WO2018209734A1 - 一种网络流量动态调度分配方法及系统 - Google Patents

一种网络流量动态调度分配方法及系统 Download PDF

Info

Publication number
WO2018209734A1
WO2018209734A1 PCT/CN2017/086805 CN2017086805W WO2018209734A1 WO 2018209734 A1 WO2018209734 A1 WO 2018209734A1 CN 2017086805 W CN2017086805 W CN 2017086805W WO 2018209734 A1 WO2018209734 A1 WO 2018209734A1
Authority
WO
WIPO (PCT)
Prior art keywords
service
node
nodes
domain name
record
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Ceased
Application number
PCT/CN2017/086805
Other languages
English (en)
French (fr)
Inventor
蔡镇河
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Wangsu Science and Technology Co Ltd
Original Assignee
Wangsu Science and Technology Co Ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Wangsu Science and Technology Co Ltd filed Critical Wangsu Science and Technology Co Ltd
Priority to US16/321,870 priority Critical patent/US10757037B2/en
Priority to EP17909729.0A priority patent/EP3522466B1/en
Publication of WO2018209734A1 publication Critical patent/WO2018209734A1/zh
Anticipated expiration legal-status Critical
Ceased legal-status Critical Current

Links

Images

Classifications

    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04LTRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L61/00Network arrangements, protocols or services for addressing or naming
    • H04L61/45Network directories; Name-to-address mapping
    • H04L61/4505Network directories; Name-to-address mapping using standardised directories; using standardised directory access protocols
    • H04L61/4511Network directories; Name-to-address mapping using standardised directories; using standardised directory access protocols using domain name system [DNS]
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04LTRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L47/00Traffic control in data switching networks
    • H04L47/10Flow control; Congestion control
    • H04L47/20Traffic policing
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04LTRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L45/00Routing or path finding of packets in data switching networks
    • H04L45/02Topology update or discovery
    • H04L45/021Ensuring consistency of routing table updates, e.g. by using epoch numbers
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04LTRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L47/00Traffic control in data switching networks
    • H04L47/70Admission control; Resource allocation
    • H04L47/76Admission control; Resource allocation using dynamic resource allocation, e.g. in-call renegotiation requested by the user or requested by the network in response to changing network conditions
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04LTRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L47/00Traffic control in data switching networks
    • H04L47/70Admission control; Resource allocation
    • H04L47/82Miscellaneous aspects
    • H04L47/821Prioritising resource allocation or reservation requests
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04LTRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L61/00Network arrangements, protocols or services for addressing or naming
    • H04L61/50Address allocation
    • H04L61/5007Internet protocol [IP] addresses

