of each of the plurality of secondary nodes stored locally based on traffic scheduling information carried in the probing response.

    [0013] Optionally, the method further includes:

    receiving, by the first node, a request response for a second data resource sent by a third node; and

    updating, by the first node, traffic scheduling information of the third node stored locally based on traffic scheduling information of the third node carried in the request response.



    [0014] In a second aspect, a system for transmitting a data resource acquisition request is provided. The system comprises a plurality of nodes, where the plurality of nodes includes a first node that is configured to:

    when there is a need to transmit an acquisition request for a first data source, acquire traffic scheduling information of a plurality of secondary nodes stored locally corresponding to a source server to which the first data source belongs, where the traffic scheduling information includes traffic-to-be-split domain name level and traffic-to-be-split ratio;

    select a target node among the plurality of secondary nodes based on the traffic scheduling information of the plurality of secondary nodes; and

    transmit the acquisition request for the first data source to the target node.



    [0015] Optionally, the plurality of nodes further includes a second node, and the second node is configured to:

    periodically acquire historic traffic corresponding to each domain name level in a historic cycle, to predict latest traffic corresponding to each domain name level in a current cycle; and

    determine traffic scheduling information of the second node based on a preset traffic-splitting threshold and the latest traffic corresponding to each domain name level.



    [0016] Optionally, the second node is specifically configured to:

    add latest traffic corresponding to each domain name level starting from a highest domain name level;

    when an adding result is greater than the preset traffic-splitting threshold, stop adding and determine a domain name level corresponding to lastly added latest traffic as the traffic-to-be-split domain name level;

    determine a ratio of a difference between the adding result and the preset traffic-splitting threshold to latest traffic corresponding to the traffic-to-be-split domain name level as the traffic-to-be-split ratio.



    [0017] Optionally, the first node is further configured to:

    periodically transmit a probing message to each of the plurality of secondary nodes;

    receive a probing response fed back by each of the plurality of secondary nodes, and determine a network delay between the first node and each of the plurality of secondary nodes; and

    configure a priority level for each of the plurality of secondary nodes based on the network delay, where a secondary node with a higher priority level corresponds to a smaller network delay.



    [0018] Optionally, the first node is specifically configured to:

    sequentially select a candidate node from the plurality of secondary nodes based on a descending order of priority levels; and

    when a domain name level of the resource server to which the first data resource belongs is equal to traffic-to-be-split domain name level included in the traffic scheduling information of the candidate node and the acquisition request of the first data resource does not satisfy preset traffic-splitting criteria, or the domain name level of the resource server to which the first data resource belongs is greater than the traffic-to-be-split domain name level included in the traffic scheduling information of the candidate node, determine the candidate node to be the target node.



    [0019] Optionally, the probing response fed back by each of the plurality of secondary nodes includes traffic scheduling information carried thereby respectively, and
    the first node is further configured to:
    based on traffic scheduling information carried in the probing response, update traffic scheduling information of each of the plurality of secondary nodes that is stored locally.

    [0020] Optionally, the plurality of nodes further includes a third node, and the first node is further configured to:

    receive a request response for a second data resource sent by a third node; and

    update traffic scheduling information of the third node stored locally based on traffic scheduling information of the third node carried in the request response.



    [0021] Beneficial effects brought about by technical solutions of the present disclosure are as follows.

    [0022] In the disclosed embodiments, when there is a need to transmit the acquisition request for the first data resource, the first node acquires traffic scheduling information of a plurality of secondary nodes stored locally corresponding to the resource server to which the first data resource belongs, where the traffic scheduling information is generated by nodes based on a local traffic loading condition; the first node selects a target node among the plurality of secondary nodes based on the traffic scheduling information of the plurality of secondary nodes; and the first node transmits the acquisition request for the first data resource to the target node. In this way, each mode may split traffic with a relatively low domain name level based on the traffic scheduling information of the secondary nodes, thereby ensuring that the traffic with a relatively high domain name level may obtain relatively satisfying back-to-source acceleration service. Accordingly, the occurrence of a reduced back-to-source efficiency of a dynamic parent node caused by the continuous overloading of the traffic may be avoid effectively, thereby improving the quality of the back-to-source acceleration service in the CDN service cluster.

    BRIEF DESCRIPTION OF THE DRAWINGS



    [0023] To make the technical solutions in the embodiments of the present disclosure clearer, a brief introduction of the accompanying drawings consistent with descriptions of the embodiments will be provided hereinafter. It is to be understood that the following described drawings are merely some embodiments of the present disclosure. Based on the accompanying drawings and without creative efforts, persons of ordinary skill in the art may derive other drawings.

    FIG. 1 is a schematic diagram of a network architecture of a CDN service cluster according to some embodiments of the present disclosure;

    FIG. 2 is a flowchart of a method for transmitting a data resource acquisition request according to some embodiments of the present disclosure; and

    FIG. 3 is a principle diagram showing traffic dividing of a transmitted data resource acquisition request according to some embodiments of the present disclosure.


    DETAILED DESCRIPTION



    [0024] To make the objectives, technical solutions, and advantages of the present disclosure clearer, specific embodiments of the present disclosure will be made in detail with reference to the accompanying drawings.

    [0025] Embodiments of the present disclosure provide a method for transmitting a data resource acquisition request. The method may be cooperatively implemented by a plurality of node servers in a CDN service cluster. The network architecture may be as shown in FIG. 1. The node server may include edge node servers (hereinafter referred to as "edge nodes") and multilevel parent node servers (hereinafter referred to as "parent nodes"). Further, an edge node is an ingress node for a user request in the CDN service cluster, and a parent node is a node having a deployment level of the CDN service cluster after the edge node. The parent nodes include static parent nodes for caching the data resources and dynamic parent nodes for providing the back-to-source acceleration service. In the disclosed embodiments, the static parent nodes are not considered in subsequent descriptions that relate to nodes. That is, all nodes in the disclosed embodiments refer to the node servers in the CDN service cluster, but do not include terminals and resource servers. The aforementioned node server may include a processor, a memory, and a transceiver. The processor may be configured to process the transmitted data resource acquisition requests to be described hereinafter. The memory may be configured to store data required and generated in the following processes. The transceiver may be configured to receive and transmit the data involved in the following processes. In certain situations, the functions of a node server may be implemented by a server group that includes a plurality of servers. In some embodiments, a node server is described as a single server for illustrative purposes, and rest situation are similar thereto, which is not repeated herein.

    [0026] Specific illustrations of the processing processes in FIG. 2 are provided hereinafter with reference to some embodiments, which include:

    [0027] Step 201: when there is a need to transmit an acquisition request for a first data resource, acquiring, by a first node, traffic scheduling information of a plurality of secondary nodes stored locally corresponding to a resource server to which the first data resource belongs.

    [0028] The first node may be any node among edge nodes and dynamic parent nodes that have secondary nodes.

    [0029] In one implementation, when providing the back-to-source acceleration service, a dynamic parent node may generate traffic scheduling information based on a local traffic loading condition, and feed back the traffic scheduling information to a higher level node in a predefined manner. After acquiring the traffic scheduling information of the dynamic parent node, the higher level node may store the acquired traffic scheduling information locally, such that a transmission path can be selected based on the traffic scheduling information when transmitting a data source acquisition request. Further, the dynamic parent node may divide domain names corresponding to each resource server into different levels based on an importance degree, thereby determining a plurality of domain name levels. The plurality of domain name levels may include an ordinary level, a V1 level, a V2 level, and a V3 level, which may be denoted by digits 0∼3. A higher domain name level indicates a higher importance degree. Accordingly, the traffic scheduling information may include traffic-to-be-split domain name level and a corresponding traffic-to-be-split ratio. Specifically, the traffic scheduling information may be recorded in a manner of "lv1:lv2", where lv1 is traffic-to-be-split domain name level, and lv2 is traffic-to-be-split ratio. For example, the traffic scheduling information "0:10" indicates that the 10% traffic of the ordinary-level is spit, and the traffic of V1∼V3 levels is not split. The traffic scheduling information "1:20" indicates that in the ordinary-level, traffic is fully spit, 20% traffic of the VI-level is split out, and the traffic of V2∼V3 levels is not split. The traffic scheduling information "3:50" indicates that the traffic of ordinary level∼V2 level is fully spit, and 50% traffic of the V3-level is split.

    [0030] As such, when the first node is an edge node, after the first node receives a user's acquisition request for a certain data resource (e.g., a first data resource), if the first data source is not stored locally, the first node needs to transmit the corresponding acquisition request through other nodes to the resource server to which the first data source belongs. The first node may first determine a plurality of secondary nodes corresponding to the resource server to which the first data source belongs, and acquire the traffic scheduling information of the plurality of secondary nodes stored locally. When the first node is a dynamic parent node, after receiving the acquisition request for the first data source sent by a higher level node, the first node needs to forward the acquisition request through other nodes to the resource server to which the first data source belongs. The first node may first determine the plurality of secondary nodes corresponding to the resource server to which the first data source belongs, and acquire the traffic scheduling information of a plurality of secondary nodes stored locally. It is to be understood that for each resource server, the technical staff of the CDN service cluster may configure corresponding secondary nodes corresponding to the resource server in each edge node and dynamic parent nodes. That is, the present node may transmit the data resource acquisition request to the resource server through any of these secondary nodes.

    [0031] Step 202: selecting, by the first node, a target node among the plurality of secondary nodes based on the traffic scheduling information of the plurality of secondary nodes.

    [0032] In one implementation, after acquiring the traffic scheduling information of the plurality of secondary nodes, the first node may select, based on the traffic scheduling information, a target node among the plurality of secondary nodes for forwarding the acquisition request for the first data resource.

    [0033] Step 203: transmitting, by the first node, the acquisition request for the first data resource to the target node.

    [0034] In one implementation, after selecting the target node, the first node may transmit the acquisition request for the first data resource to the target node.

    [0035] Optionally, a dynamic parent node may determine the traffic scheduling information based on the historical traffic of each domain name level, and corresponding processes are as follows: periodically acquiring, by the second node, historic traffic corresponding to each domain name level in a historic cycle, to predict latest traffic corresponding to each domain name level in a current cycle; and determining, by the second node, traffic scheduling information of the second node based on a preset traffic-splitting threshold and the latest traffic corresponding to each domain name level.

    [0036] In particular, the second node may be any of the dynamic parent nodes for providing the back-to-source acceleration service, or may be any secondary node of any node in the CDN service cluster.

    [0037] In one implementation, traffic server may be further configured in the service cluster. The traffic server is configured to record the actual traffic load status of each dynamic parent node, which includes the historical traffic corresponding to each domain name level at the dynamic parent node. When providing the back-to-source acceleration service, a dynamic parent node needs to periodically update the traffic scheduling information. Given the second node as an example, the second node may periodically acquire the local historical traffic corresponding to each domain name level in the historic cycle from traffic server, and based on such historical traffic, the second node may predict the latest traffic corresponding to each domain name level of a current cycle. Specifically, the least square fitting method or other feasible prediction algorithms may be used herein. Further, a second node may determine the traffic scheduling information of the second node in the current cycle based on the predicted latest total local traffic and the preset traffic-splitting threshold. It should be understood that, the preset traffic-splitting threshold may be traffic-dividing triggering value pre-configured by the technical staff of the CDN service cluster for each dynamic parent node, based on machine performances of each dynamic parent node. That is, when the total traffic in the current cycle exceeds the preset traffic-splitting threshold, the dynamic parent node may enter an overloading status, and the traffic needs to be split.

    [0038] Optionally, the dynamic parent node may give priority to traffic splitting corresponding to an acquisition request of a low domain name level. Correspondingly, the process of determining the traffic scheduling information may be as follows: adding, by the second node, the latest traffic corresponding to each domain name level starting from a highest domain name level; when an adding result is greater than the preset traffic-splitting threshold, stopping adding and determining, by the second node, a domain name level corresponding to lastly added latest traffic as the traffic-to-be-split domain name level; determining, by the second node, a ratio of a difference between the adding result and the preset traffic-splitting threshold to latest traffic corresponding to the traffic-to-be-split domain name level as the traffic-to-be-split ratio.