Definitions

  • the present invention relates to the field of Internet technologies, and in particular, to a network traffic dynamic scheduling allocation method and system.
  • DNS is used to name computers and network services organized into domain hierarchies, and is an important basic service system of the Internet.
  • DNS can make it easier for users to access the Internet without having to remember the IP strings that can be read directly by the machine.
  • the conversion of domain names and IP strings is called domain name resolution.
  • the host that performs this function is called a DNS server.
  • the DNS service can resolve the domain name into other related information, such as an IP address, to complete the website access.
  • the task of the CDN is mainly to transfer the content from the source station to the client as quickly as possible.
  • CDN technology is an effective means to solve the problem of poor Internet performance and excessive content delivery delay in the United States in recent years.
  • the basic idea is to avoid the bottlenecks and links on the Internet that may affect the speed and stability of data transmission, so that content transmission is faster and better.
  • By placing a content distribution network composed of edge node servers throughout the network it can redirect user access requests to users in real time based on network traffic and load conditions of each edge node and the distance to the user and response time. The most recent and best edge node.
  • the system adds a global scheduling layer to the existing network architecture, distributes the source station content to the edge of the network closest to the user, so that the user can obtain the required content nearby, solve the congestion of the Internet network, and improve the response speed of the user visiting the website.
  • the problem of slow response time of users visiting websites caused by small bandwidth of the source station exit network, large user access, uneven distribution of network points, complicated carrier network, and small user access network bandwidth is solved.
  • the existing scheduling systems rely on static policies, that is, through a single DNS mapping table, through these DNS mapping tables to distinguish different domain name coverage, and according to these tables to guide access traffic to different nodes. Although this can solve a large part of the website traffic distribution problem, but it has the following shortcomings:
  • an embodiment of the present invention provides a network traffic dynamic scheduling allocation method and system.
  • the technical solution is as follows:
  • a method for dynamically scheduling allocation of network traffic comprising:
  • the dispatch center system delivers the domain name initial configuration table and the judgment policy to each probe node;
  • each of the detecting nodes Based on the determining policy and the obtained performance information, each of the detecting nodes generates its own optimal coverage record, and uploads the optimal coverage record to the dispatch center system;
  • the dispatch center system converts the best coverage record into a target service node and feeds back to the local DNS server.
  • the determining policy includes the number of service nodes corresponding to each detecting node, whether to allow network access across operators, the detecting time interval, the expected bandwidth usage rate of the serving node, and the expected response time of the service node;
  • the performance information of each of the service nodes includes actual response time, actual availability, and actual bandwidth usage of each of the service nodes.
  • each of the detection sections is based on the determination policy and the acquired performance information
  • the specific steps for the point to generate its own best coverage record include:
  • the probe node calculates a priority corresponding to each of the service nodes based on a desired bandwidth usage of the service node, an expected response time of the service node, and performance information of the service node;
  • the top N service nodes are selected from the sorted service nodes, and the N service nodes are determined to be the best corresponding to the detecting nodes.
  • a coverage node where N is the number of service nodes corresponding to each detection node in the judgment policy;
  • the probe node establishes its own best coverage record, and the optimal coverage record includes the preset domain name, the geographic location of the probe node, and a network service operator and the N service nodes.
  • P i represents a priority corresponding to the i-th service node
  • T s represents an expected response time of the service node in the judgment policy
  • T i represents an actual response time of the i-th service node
  • a i represents The actual availability of the i-th service node is taken as 1 if available, 0 is taken if not available
  • B s represents the expected bandwidth usage rate of the serving node in the judgment policy
  • B i represents the actual content of the i-th service node Bandwidth usage.
  • the method further includes:
  • the service node that is inconsistent with the network service operator of the probe node is removed from the sorted service node;
  • the specific step of determining the target service node address corresponding to the domain name resolution request according to the domain name resolution record includes:
  • the preset DNS server queries, from the domain name resolution record, a target record that matches the geographic location of the local DNS server and the network service operator, and uses the IP address in the target record as the target service. Node address.
  • the method further includes:
  • Each of the detecting nodes acquires performance information of each of the service nodes again according to a detection time interval in the determining policy;
  • Each of the detecting nodes regenerates its own optimal coverage record according to the acquired performance information and the determination policy, and uploads the regenerated optimal coverage record to the dispatch center system.
  • each of the detecting nodes generates an own optimal coverage record
  • uploading the optimal coverage record to the scheduling center system further includes the following steps: the scheduling center system according to the optimal coverage record
  • the initial configuration table is refreshed, and the dispatch center system identifies the target service node according to the refreshed configuration table.
  • the scheduling center system converting the optimal coverage record into a target service node and feeding back to the local DNS server includes:
  • the dispatching center system converts each of the best coverage records into a domain name resolution record, and pushes the domain name resolution record to a preset DNS server;
  • a network traffic dynamic scheduling distribution system the system includes a dispatch center system, at least one probe node, and a preset DNS server, wherein:
  • the dispatching center system is configured to send a domain name initial configuration table and a determining policy to each of the detecting nodes and receive an optimal coverage record uploaded by each of the detecting nodes; and convert each of the best coverage records into domain name resolution Recording, and pushing the domain name resolution record to the preset DNS server;
  • Each of the detecting nodes is configured to initiate, for a preset domain name in the initial configuration table of the domain name, a detection of each service node of the preset domain name, to obtain performance information of each service node; The policy and the obtained performance information, generate an optimal coverage record of the self, and upload the optimal coverage record to the dispatch center system;
  • the preset DNS server is configured to receive a domain name resolution request sent by a local DNS server, determine a target service node address corresponding to the domain name resolution request according to the domain name resolution record, and feed back the target service node address to the Describe the local DNS server.
  • the determining policy includes the number of service nodes corresponding to each detecting node, whether to allow network access across operators, the detecting time interval, the expected bandwidth usage rate of the serving node, and the expected response time of the service node; each of the service nodes
  • the performance information includes the actual response time, actual availability, and actual bandwidth usage of each of the service nodes.
  • each of the detecting nodes includes:
  • a location information obtaining unit configured to acquire its own geographical location and a network service operator according to its own IP address
  • a priority calculation unit configured to calculate a priority corresponding to each of the service nodes based on a desired bandwidth usage rate of the service node, an expected response time of the service node, and performance information of the service node;
  • a sorting unit configured to sort each of the service nodes according to a priority from high to low
  • a service node screening unit configured to: when the determining policy allows network access across operators, select the top N service nodes from the sorted service nodes, and determine the N service nodes as themselves The best coverage node; where N is the number of service nodes corresponding to each detection node in the judgment policy;
  • the coverage record establishing unit is configured to establish an optimal coverage record of the user, where the optimal coverage record includes the preset domain name, the geographic location of the probe node, and a network service operator and the N service nodes.
  • the priority calculation unit calculates a priority corresponding to each of the service nodes according to the following formula:
  • P i represents a priority corresponding to the i-th service node
  • T s represents an expected response time of the service node in the judgment policy
  • T i represents an actual response time of the i-th service node
  • a i represents The actual availability of the i-th service node is taken as 1 if available, 0 is taken if not available
  • B s represents the expected bandwidth usage rate of the serving node in the judgment policy
  • B i represents the actual content of the i-th service node Bandwidth usage.
  • system further includes:
  • a service node removal unit configured to remove, from the sorted service node, a service node that is inconsistent with the network service operator of the probe node when the judgment policy does not allow network access across the operator;
  • a remaining node screening unit configured to filter out the top N service nodes from the remaining service nodes, and determine the N service nodes as the best coverage node corresponding to the probe node; where N is The number of service nodes corresponding to each probe node in the judgment strategy.
  • the preset DNS server includes:
  • a local DNS address obtaining unit configured to obtain an IP address of the local DNS server from the domain name resolution request
  • a location determining unit configured to determine a geographic location of the local DNS server and a network service operator according to an IP address of the local DNS server;
  • a target record query unit configured to query, from the domain name resolution record, a target record that matches a geographic location of the local DNS server and a network service operator, and use the IP address in the target record as the target Service node address.
  • the technical solution provided by the embodiment of the present invention has the beneficial effects that the present invention detects an ungrouped service node group by establishing an extensive coverage network, and filters and dynamically groups the service nodes according to the detected data. Establish an optimal coverage form.
  • the administrator only needs to establish a simple initial configuration table for the domain name, without having to build a large and complex coverage table according to different network service operators and different geographical locations.
  • the present invention breaks the service node IP address and the limitation of the network service operator, and all factors are based on response time and resource usage rate. In the case of reliable service quality, the policy can also be allowed to restrict the network across operators. access.
  • the present invention can automatically adjust the coverage table according to a preset period, so that the coverage table is gradually optimized in actual use, and the process does not require the intervention of an experienced operation and maintenance personnel, thereby reducing the operation and maintenance cost.
  • FIG. 1 is a flowchart of a network traffic dynamic scheduling allocation method according to Embodiment 1 of the present invention
  • FIG. 3 is a flowchart of a method for determining a target service node address in Embodiment 1 of the present invention
  • FIG. 4 is a schematic structural diagram of a network traffic dynamic scheduling distribution system according to Embodiment 2 of the present invention.
  • an embodiment of the present application discloses a network traffic dynamic scheduling allocation method, where the method includes:
  • S1 The dispatch center system sends the domain name initial configuration table and the judgment policy to each probe node.
  • the domain name initial configuration table may be preset by a configuration administrator in a server of the dispatch center system.
  • the domain name initial configuration table may include a preset domain name and an IP address of multiple service nodes pointed to by the preset domain name.
  • Table 1 It can be seen from Table 1 that the three nodes of the preset domain name can respectively correspond to different Internet Service Providers (ISPs) and geographical locations.
  • ISPs Internet Service Providers
  • Table 2 a domain name initial configuration table as shown in Table 2 can be generated:
  • the preset domain name can correspond to the IP addresses of the three service nodes.
  • the configuration administrator can also customize the corresponding judgment policy in the server.
  • the judgment policy may include information such as the number of service nodes corresponding to each probe node, whether to allow network access across operators, the detection time interval, the expected bandwidth usage rate of the service node, and the expected response time of the service node.
  • the judgment strategy can be as follows:
  • the number of service nodes corresponding to each probe node is one;
  • the expected bandwidth usage of the service node is 80%;
  • the expected response time is 100ms.
  • the server of the dispatch center system After receiving the above configuration, the server of the dispatch center system will synchronize the domain name initial configuration table and the judgment policy to all the probe nodes through the reliable data transmission channel.
  • the number of the detecting nodes is seven, and the information of the seven detecting nodes may be as shown in Table 3:
  • Each of the probe nodes in Table 3 can receive the initial configuration table and the judgment policy of the domain name sent by the dispatch center system.
  • each of the detecting nodes initiates detection of each service node of the preset domain name to obtain performance information of each service node.
  • each of the detecting nodes Based on the determining policy and the obtained performance information, each of the detecting nodes generates an own optimal coverage record, and uploads the optimal coverage record to the dispatch center system.
  • each detecting node may automatically enter the optimal coverage table generating process.
  • the probe node fjlt_node1 is taken as an example for description:
  • fjlt_node1 After receiving the domain name initial configuration table and the judgment policy, fjlt_node1 can detect the IP address (1.1.1.1, 2.2.2.1, 3.3.3.3) of the service node. At this point, it can be assumed that the service node has an SNMP or open bandwidth query interface enabled, so that the probe node can detect normally.
  • the detected performance information may include actual response time, actual availability, and actual bandwidth usage of each of the service nodes. Among them, the actual availability can be represented by 1 or 0, 1 means available, 0 means unavailable.
  • a probe condition table as shown in Table 4 can be formed:
  • the detecting node may combine the determining policy and the performance information. Generate your own best coverage record. Specifically, referring to FIG. 2, the steps of generating an optimal coverage record may include:
  • S31 Obtain a geographic location of the probe node and a network service operator according to the IP address of the probe node.
  • the ISP can be determined from the probe node information table according to its IP address 100.1.1.1 as China Unicom, and the geographical location is Fujian province.
  • the probe node calculates a priority corresponding to each of the service nodes based on a desired bandwidth usage of the service node, an expected response time of the service node, and performance information of the service node.
  • the priority corresponding to each of the service nodes may be calculated according to the following formula:
  • P i represents a priority corresponding to the i-th service node
  • T s represents an expected response time of the service node in the judgment policy
  • T i represents an actual response time of the i-th service node
  • a i represents said i-th actual availability of a service node, if available, then take 1, if not available then take 0,
  • B s represents the determination of the desired bandwidth usage policy service node
  • B i represents the i-th actual serving node Bandwidth usage.
  • S33 Sort each of the service nodes according to a priority from high to low;
  • the top N service nodes are selected from the sorted service nodes, and the N service nodes are determined to be corresponding to the detecting node.
  • An optimal coverage node where N is the number of service nodes corresponding to each detection node in the judgment policy;
  • the detecting node establishes its own best coverage record, where the optimal coverage record includes the preset domain name, the geographic location of the probe node, and a network service operator and the N service nodes.
  • each probe node generates an optimal coverage record and reports it to the dispatch center system.
  • the dispatch center system can perform the best coverage record table of the preset domain name www.test.com as shown in Table 6. Summary:
  • the top N service nodes may be selected from the remaining service nodes, and the N service nodes are determined as the best coverage nodes corresponding to the probe nodes; wherein N is each of the judgment policies The number of service nodes corresponding to the probe node.
  • S4 The dispatch center system converts the best coverage record into a target service node and feeds back to the local DNS server.
  • the scheduling center system may convert each of the best coverage records into a domain name resolution record, and push the domain name resolution record to a preset DNS server.
  • the optimal coverage record is generally not recognized by the DNS server, so the dispatch center system can convert the optimal coverage record into a domain name resolution record that can be recognized by the DNS server.
  • the domain name resolution record can be pushed to a preset DNS server in the dispatch center system, so that a subsequent traffic scheduling process can be performed.
  • the preset DNS server may receive a domain name resolution request sent by the local DNS server, determine a target service node address corresponding to the domain name resolution request according to the domain name resolution record, and feed back the target service node address to the Describe the local DNS server.
  • the domain name when the user accesses the preset domain name, the domain name may be sent to the local DNS server.
  • the local DNS server may send a domain name resolution request to the preset DNS server to determine an IP address of the best service node corresponding to the preset domain name.
  • the target service node address may be determined according to the following steps:
  • the preset DNS server obtains an IP address of the local DNS server from the domain name resolution request.
  • the preset DNS server determines a geographic location of the local DNS server and a network service operator according to an IP address of the local DNS server.
  • S53 the preset DNS server queries, from the domain name resolution record, a target record that matches a geographic location of the local DNS server and a network service operator, and uses the IP address in the target record as the Target service node address.
  • the target service node address When the target node is accessed through the service node, it can have a smoother network experience.
  • the dispatching center system may refresh according to the optimal coverage record.
  • the dispatch center system identifies the target service node according to the refreshed configuration table.