    [0039] In one implementation, during the process of generating the traffic scheduling information by the second node, the latest traffic corresponding to each domain name level may be added up starting from the highest domain name level. When the adding result is greater than the preset traffic-splitting threshold, the adding may be stopped and the last added domain name level may be determined to be the traffic-to-be-split domain name level. It is indicated that, when merely bearing all of the latest traffic both from traffic-to-be-split domain name level and higher domain name levels, the second node is already in an overloading status. Thus, partial traffic from the traffic-to-be-split domain name level and all latest traffic from the domain name levels lower than the traffic-to-be-split domain name level are split. As such, the second node may first calculate a difference value between the adding result and the preset traffic-splitting threshold (i.e., the amount of the partial traffic of the traffic-to-be-split domain name level), and further determine a ratio of such difference to the latest traffic corresponding to the traffic-to-be-split domain name level as the traffic-to-be-split ratio.

    [0040] Optionally, a node may rank the priorities for all secondary nodes. The corresponding process may be as follows: a first node periodically sends probing messages to all secondary nodes; the first node receives a probing response fed back by each secondary node, and determines a network delay between the first node and each secondary node; the first node configures the priority of each secondary node based on the network delay, where a secondary node with a higher priority level corresponds to a smaller network delay.

    [0041] In one implementation, a first node may periodically send probing messages to all pre-recorded secondary nodes. After receiving the probing message, a secondary node may feed back a corresponding probing response to the first node. Further, the first node may receive the probing response fed back by each secondary node, and determine the network delay between the first node and each secondary node based on the sending time of the probing message and the receiving time of each probing response. Further, the first node may determine a priority for each secondary node based on the aforementioned network delay. A secondary node with a higher priority level corresponds to a smaller network delay.

    [0042] Optionally, the first node may preferentially send an acquisition request to a secondary node with a small network delay. Correspondingly, the process of step 202 may be as follows: the first node sequentially selects a candidate node from the plurality of secondary nodes based on a descending order of priority levels; when a domain name level of the resource server to which the first data resource belongs is equal to the traffic-to-be-split domain name level in the traffic scheduling information of the candidate node and the acquisition request for the first data resource does not satisfy the preset traffic-splitting criteria, or when the domain name level of the resource server to which the first data resource belongs is greater than the traffic-to-be-split domain name level in the traffic scheduling information of the candidate node, the candidate node is determined as the target node.

    [0043] In one implementation, after acquiring the traffic scheduling information of a plurality of secondary nodes, based on the priorities of the plurality of secondary nodes, the first node may sequentially select a secondary node as a candidate node according to a descending order of the priorities. When the domain name level of the resource server to which the first data resource belongs is greater than the traffic-to-be-split domain name level in the traffic scheduling information of the candidate node, the candidate node may be determined as the target node. When the domain name level of the resource server to which the first data resource belongs is equal to the traffic-to-be-split domain name level in the traffic scheduling information of the candidate node, whether the acquisition request for the first data resource satisfies the preset traffic-splitting criteria is determined. If the acquisition request for the first data resource does not satisfy the preset traffic-splitting criteria, it means that there is no need to split the traffic for the acquisition request. Further, the candidate node may be selected as the target node. When the domain name level of the resource server to which the first data resource belongs is equal to the traffic-to-be-split domain name level in the traffic scheduling information of the candidate node and the acquisition request for the first data resource satisfies the preset traffic-splitting criteria, or when the domain name level of the resource server to which the first data resource belongs is lower than the traffic-to-be-split domain name level in the traffic scheduling information of the candidate node, it means the traffic for the acquisition request needs to be split. A next secondary node may be then selected as a candidate node. It should be noted that the foregoing preset traffic-splitting criteria may be arbitrary criteria for traffic division preset by the technical staff of the CDN service cluster. The criteria for traffic division for different secondary nodes may be the same or different. For example, a random number may be configured for each acquisition request, and if the random number is smaller than the traffic-to-be-split ratio in the traffic scheduling information, it is indicated that the corresponding acquisition request satisfies the preset traffic-splitting criteria.

    [0044] Optionally, a secondary node may send the traffic scheduling information to the first node through a probing response or through a request response for a data resource, which may be specifically as follows:
    Under situation 1, the first node updates the traffic scheduling information of each secondary node stored locally based on the traffic scheduling information carried in a probing response.

    [0045] In one implementation, after the first node periodically sends probing messages to all the secondary nodes, each secondary node may add the locally generated latest traffic scheduling information to the probing response, and feed the probing response back to the first node. In this way, after receiving the aforementioned probing response, the first node may update the traffic scheduling information of each secondary node stored locally based on traffic scheduling information carried in the probing response.

    [0046] Under situation 2, the first node receives a request response for a second data resource transmitted by a third node. The first node updates the traffic scheduling information of the third node stored locally based on traffic scheduling information of the third node carried in the request response.

    [0047] The third node may be any of the dynamic parent nodes for providing the back-to-source acceleration service, or any secondary node of any node in the CDN service cluster. The third node may be the same as or different from the aforementioned second node.

    [0048] In one implementation, if the first node successfully transmits, through the third node, an acquisition request for a second data resource to a certain resource server, after a certain period, the third node may receive a request response for the second data resource fed back by the resource server. The third node may add the locally generated latest traffic scheduling information to the request response and feed back the request response to the first node. As such, after the first node receives the request response of the second data resource transmitted by the third node, the traffic scheduling information of the third node stored locally may be updated based on the traffic scheduling information of the third node carried in the request response.

    [0049] For ease of understanding, FIG. 3 is a principle diagram showing traffic dividing according to some embodiments of the present disclosure. If the traffic of a plurality of domain name levels exceeds the preset traffic-splitting threshold, the traffic from relatively low domain name levels is split, to ensure the remaining traffic does not exceed the preset traffic-splitting threshold.

    [0050] In the disclosed embodiments, when there is a need to transmit the acquisition request for the first data resource, the first node acquires traffic scheduling information of a plurality of secondary nodes stored locally corresponding to the resource server to which the first data resource belongs, where the traffic scheduling information is generated by nodes based on the local traffic loading condition; the first node selects a target node among the plurality of secondary nodes based on the traffic scheduling information of the plurality of secondary nodes; and the first node transmits the acquisition request for the first data resource to the target node. As such, each node may split traffic of relatively low domain name levels based on the traffic scheduling information of the secondary nodes, thereby ensuring that the traffic of relatively high domain name levels receives relatively a satisfying back-to-source acceleration service. Accordingly, the occurrence of a reduced back-to-source efficiency of a dynamic parent node caused by the continuous overloading of the traffic may be avoid effectively, thereby improving the quality of the back-to-source acceleration service in the CDN service cluster.

    [0051] Based on the similar technical concepts, the present disclosure further provides a system for transmitting a data source acquisition request. The system comprises a plurality of nodes, where the plurality of nodes includes a first node. The first node is configured to:

    when there is a need to transmit an acquisition request for a first data source, acquire traffic scheduling information of a plurality of secondary nodes stored locally corresponding to a source server to which the first data source belongs, where the traffic scheduling information includes traffic-to-be-split domain name level and traffic-to-be-split ratio;

    select a target node among the plurality of secondary nodes based on the traffic scheduling information of the plurality of secondary nodes; and

    transmit the acquisition request for the first data source to the target node.



    [0052] Optionally, the plurality of nodes further includes a second node. The second node is configured to:

    periodically acquire historic traffic corresponding to each domain name level in a historic cycle, to predict latest traffic corresponding to each domain name level in a current cycle; and

    determine traffic scheduling information of the second node based on a preset traffic-splitting threshold and the latest traffic corresponding to each domain name level.



    [0053] Optionally, the second node is specifically configured to:

    add latest traffic corresponding to each domain name level starting from a highest domain name level;

    when an adding result is greater than the preset traffic-splitting threshold, stop adding and determine a domain name level corresponding to lastly added latest traffic as the traffic-to-be-split domain name level;

    determine a ratio of a difference between the adding result and the preset traffic-splitting threshold to latest traffic corresponding to the traffic-to-be-split domain name level as the traffic-to-be-split ratio.



    [0054] Optionally, the first node is further configured to:

    periodically transmit a probing message to each of the plurality of secondary nodes;

    receive a probing response fed back by each of the plurality of secondary nodes, and determine a network delay between the first node and each of the plurality of secondary nodes; and

    configure a priority level for each of the plurality of secondary nodes based on the network delay, where a secondary node with a higher priority level corresponds to a smaller network delay.



    [0055] Optionally, the first node is specifically configured to:

    sequentially select a candidate node from the plurality of secondary nodes based on a descending order of priority levels; and

    when a domain name level of the resource server to which the first data resource belongs is equal to traffic-to-be-split domain name level included in the traffic scheduling information of the candidate node and the acquisition request of the first data resource does not satisfy preset traffic-splitting criteria, or when the domain name level of the resource server to which the first data resource belongs is greater than the traffic-to-be-split domain name level included in the traffic scheduling information of the candidate node, determine the candidate node as the target node.



    [0056] Optionally, the probing response fed back by each of the plurality of secondary nodes includes traffic scheduling information carried thereby respectively, and
    the first node is further configured to:
    update traffic scheduling information of each of the plurality of secondary nodes that is stored locally based on traffic scheduling information carried in the probing response.

    [0057] Optionally, the plurality of nodes further includes a third node, and the first node is further configured to:

    receive a request response for a second data resource transmitted by a third node; and

    update traffic scheduling information of the third node that is stored locally based on the traffic scheduling information of the third node carried in the request response.



    [0058] In the disclosed embodiments, when there is a need to transmit the acquisition request for the first data resource, the first node acquires traffic scheduling information of a plurality of secondary nodes stored locally corresponding to the resource server to which the first data resource belongs, where the traffic scheduling information is generated based on the local traffic loading condition; the first node selects a target node among the plurality of secondary nodes based on the traffic scheduling information of the plurality of secondary nodes; and the first node transmits the acquisition request for the first data resource to the target node. As such, each node may split traffic of relatively low domain name levels based on the traffic scheduling information of the secondary nodes, thereby ensuring that the traffic of relatively high domain name levels receives relatively a satisfying back-to-source acceleration service. Accordingly, the occurrence of a reduced back-to-source efficiency of a dynamic parent node caused by the continuous overloading of the traffic may be avoid effectively, thereby improving the quality of the back-to-source acceleration service in the CDN service cluster.

    [0059] It is clear to those skilled in the art that the entire or partial steps of the disclosed embodiments may be implemented in the form of hardware or may be implemented in the form of software programs that instruct relevant hardware components. The software programs may be stored in a computer-readable storage medium. The storage medium may be a read-only memory, a magnetic disk, or an optic disk, etc.

    [0060] Although the present disclosure has been described as above with reference to preferred embodiments, these embodiments are not constructed as limiting the present disclosure. Any modifications, equivalent replacements, and improvements made shall fall within the scope of the protection of the present disclosure.


    Claims

    1. A method for transmitting a data resource acquisition request, the method comprising:

    when (201) there is a need to transmit an acquisition request for a first data source, acquiring, by a first node, traffic scheduling information of a plurality of secondary nodes stored locally corresponding to a source server to which the first data source belongs, wherein the traffic scheduling information includes traffic-to-be-split domain name level and traffic-to-be-split ratio;

    selecting (202), by the first node, a target node among the plurality of secondary nodes based on the traffic scheduling information of the plurality of secondary nodes; and

    transmitting (203), by the first node, the acquisition request for the first data source to the target node.


     
    2. The method according to claim 1, further comprising:

    periodically acquiring, by a second node, historic traffic corresponding to each domain name level in a historic cycle, to predict latest traffic corresponding to each domain name level in a current cycle;

    determining, by the second node, traffic scheduling information of the second node based on a preset traffic-splitting threshold and latest traffic corresponding to each domain name level.


     
    3. The method according to claim 2, wherein the determining, by the second node, traffic scheduling information of the second node based on a preset traffic-splitting threshold and latest traffic corresponding to each domain name level includes:

    adding, by the second node, latest traffic corresponding to each domain name level starting from a highest domain name level;

    when an adding result is greater than the preset traffic-splitting threshold, stopping adding and determining, by the second node, a domain name level corresponding to lastly added latest traffic as the traffic-to-be-split domain name level;

    determining, by the second node, a ratio of a difference between the adding result and the preset traffic-splitting threshold to latest traffic corresponding to the traffic-to-be-split domain name level as the traffic-to-be-split ratio.


     
    4. The method according to claim 1, further comprising:

    periodically sending, by the first node, a probing message to each of the plurality of secondary nodes;

    receiving, by the first node, a probing response fed back by each of the plurality of secondary nodes, and determining, by the first node, a network delay between the first node and each of the plurality of secondary nodes; and

    configuring, by the first node, a priority level for each of the plurality of secondary nodes based on the network delay, wherein a secondary node with a higher priority corresponds to a smaller network delay.