  • the initial configuration table may be deleted, and the optimal coverage record is written in a path of the initial configuration table, so that the optimal coverage record is used to update the Initial configuration table.
  • each of the probe nodes may acquire again according to the detection time interval in the judgment policy. Performance information of each of the service nodes. Then, each of the detecting nodes may regenerate its own optimal coverage record according to the acquired performance information and the determination policy, and upload the regenerated optimal coverage record to the dispatch center system. In this way, the dispatching center can thus update the best coverage record previously stored locally based on the regenerated optimal coverage record.
  • the scheduling can be made more accurate, and the traffic can be tilted to the spare service node with good response quality, and the service access rate can be improved while the service access speed can be improved and reduced. Dispatch operation and maintenance difficulty.
  • the present application further discloses a network traffic dynamic scheduling distribution system, where the system includes a dispatch center system 100, at least one probe node 200, and a preset DNS server 300, where:
  • the dispatching center system 100 is configured to send a domain name initial configuration table and a determining policy to each of the detecting nodes, and receive an optimal coverage record uploaded by each of the detecting nodes; and convert each of the best coverage records into a domain name. Parsing the record and pushing the domain name resolution record to the preset DNS server;
  • Each of the detecting nodes 200 is configured to initiate, for a preset domain name in the domain name initial configuration table, detection of each service node of the preset domain name, to obtain performance information of each of the service nodes; Determining the policy and the obtained performance information, generating an optimal coverage record of the self, and uploading the optimal coverage record to the dispatch center system;
  • the preset DNS server 300 is configured to receive a domain name resolution request sent by a local DNS server, determine a target service node address corresponding to the domain name resolution request according to the domain name resolution record, and feed back the target service node address to The local DNS server.
  • the determining policy includes the number of service nodes corresponding to each detecting node, whether to allow network access across operators, the detecting time interval, the expected bandwidth usage rate of the serving node, and the expected response time of the service node;
  • the performance information of each of the service nodes includes actual response time, actual availability, and actual bandwidth usage of each of the service nodes.
  • each of the detecting nodes 200 includes:
  • a location information obtaining unit configured to acquire its own geographical location and a network service operator according to its own IP address
  • a priority calculation unit configured to calculate a priority corresponding to each of the service nodes based on a desired bandwidth usage rate of the service node, an expected response time of the service node, and performance information of the service node;
  • a sorting unit configured to sort each of the service nodes according to a priority from high to low
  • a service node screening unit configured to: when the determining policy allows network access across operators, select the top N service nodes from the sorted service nodes, and select the N service nodes Determining the best coverage node for itself; wherein N is the number of service nodes corresponding to each detection node in the judgment policy;
  • the coverage record establishing unit is configured to establish an optimal coverage record of the user, where the optimal coverage record includes the preset domain name, the geographic location of the probe node, and a network service operator and the N service nodes.
  • the priority calculation unit calculates a priority corresponding to each of the service nodes according to the following formula:
  • P i represents a priority corresponding to the i-th service node
  • T s represents an expected response time of the service node in the judgment policy
  • T i represents an actual response time of the i-th service node
  • a i represents The actual availability of the i-th service node is taken as 1 if available, 0 is taken if not available
  • B s represents the expected bandwidth usage rate of the serving node in the judgment policy
  • B i represents the actual content of the i-th service node Bandwidth usage.
  • system further includes:
  • a service node removal unit configured to remove, from the sorted service node, a service node that is inconsistent with the network service operator of the probe node when the judgment policy does not allow network access across the operator;
  • a remaining node screening unit configured to filter out the top N service nodes from the remaining service nodes, and determine the N service nodes as the best coverage node corresponding to the probe node; where N is The number of service nodes corresponding to each probe node in the judgment strategy.
  • the preset DNS server includes:
  • a local DNS address obtaining unit configured to obtain an IP address of the local DNS server from the domain name resolution request
  • a location determining unit configured to determine a geographic location of the local DNS server and a network service operator according to an IP address of the local DNS server;
  • a target record query unit configured to query, from the domain name resolution record, a target record that matches a geographic location of the local DNS server and a network service operator, and use the IP address in the target record as the target Service node address.
  • the technical solution provided by the embodiment of the present invention has the beneficial effects that the present invention passes An extensive coverage network is established to detect ungrouped service node groups, and these service nodes are filtered and dynamically grouped according to the detected data to establish an optimal DNS coverage table.
  • the configuration administrator only needs to establish a simple initial configuration table for the domain name, and does not need to establish a large and complex coverage table according to different network service operators and different geographical locations.
  • the present invention breaks the service node IP address and the limitation of the network service operator, and all factors are based on response time and resource usage rate. In the case of reliable service quality, the policy can also be allowed to restrict the network across operators. access.
  • the present invention can automatically adjust the coverage table according to a preset period, so that the coverage table is gradually optimized in actual use, and the process does not require the intervention of an experienced operation and maintenance personnel, thereby reducing the operation and maintenance cost.