     
    5. The method according to claim 4, wherein the selecting, by the first node, the target node among the plurality of secondary nodes based on the traffic scheduling information of the plurality of secondary nodes includes:

    sequentially selecting, by the first node, a candidate node from the plurality of secondary nodes based on a descending order of priority levels; and

    when a domain name level of the resource server to which the first data resource belongs is equal to traffic-to-be-split domain name level included in the traffic scheduling information of the candidate node and the acquisition request of the first data resource does not satisfy preset traffic-splitting criteria, or when the domain name level of the resource server to which the first data resource belongs is greater than the traffic-to-be-split domain name level included in the traffic scheduling information of the candidate node, determining, by the first node, the candidate node to be the target node.


     
    6. The method according to claim 4, wherein the probing response fed back by each of the plurality of secondary nodes includes traffic scheduling information carried thereby respectively, and after receiving, by the first node, the probing response fed back by each of the plurality of secondary nodes, the method further includes:
    updating, by the first node, traffic scheduling information of each of the plurality of secondary nodes based on traffic scheduling information carried in the probing response.
     
    7. The method according to claim 1, further comprising:

    receiving, by the first node, a request response for a second data resource sent by a third node; and

    updating, by the first node, traffic scheduling information of the third node stored locally based on traffic scheduling information of the third node carried in the request response.


     
    8. A system for transmitting a data source acquisition request, the system comprising a plurality of nodes, wherein the plurality of nodes includes a first node and the system configured to:

    when (201) there is a need to transmit an acquisition request for a first data source, acquire traffic scheduling information of a plurality of secondary nodes stored locally corresponding to a source server to which the first data source belongs, wherein the traffic scheduling information includes traffic-to-be-split domain name level and traffic-to-be-split ratio;

    select (202) a target node among the plurality of secondary nodes based on the traffic scheduling information of the plurality of secondary nodes; and

    transmit (203) the acquisition request for the first data source to the target node.


     
    9. The system according to claim 8, wherein the plurality of nodes further includes a second node, and the second node is configured to:

    periodically acquire historic traffic corresponding to each domain name level in a historic cycle, to predict latest traffic corresponding to each domain name level in a current cycle;

    determine traffic scheduling information of the second node based on a preset traffic-splitting threshold and the latest traffic corresponding to each domain name level.


     
    10. The system according to claim 9, wherein the second node is specifically configured to:

    add the latest traffic corresponding to each domain name level starting from a highest domain name level;

    when an adding result is greater than the preset traffic-splitting threshold, stop adding and determine a domain name level corresponding to lastly added latest traffic as the traffic-to-be-split domain name level;

    determine a ratio of a difference between the adding result and the preset traffic-splitting threshold to latest traffic corresponding to the traffic-to-be-split domain name level as the traffic-to-be-split ratio.


     
    11. The system according to claim 8, wherein the first node is further configured to:

    periodically send a probing message to each of the plurality of secondary nodes;

    receive a probing response fed back by each of the plurality of secondary nodes, and determine a network delay between the first node and each of the plurality of secondary nodes; and

    determine a priority level for each of the plurality of secondary nodes based on the network delay, wherein a secondary node with a higher priority level corresponds to a smaller network delay.


     
    12. The system according to claim 11, wherein the first node is specifically configured to:

    sequentially select a candidate node from the plurality of secondary nodes based on a descending order of priority levels; and

    when a domain name level of the resource server to which the first data resource belongs is equal to traffic-to-be-split domain name level included in the traffic scheduling information of the candidate node and the acquisition request of the first data resource does not satisfy preset traffic-splitting criteria, or the domain name level of the resource server to which the first data resource belongs is greater than the traffic-to-be-split domain name level included in the traffic scheduling information of the candidate node, determine the candidate node to be the target node.


     
    13. The system according to claim 11, wherein the probing response fed back by each of the plurality of secondary nodes includes traffic scheduling information carried thereby respectively, and the first node is further configured to:
    update traffic scheduling information of each of the plurality of secondary nodes based on traffic scheduling information carried in the probing response.
     
    14. The system according to claim 8, wherein the plurality of nodes further includes a third node, and the first node is further configured to:

    receive a request response for a second data resource transmitted by a third node; and

    update traffic scheduling information of the third node stored locally based on traffic scheduling information of the third node carried in the request response.


     


    Ansprüche

    1. Verfahren zur Übertragung einer Datenressourcenbezugsanforderung, wobei das Verfahren umfasst:

    wenn (201) Bedarf daran besteht, eine Bezugsanforderung für eine erste Datenquelle zu übertragen, Beziehen von Verkehrsplanungsinformationen einer Vielzahl von sekundären Knoten, die lokal gespeichert sind, entsprechend einem Quellenserver, zu dem die erste Datenquelle gehört, durch einen ersten Knoten, wobei die Verkehrsplanungsinformationen eine Aufzuteilender-Verkehr-Domänennamensebene und ein Aufzuteilender-Verkehr-Verhältnis beinhalten;

    Auswählen (202) eines Zielknotens aus der Vielzahl von sekundären Knoten auf der Basis der Verkehrsplanungsinformationen der Vielzahl von sekundären Knoten durch den ersten Knoten und

    Übertragen (203) der Bezugsanforderung für die erste Datenquelle an den Zielknoten durch den ersten Knoten.


     
    2. Verfahren nach Anspruch 1, weiterhin umfassend:

    periodisches Beziehen von historischem Verkehr entsprechend jeder Domänennamensebene in einem historischen Zyklus durch einen zweiten Knoten, um den jüngsten Verkehr entsprechend jeder Domänennamensebene in einem aktuellen Zyklus vorherzusagen;

    Bestimmen von Verkehrsplanungsinformationen des zweiten Knotens auf der Basis eines voreingestellten Verkehrsaufteilungsgrenzwerts und des jüngsten Verkehrs entsprechend jeder Domänennamensebene durch den zweiten Knoten.


     
    3. Verfahren nach Anspruch 2, wobei das Bestimmen von Verkehrsplanungsinformationen des zweiten Knotens auf der Basis eines voreingestellten Verkehrsaufteilungsgrenzwerts und dem jüngsten Verkehr entsprechend jeder Domänennamensebene durch den zweiten Knoten beinhaltet:

    Hinzufügen des jüngsten Verkehrs entsprechend jeder Domänennamensebene ausgehend von einer höchsten Domänennamensebene durch den zweiten Knoten;

    wenn ein Hinzufügungsergebnis höher als der voreingestellte Verkehrsaufteilungsgrenzwert ist, Stoppen des Hinzufügens und Bestimmen einer Domänennamensebene entsprechend dem zuletzt hinzugefügten jüngsten Verkehr als die Aufzuteilender-Verkehr-Domänennamensebene durch den zweiten Knoten;

    Bestimmen eines Verhältnisses eines Unterschieds zwischen dem Hinzufügungsergebnis und dem voreingestellten Verkehrsaufteilungsgrenzwert zu dem jüngsten Verkehr entsprechend der Aufzuteilender-Verkehr-Domänennamensebene als das Aufzuteilender-Verkehr-Verhältnis durch den zweiten Knoten.


     
    4. Verfahren nach Anspruch 1, weiterhin umfassend:

    periodisches Senden einer Sondierungsnachricht an jeden der Vielzahl von sekundären Knoten durch den ersten Knoten;

    Empfangen einer Sondierungsantwort, die durch jeden der Vielzahl von sekundären Knoten zurückgesendet wird, durch den ersten Knoten und Bestimmen einer Netzwerkverzögerung zwischen dem ersten Knoten und jedem der Vielzahl von sekundären Knoten durch den ersten Knoten und

    Konfigurieren eines Prioritätsniveaus für jeden der Vielzahl von sekundären Knoten auf der Basis der Netzwerkverzögerung durch den ersten Knoten, wobei ein sekundärer Knoten mit einer höheren Priorität einer geringeren Netzwerkverzögerung entspricht.


     
    5. Verfahren nach Anspruch 4, wobei das Auswählen des Zielknotens aus der Vielzahl von sekundären Knoten auf der Basis der Verkehrsplanungsinformationen der Vielzahl von sekundären Knoten durch den ersten Knoten beinhaltet:

    sequentielles Auswählen eines Kandidatenknotens aus der Vielzahl von sekundären Knoten auf der Basis einer absteigenden Reihenfolge von Prioritätsniveaus durch den ersten Knoten; und

    wenn eine Domänennamensebene des Ressourcenservers, zu dem die erste Datenressource gehört, gleich einer Aufzuteilender-Verkehr-Domänennamensebene ist, die in den Verkehrsplanungsinformationen des Kandidatenknotens enthalten ist, und die Bezugsanforderung der ersten Datenressource voreingestellte Verkehrsaufteilungskriterien nicht erfüllt, oder wenn die Domänennamensebene des Ressourcenservers, zu dem die erste Datenressource gehört, höher als die Aufzuteilender-Verkehr-Domänennamensebene ist, die in den Verkehrsplanungsinformationen des Kandidatenknotens enthalten ist, Bestimmen des Kandidatenknotens als der Zielknoten durch den ersten Knoten.


     
    6. Verfahren nach Anspruch 4, wobei die Sondierungsantwort, die durch jeden der Vielzahl von sekundären Knoten zurückgesendet wird, Verkehrsplanungsinformationen beinhaltet, die jeweils dadurch getragen werden, und nach Empfangen der Sondierungsantwort, die durch jeden der Vielzahl von sekundären Knoten zurückgesendet wird, durch den ersten Knoten das Verfahren weiterhin beinhaltet:
    Aktualisieren von Verkehrsplanungsinformationen jedes der Vielzahl von sekundären Knoten auf der Basis von Verkehrsplanungsinformationen, die in der Sondierungsantwort getragen werden, durch den ersten Knoten.
     
    7. Verfahren nach Anspruch 1, weiterhin umfassend:

    Empfangen einer Anforderungsantwort für eine zweite Datenressource, die von einem dritten Knoten gesendet wird, durch den ersten Knoten und

    Aktualisieren von Verkehrsplanungsinformationen des dritten Knotens, die lokal gespeichert sind, auf der Basis von Verkehrsplanungsinformationen des dritten Knotens, die in der Anforderungsantwort getragen werden, durch den ersten Knoten.


     
    8. System zur Übertragung einer Datenressourcenbezugsanforderung, wobei das System eine Vielzahl von Knoten umfasst, wobei die Vielzahl von Knoten einen ersten Knoten beinhaltet und das System konfiguriert ist zum:

    wenn (201) Bedarf daran besteht, eine Bezugsanforderung für eine erste Datenquelle zu übertragen, Beziehen von Verkehrsplanungsinformationen einer Vielzahl von sekundären Knoten, die lokal gespeichert sind, entsprechend einem Quellenserver, zu dem die erste Datenquelle gehört, wobei die Verkehrsplanungsinformationen eine Aufzuteilender-Verkehr-Domänennamensebene und ein Aufzuteilender-Verkehr-Verhältnis beinhalten;

    Auswählen (202) eines Zielknotens aus der Vielzahl von sekundären Knoten auf der Basis der Verkehrsplanungsinformationen der Vielzahl von sekundären Knoten und

    Übertragen (203) der Bezugsanforderung für die erste Datenquelle an den Zielknoten.


     
    9. System nach Anspruch 8, wobei die Vielzahl von Knoten weiterhin einen zweiten Knoten beinhaltet und der zweite Knoten konfiguriert ist zum:

    periodischen Beziehen von historischem Verkehr entsprechend jeder Domänennamensebene in einem historischen Zyklus, um den jüngsten Verkehr entsprechend jeder Domänennamensebene in einem aktuellen Zyklus vorherzusagen;

    Bestimmen von Verkehrsplanungsinformationen des zweiten Knotens auf der Basis eines voreingestellten Verkehrsaufteilungsgrenzwerts und des jüngsten Verkehrs entsprechend jeder Domänennamensebene.


     
    10. System nach Anspruch 9, wobei der zweite Knoten spezifisch konfiguriert ist zum:

    Hinzufügen des jüngsten Verkehrs entsprechend jeder Domänennamensebene ausgehend von einer höchsten Domänennamensebene;

    wenn ein Hinzufügungsergebnis höher als der voreingestellte Verkehrsaufteilungsgrenzwert ist, Stoppen des Hinzufügens und Bestimmen einer Domänennamensebene entsprechend dem zuletzt hinzugefügten jüngsten Verkehr als die Aufzuteilender-Verkehr-Domänennamensebene;

    Bestimmen eines Verhältnisses eines Unterschieds zwischen dem Hinzufügungsergebnis und dem voreingestellten Verkehrsaufteilungsgrenzwert zu dem jüngsten Verkehr entsprechend der Aufzuteilender-Verkehr-Domänennamensebene als das Aufzuteilender-Verkehr-Verhältnis.