Landscapes

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

Abstract

本发明公开了一种网络流量动态调度分配方法及系统,其中,所述方法包括:调度中心系统将域名初始配置表及判断策略下发至各个探测节点;各个所述探测节点获取各个服务节点的性能信息;各个所述探测节点生成自身的最佳覆盖记录;所述调度中心系统将所述最佳覆盖记录转换成目标服务节点并反馈至本地DNS服务器。本发明公开的网络流量动态调度分配方法及系统,能够提高网络访问的速率并降低运维成本。

Description

一种网络流量动态调度分配方法及系统 技术领域
本发明涉及互联网技术领域,特别涉及一种网络流量动态调度分配方法及系统。
背景技术
互联网出现至今,互联网应用提供者一直在追求为用户提供更快更便捷的服务体验,在此过程中,出现了许多技术,比如DNS(Domain Name System,域名系统)、CDN(Content Distribute Network,内容分发网络)等等。
其中,DNS用于命名组织到域层次结构中的计算机和网络服务,是Internet的重要基础服务系统。DNS的出现,能够使用户更方便的访问互联网,而不用去记住能够被机器直接读取的IP数串。域名与IP数串的转换工作称为域名解析,执行此项功能的主机被称为DNS服务器。当用户在应用程序中输入域名时,DNS服务可将此域名解析成与之对应的其他相关信息,如IP地址,完成网站访问。
在DNS系统之上,衍生了多个业务系统来提高互联网访问速度,比如各种流量调度系统,这其中最出名的流量调度分配系统就是CDN。
CDN的任务主要是内容从源站尽可能快的传递到用户端。CDN技术是近年来在美国首先兴起并迅速发展起来的一种解决互联网性能不佳,内容交付延时过大的有效手段。其基本思想就是尽可能避开互联网上有可能影响数据传输速度和稳定性的瓶颈和环节,使内容传输的更快,更好。通过在网络各处放置边缘节点服务器所构成的内容分发网络,其能够实时根据网络流量和各边缘节点的负载情况以及到用户的距离和响应时间等综合信息将用户的访问请求重定向至离用户最近且最好的边缘节点上。该系统在现有网络架构上增加一个全局调度层,将源站内容分发到最接近用户的网络边缘,使用户可以就近取得所需内容,解决Internet网络拥挤的状况,提高用户访问网站的响应速度,解决因源站出口网络带宽小、用户访问量大、网点分布不均、复杂的运营商网络、用户接入网络带宽小所造成的用户访问网站响应速度慢的问题。
现有调度系统均是依赖静态策略,即通过一张张DNS映射表,通过这些DNS映射表来区分不同的域名覆盖,并根据这些表来将访问流量引导到不同的节点上。虽这样能够解决一大部分网站流量分配问题,但是,却存在如下几个缺点:
1、基于DNS映射表的静态策略,需要花费大量人力进行维护调整,而且调整也不够及时;
2、策略过于固定,而且都是按照地域运营商进行严格切分,造成用户无法访问到最优节点
3、构建准确的策略门槛高,需要有丰富经验的技术人员经过不断长期优化才能逐渐实现。
发明内容
为了解决现有技术的问题,本发明实施例提供了一种网络流量动态调度分配方法及系统。所述技术方案如下:
一方面,一种网络流量动态调度分配方法,所述方法包括:
调度中心系统将域名初始配置表及判断策略下发至各个探测节点;
针对所述域名初始配置表中的预设域名,各个所述探测节点发起对所述预设域名的各个服务节点的探测,以获取各个所述服务节点的性能信息;
基于所述判断策略以及获取的所述性能信息,各个所述探测节点生成自身的最佳覆盖记录,并将所述最佳覆盖记录上传至所述调度中心系统;
所述调度中心系统将所述最佳覆盖记录转换成目标服务节点并反馈至本地DNS服务器。
进一步地,所述判断策略包括各个探测节点对应的服务节点数量、是否允许跨运营商进行网络访问、探测时间间隔、服务节点的期望带宽使用率以及服务节点的期望响应时间;
各个所述服务节点的性能信息包括各个所述服务节点的实际响应时间、实际可用性以及实际带宽使用率。
进一步地,基于所述判断策略以及获取的所述性能信息,各个所述探测节 点生成自身的最佳覆盖记录的具体步骤包括:
根据探测节点的IP地址,获取所述探测节点的地理位置以及网络服务运营商;
基于服务节点的期望带宽使用率、服务节点的期望响应时间以及服务节点的性能信息,所述探测节点计算各个所述服务节点对应的优先级;
根据优先级从高到低的顺序,对各个所述服务节点进行排序;
当所述判断策略允许跨运营商进行网络访问时,从排序后的服务节点中筛选出最靠前的N个服务节点,并将所述N个服务节点确定为所述探测节点对应的最佳覆盖节点;其中,N为所述判断策略中各个探测节点对应的服务节点数量;
所述探测节点建立自身的最佳覆盖记录,所述最佳覆盖记录中包括所述预设域名、所述探测节点的地理位置和网络服务运营商以及所述N个服务节点。
进一步地,按照下述公式计算各个所述服务节点对应的优先级:
Pi=(Ts/Ti)*Ai*(Bs/Bi)
其中,Pi表示第i个服务节点对应的优先级,Ts表示所述判断策略中的服务节点的期望响应时间,Ti表示所述第i个服务节点的实际响应时间,Ai表示所述第i个服务节点的实际可用性,若可用则取1,若不可用则取0,Bs表示所述判断策略中的服务节点的期望带宽使用率,Bi表示第i个服务节点的实际带宽使用率。
进一步地,所述方法还包括:
当所述判断策略不允许跨运营商进行网络访问时,从排序后的服务节点中将与所述探测节点的网络服务运营商不一致的服务节点去除;
从剩余的服务节点中筛选出最靠前的N个服务节点,并将所述N个服务节点确定为所述探测节点对应的最佳覆盖节点;其中,N为所述判断策略中各个探测节点对应的服务节点数量。
进一步地,根据所述域名解析记录确定所述域名解析请求对应的目标服务节点地址的具体步骤包括:
所述预设DNS服务器从所述域名解析请求中获取所述本地DNS服务器的IP地址;
所述预设DNS服务器根据所述本地DNS服务器的IP地址,确定所述本地DNS服务器的地理位置和网络服务运营商;
所述预设DNS服务器从所述域名解析记录中,查询与所述本地DNS服务器的地理位置和网络服务运营商相匹配的目标记录,并将所述目标记录中的IP地址作为所述目标服务节点地址。