     
    11. System nach Anspruch 8, wobei der erste Knoten weiterhin konfiguriert ist zum:

    periodischen Senden einer Sondierungsnachricht an jeden der Vielzahl von sekundären Knoten;

    Empfangen einer Sondierungsantwort, die durch jeden der Vielzahl von sekundären Knoten zurückgesendet wird, und Bestimmen einer Netzwerkverzögerung zwischen dem ersten Knoten und jedem der Vielzahl von sekundären Knoten und

    Bestimmen eines Prioritätsniveaus für jeden der Vielzahl von sekundären Knoten auf der Basis der Netzwerkverzögerung, wobei ein sekundärer Knoten mit einem höheren Prioritätsniveau einer geringeren Netzwerkverzögerung entspricht.


     
    12. System nach Anspruch 11, wobei der erste Knoten spezifisch konfiguriert ist zum:

    sequentiellen Auswählen eines Kandidatenknotens aus der Vielzahl von sekundären Knoten auf der Basis einer absteigenden Reihenfolge von Prioritätsniveaus; und

    wenn eine Domänennamensebene des Ressourcenservers, zu dem die erste Datenressource gehört, gleich einer Aufzuteilender-Verkehr-Domänennamensebene ist, die in den Verkehrsplanungsinformationen des Kandidatenknotens enthalten ist, und die Bezugsanforderung der ersten Datenressource voreingestellte Verkehrsaufteilungskriterien nicht erfüllt, oder die Domänennamensebene des Ressourcenservers, zu dem die erste Datenressource gehört, höher als die Aufzuteilender-Verkehr-Domänennamensebene ist, die in den Verkehrsplanungsinformationen des Kandidatenknotens enthalten ist, Bestimmen des Kandidatenknotens als der Zielknoten.


     
    13. System nach Anspruch 11, wobei die Sondierungsantwort, die durch jeden der Vielzahl von sekundären Knoten zurückgesendet wird, Verkehrsplanungsinformationen beinhaltet, die jeweils dadurch getragen werden, und der erste Knoten weiterhin konfiguriert ist zum:
    Aktualisieren von Verkehrsplanungsinformationen jedes der Vielzahl von sekundären Knoten auf der Basis von Verkehrsplanungsinformationen, die in der Sondierungsantwort getragen werden.
     
    14. System nach Anspruch 8k, wobei die Vielzahl von Knoten weiterhin einen dritten Knoten beinhaltet und der erste Knoten weiterhin konfiguriert ist zum:

    Empfangen einer Anforderungsantwort für eine zweite Datenressource, die durch den dritten Knoten übertragen wird; und

    Aktualisieren von Verkehrsplanungsinformationen des dritten Knotens, die lokal gespeichert sind, auf der Basis von Verkehrsplanungsinformationen des dritten Knotens, die in der Anforderungsantwort getragen werden.


     


    Revendications

    1. Procédé de transmission d'une demande d'acquisition de ressource de données, le procédé comprenant :

    lorsque (201) il est nécessaire de transmettre une demande d'acquisition pour une première source de données, acquérir, par un premier nœud, des informations de planification de trafic d'une pluralité de nœuds secondaires stockées localement correspondant à un serveur source auquel la première source de données appartient, les informations de planification de trafic comprenant un niveau de nom de domaine de trafic à diviser et un rapport de trafic à diviser ;

    sélectionner (202), par le premier nœud, un nœud cible parmi la pluralité de nœuds secondaires sur la base des informations de planification de trafic de la pluralité de nœuds secondaires ; et

    transmettre (203), par le premier nœud, la demande d'acquisition pour la première source de données au nœud cible.


     
    2. Procédé selon la revendication 1, comprenant en outre :

    acquérir périodiquement, par un second nœud, un trafic historique correspondant à chaque niveau de nom de domaine dans un cycle historique, pour prédire un dernier trafic correspondant à chaque niveau de nom de domaine dans un cycle actuel ;

    déterminer, par le second nœud, des informations de planification de trafic du second nœud sur la base d'un seuil de division de trafic prédéfini et d'un dernier trafic correspondant à chaque niveau de nom de domaine.


     
    3. Procédé selon la revendication 2, dans lequel la détermination, par le second nœud, d'informations de planification de trafic du second nœud sur la base d'un seuil de division de trafic prédéfini et d'un dernier trafic correspondant à chaque niveau de nom de domaine comprend :

    ajouter, par le second nœud, un dernier trafic correspondant à chaque niveau de nom de domaine à partir d'un niveau de nom de domaine le plus élevé ;

    lorsqu'un résultat d'ajout est supérieur au seuil de division de trafic prédéfini, arrêter l'ajout et déterminer, par le second nœud, un niveau de nom de domaine correspondant à un dernier trafic ajouté en dernier en tant que niveau de nom de domaine de trafic à diviser ;

    déterminer, par le second nœud, un rapport d'une différence entre le résultat d'ajout et le seuil de division de trafic prédéfini sur un dernier trafic correspondant au niveau de nom de domaine de trafic à diviser en tant que rapport de trafic à diviser.


     
    4. Procédé selon la revendication 1, comprenant en outre :

    envoyer périodiquement, par le premier nœud, un message de sondage à chacun de la pluralité de nœuds secondaires ;

    recevoir, par le premier nœud, une réponse de sondage renvoyée par chacun de la pluralité de nœuds secondaires, et déterminer, par le premier nœud, un retard de réseau entre le premier nœud et chacun de la pluralité de nœuds secondaires ; et

    configurer, par le premier nœud, un niveau de priorité pour chacun de la pluralité de nœuds secondaires sur la base du retard de réseau, un nœud secondaire avec une priorité plus élevée correspondant à un retard de réseau plus faible.


     
    5. Procédé selon la revendication 4, dans lequel la sélection, par le premier nœud, du nœud cible parmi la pluralité de nœuds secondaires sur la base des informations de planification de trafic de la pluralité de nœuds secondaires comprend :

    sélectionner de manière séquentielle, par le premier nœud, un nœud candidat à partir de la pluralité de nœuds secondaires sur la base d'un ordre décroissant de niveaux de priorité ; et

    lorsqu'un niveau de nom de domaine du serveur de ressource auquel la première ressource de données appartient est égal à un niveau de nom de domaine de trafic à diviser inclus dans les informations de planification de trafic du nœud candidat et que la demande d'acquisition de la première ressource de données ne satisfait pas des critères de division de trafic prédéfinis, ou lorsque le niveau de nom de domaine du serveur de ressource auquel la première ressource de données appartient est supérieur au niveau de nom de domaine de trafic à diviser inclus dans les informations de planification de trafic du nœud candidat, déterminer, par le premier nœud, le nœud candidat comme étant le nœud cible.


     
    6. Procédé selon la revendication 4, dans lequel la réponse de sondage renvoyée par chacun de la pluralité de nœuds secondaires comprend des informations de planification de trafic transportées par celle-ci respectivement, et après la réception, par le premier nœud, de la réponse de sondage renvoyée par chacun de la pluralité de nœuds secondaires, le procédé comprend en outre :
    mettre à jour, par le premier nœud, des informations de planification de trafic de chacun de la pluralité de nœuds secondaires sur la base des informations de planification de trafic transportées dans la réponse de sondage.
     
    7. Procédé selon la revendication 1, comprenant en outre :

    recevoir, par le premier nœud, une réponse de demande pour une seconde ressource de données envoyée par un troisième nœud ; et

    mettre à jour, par le premier nœud, des informations de planification de trafic du troisième nœud stockées localement sur la base d'informations de planification de trafic du troisième nœud transportées dans la réponse de demande.


     
    8. Système de transmission d'une demande d'acquisition de source de données, le système comprenant une pluralité de nœuds, dans lequel la pluralité de nœuds comprend un premier nœud et le système est configuré pour :

    lorsque (201) il est nécessaire de transmettre une demande d'acquisition pour une première source de données, acquérir des informations de planification de trafic d'une pluralité de nœuds secondaires stockées localement correspondant à un serveur source auquel la première source de données appartient, les informations de planification de trafic comprenant un niveau de nom de domaine de trafic à diviser et un rapport de trafic à diviser ;

    sélectionner (202) un nœud cible parmi la pluralité de nœuds secondaires sur la base des informations de planification de trafic de la pluralité de nœuds secondaires ; et

    transmettre (203) la demande d'acquisition pour la première source de données au nœud cible.


     
    9. Système selon la revendication 8, dans lequel la pluralité de nœuds comprend en outre un second nœud, et le second nœud est configuré pour :

    acquérir périodiquement un trafic historique correspondant à chaque niveau de nom de domaine dans un cycle historique, pour prédire un dernier trafic correspondant à chaque niveau de nom de domaine dans un cycle actuel ;

    déterminer des informations de planification de trafic du second nœud sur la base d'un seuil de division de trafic prédéfini et du dernier trafic correspondant à chaque niveau de nom de domaine.


     
    10. Système selon la revendication 9, dans lequel le second nœud est configuré spécifiquement pour :

    ajouter le dernier trafic correspondant à chaque niveau de nom de domaine à partir d'un niveau de nom de domaine le plus élevé ;

    lorsqu'un résultat d'ajout est supérieur au seuil de division de trafic prédéfini, arrêter l'ajout et déterminer un niveau de nom de domaine correspondant à un dernier trafic ajouté en dernier en tant que niveau de nom de domaine de trafic à diviser ;

    déterminer un rapport d'une différence entre le résultat d'ajout et le seuil de division de trafic prédéfini sur un dernier trafic correspondant au niveau de nom de domaine de trafic à diviser en tant que rapport de trafic à diviser.


     
    11. Système selon la revendication 8, dans lequel le premier nœud est en outre configuré pour :

    envoyer périodiquement un message de sondage à chacun de la pluralité de nœuds secondaires ;

    recevoir une réponse de sondage renvoyée par chacun de la pluralité de nœuds secondaires, et déterminer un retard de réseau entre le premier nœud et chacun de la pluralité de nœuds secondaires ; et

    déterminer un niveau de priorité pour chacun de la pluralité de nœuds secondaires sur la base du retard de réseau, un nœud secondaire avec un niveau de priorité plus élevé correspondant à un retard de réseau plus faible.


     
    12. Système selon la revendication 11, dans lequel le premier nœud est configuré spécifiquement pour :

    sélectionner de manière séquentielle un nœud candidat à partir de la pluralité de nœuds secondaires sur la base d'un ordre décroissant de niveaux de priorité ; et

    lorsqu'un niveau de nom de domaine du serveur de ressource auquel la première ressource de données appartient est égal à un niveau de nom de domaine de trafic à diviser inclus dans les informations de planification de trafic du nœud candidat et que la demande d'acquisition de la première ressource de données ne satisfait pas des critères de division de trafic prédéfinis, ou que le niveau de nom de domaine du serveur de ressource auquel la première ressource de données appartient est supérieur au niveau de nom de domaine de trafic à diviser inclus dans les informations de planification de trafic du nœud candidat, déterminer le nœud candidat comme étant le nœud cible.


     
    13. Système selon la revendication 11, dans lequel la réponse de sondage renvoyée par chacun de la pluralité de nœuds secondaires comprend des informations de planification de trafic transportées par celle-ci respectivement, et le premier nœud est en outre configuré pour :
    mettre à jour des informations de planification de trafic de chacun de la pluralité de nœuds secondaires sur la base d'informations de planification de trafic transportées dans la réponse de sondage.
     
    14. Système selon la revendication 8, dans lequel la pluralité de nœuds comprend en outre un troisième nœud, et le premier nœud est en outre configuré pour :

    recevoir une réponse de demande pour une seconde ressource de données transmise par un troisième nœud ; et

    mettre à jour des informations de planification de trafic du troisième nœud stockées localement sur la base d'informations de planification de trafic du troisième nœud transportées dans la réponse de demande.


     




    Drawing











    Cited references

    REFERENCES CITED IN THE DESCRIPTION



    This list of references cited by the applicant is for the reader's convenience only. It does not form part of the European patent document. Even though great care has been taken in compiling the references, errors or omissions cannot be excluded and the EPO disclaims all liability in this regard.

    Patent documents cited in the description