进一步地,各个所述探测节点在将所述最佳覆盖记录上传至所述调度中心系统之后,所述方法还包括:
各个所述探测节点按照所述判断策略中的探测时间间隔,再次获取各个所述服务节点的性能信息;
各个所述探测节点根据再次获取的性能信息以及所述判断策略,重新生成自身的最佳覆盖记录,并将重新生成的最佳覆盖记录上传至所述调度中心系统。
进一步地,所述各个所述探测节点生成自身的最佳覆盖记录,并将所述最佳覆盖记录上传至所述调度中心系统还包括以下步骤:所述调度中心系统根据所述最佳覆盖记录刷新所述初始配置表,所述调度中心系统按刷新后的配置表识别目标服务节点。
进一步地,所述调度中心系统将所述最佳覆盖记录转换成目标服务节点并反馈至本地DNS服务器包括:
所述调度中心系统将各个所述最佳覆盖记录转化为域名解析记录,并将所述域名解析记录推送至预设DNS服务器;
所述预设DNS服务器接收本地DNS服务器发来的域名解析请求,根据所述域名解析记录确定所述域名解析请求对应的目标服务节点地址,并将所述目标服务节点地址反馈给所述本地DNS服务器。
另一方面,一种网络流量动态调度分配系统,所述系统包括调度中心系统、至少一个探测节点以及预设DNS服务器,其中:
所述调度中心系统,用于将域名初始配置表及判断策略下发至各个所述探测节点并接收各个所述探测节点上传的最佳覆盖记录;将各个所述最佳覆盖记录转化为域名解析记录,并将所述域名解析记录推送至所述预设DNS服务器;
各个所述探测节点,用于针对所述域名初始配置表中的预设域名,发起对所述预设域名的各个服务节点的探测,以获取各个所述服务节点的性能信息;基于所述判断策略以及获取的所述性能信息,生成自身的最佳覆盖记录,并将所述最佳覆盖记录上传至所述调度中心系统;
所述预设DNS服务器,用于接收本地DNS服务器发来的域名解析请求,根据所述域名解析记录确定所述域名解析请求对应的目标服务节点地址,并将所述目标服务节点地址反馈给所述本地DNS服务器。
进一步地,所述判断策略包括各个探测节点对应的服务节点数量、是否允许跨运营商进行网络访问、探测时间间隔、服务节点的期望带宽使用率以及服务节点的期望响应时间;各个所述服务节点的性能信息包括各个所述服务节点的实际响应时间、实际可用性以及实际带宽使用率。
进一步地,各个所述探测节点包括:
位置信息获取单元,用于根据自身的IP地址,获取自身的地理位置以及网络服务运营商;
优先级计算单元,用于基于服务节点的期望带宽使用率、服务节点的期望响应时间以及服务节点的性能信息,计算各个所述服务节点对应的优先级;
排序单元,用于根据优先级从高到低的顺序,对各个所述服务节点进行排序;
服务节点筛选单元,用于当所述判断策略允许跨运营商进行网络访问时,从排序后的服务节点中筛选出最靠前的N个服务节点,并将所述N个服务节点确定为自身的最佳覆盖节点;其中,N为所述判断策略中各个探测节点对应的服务节点数量;
覆盖记录建立单元,用于建立自身的最佳覆盖记录,所述最佳覆盖记录中包括所述预设域名、所述探测节点的地理位置和网络服务运营商以及所述N个服务节点。
进一步地,所述优先级计算单元按照下述公式计算各个所述服务节点对应的优先级:
Pi=(Ts/Ti)*Ai*(Bs/Bi)
其中,Pi表示第i个服务节点对应的优先级,Ts表示所述判断策略中的服务节点的期望响应时间,Ti表示所述第i个服务节点的实际响应时间,Ai表示所述第i个服务节点的实际可用性,若可用则取1,若不可用则取0,Bs表示所述判断策略中的服务节点的期望带宽使用率,Bi表示第i个服务节点的实际带宽使用率。
进一步地,所述系统还包括:
服务节点去除单元,用于当所述判断策略不允许跨运营商进行网络访问时,从排序后的服务节点中将与所述探测节点的网络服务运营商不一致的服务节点去除;
剩余节点筛选单元,用于从剩余的服务节点中筛选出最靠前的N个服务节点,并将所述N个服务节点确定为所述探测节点对应的最佳覆盖节点;其中,N为所述判断策略中各个探测节点对应的服务节点数量。
进一步地,所述预设DNS服务器包括:
本地DNS地址获取单元,用于从所述域名解析请求中获取所述本地DNS服务器的IP地址;
位置确定单元,用于根据所述本地DNS服务器的IP地址,确定所述本地DNS服务器的地理位置和网络服务运营商;
目标记录查询单元,用于从所述域名解析记录中,查询与所述本地DNS服务器的地理位置和网络服务运营商相匹配的目标记录,并将所述目标记录中的IP地址作为所述目标服务节点地址。
本发明实施例提供的技术方案带来的有益效果是:本发明通过建立一个覆盖广泛的探测网络来探测未分组的服务节点组,根据探测得到的数据来对这些服务节点进行筛选及动态分组,建立最佳覆盖表。根据本发明的技术方案,配 置管理员只需为域名建立一个简单的初始配置表即可,无需根据不同的网络服务运营商以及不同的地理位置建立庞大且复杂的覆盖表。此外,本发明打破服务节点IP地址及网络服务运营商的限制,一切以响应时间、资源使用率等因素为准,在服务质量可靠的情况下,还可以通过判断策略限定允许跨运营商进行网络访问。再者,本发明可以按照预设的周期,自动调整覆盖表,使覆盖表在实际使用中逐渐达到最佳,且这个过程无需有经验运维人员介入,从而能够降低运维成本。
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是本发明实施例一中网络流量动态调度分配方法的流程图;
图2是本发明实施例一中最佳覆盖记录生成的流程图;
图3是本发明实施例一中目标服务节点地址的确定方法流程图;
图4是本发明实施例二中网络流量动态调度分配系统的结构示意图。
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。
实施例一
请参阅图1,本申请实施例公开一种网络流量动态调度分配方法,所述方法包括:
S1:调度中心系统将域名初始配置表及判断策略下发至各个探测节点。
在本实施方式中,所述域名初始配置表可以是配置管理员在调度中心系统的服务器中预先设置的。该域名初始配置表可以包括预设域名以及该预设域名指向的多个服务节点的IP地址。
在本实施方式中,假设针对预设域名www.test.com可以具备三个服务节点, 这三个服务节点的信息可以如表1所示:
表1服务节点信息
ISP 地理位置 服务节点IP地址 机房出口带宽
中国联通 福建 1.1.1.1 100Mbps
中国电信 北京 2.2.2.1 1Gbps
中国移动 上海 3.3.3.3 2Gbps
由表1可见,该预设域名的这三个节点可以分别对应不同的网络服务运营商(Internet Service Provider,ISP)以及地理位置。这样,根据上述的服务节点信息,在本实施方式中可以生成如表2所示的域名初始配置表:
表2域名初始配置表
预设域名 服务节点IP地址
www.test.com 1.1.1.1;2.2.2.1;3.3.3.3
其中,该预设域名可以对应三个服务节点的IP地址。
在本实施方式中,配置管理员还可以在服务器中定制相应的判断策略。所述判断策略可以包括各个探测节点对应的服务节点数量、是否允许跨运营商进行网络访问、探测时间间隔、服务节点的期望带宽使用率以及服务节点的期望响应时间等信息。例如,所述判断策略可以如下所示:
1、每3分钟探测一次;
2、每个探测节点对应的服务节点的数量是1个;
3、允许跨运营商进行网络访问;
4、服务节点的期望带宽使用率为80%;
5、期望响应时间为100ms。
调度中心系统的服务器收到如上配置后,将会将域名初始配置表及判断策略通过可靠数据传输通道同步至所有探测节点。
在本实施方式中,假设所述探测节点的数量为7个,这7个探测节点的信息可以如表3所示:
表3探测节点信息表
探测节点名称 IP地址 ISP 地理位置
fjlt_node1 100.1.1.1 中国联通 福建省
bjdx_node1 110.1.1.1 中国电信 广东省
ahdx_node1 120.1.1.1 中国电信 安徽省
shdx_node1 130.1.1.1 中国电信 上海市
zjdx_node1 140.1.1.1 中国电信 浙江省
shyd_node1 150.1.1.1 中国移动 上海市
bjlt_node1 160.1.1.1 中国联通 北京市
表3中的各个探测节点均可以接收到调度中心系统发来的域名初始配置表及判断策略。
S2:针对所述域名初始配置表中的预设域名,各个所述探测节点发起对所述预设域名的各个服务节点的探测,以获取各个所述服务节点的性能信息;
S3:基于所述判断策略以及获取的所述性能信息,各个所述探测节点生成自身的最佳覆盖记录,并将所述最佳覆盖记录上传至所述调度中心系统。
在本实施方式中,各个探测节点接收到域名初始配置表以及判断策略后,可以自动进入最佳覆盖表生成流程。具体地,以探测节点fjlt_node1为例进行说明:
fjlt_node1接收到域名初始配置表和判断策略后,可以对服务节点的IP地址(1.1.1.1,2.2.2.1,3.3.3.3)进行探测。此时,可以假定服务节点均有开启SNMP或开放带宽查询接口,从而使得探测节点能够正常进行探测。探测到的性能信息可以包括各个所述服务节点的实际响应时间、实际可用性以及实际带宽使用率。其中,实际可用性可以通过1或者0来表示,1表示可用,0则表示不可用。该探测节点探测之后,可以形成如表4所示探测情况表:
表4探测情况表
服务节点 实际响应时间 实际可用性 实际带宽使用率
1.1.1.1 60ms 1 40%
2.2.2.1 100ms 1 60%
3.3.3.3 180ms 1 70%
在本实施方式中,该探测节点可以结合所述判断策略以及所述性能信息, 生成自身的最佳覆盖记录。具体地,请参阅图2,生成最佳覆盖记录的步骤可以包括:
S31:根据探测节点的IP地址,获取所述探测节点的地理位置以及网络服务运营商。
以fjlt_node1为例,可以根据其IP地址100.1.1.1从探测节点信息表中确定其ISP为中国联通,地理位置为福建省。
S32:基于服务节点的期望带宽使用率、服务节点的期望响应时间以及服务节点的性能信息,所述探测节点计算各个所述服务节点对应的优先级。
在本实施方式中,可以按照下述公式计算各个所述服务节点对应的优先级:
Pi=(Ts/Ti)*Ai*(Bs/Bi)
其中,Pi表示第i个服务节点对应的优先级,Ts表示所述判断策略中的服务节点的期望响应时间,Ti表示所述第i个服务节点的实际响应时间,Ai表示所述第i个服务节点的实际可用性,若可用则取1,若不可用则取0,Bs表示所述判断策略中的服务节点的期望带宽使用率,Bi表示第i个服务节点的实际带宽使用率。
这样,对于节点1.1.1.1,其优先级可以为(100ms/60ms)*1*(80%/40%)=3.3;对于节点2.2.2.1,其优先级可以为(100ms/100ms)*1*(80%/60%)=1.3;对于节点3.3.3.3,其优先级可以为(100ms/180ms)*1*(80%/70%)=0.63。
S33:根据优先级从高到低的顺序,对各个所述服务节点进行排序;
S34:当所述判断策略允许跨运营商进行网络访问时,从排序后的服务节点中筛选出最靠前的N个服务节点,并将所述N个服务节点确定为所述探测节点对应的最佳覆盖节点;其中,N为所述判断策略中各个探测节点对应的服务节点数量;
S35:所述探测节点建立自身的最佳覆盖记录,所述最佳覆盖记录中包括所述预设域名、所述探测节点的地理位置和网络服务运营商以及所述N个服务节点。
在本实施方式中,根据判断策略中指定“每个探测节点对应的服务节点是1个”,“允许跨运营商”等参数,可以确定服务节点1.1.1.1的优先级最高,从而可以筛选出如表5所示的该探测节点对应的最佳覆盖记录:
表5最佳覆盖记录
预设域名 ISP 地理位置 服务节点
www.test.com 中国联通 福建 1.1.1.1
同理,每个探测节点均会生成一个最佳覆盖记录,并上报给调度中心系统,最后调度中心系统可以将预设域名www.test.com的最佳覆盖记录表进行如表6所示的汇总:
表6最佳覆盖记录汇总表
域名 ISP 地理位置 服务IP
www.test.com 中国电信 福建省 1.1.1.1
www.test.com 中国电信 广东省 1.1.1.1
www.test.com 中国电信 安徽省 2.2.2.1
www.test.com 中国电信 上海市 2.2.2.1
www.test.com 中国电信 浙江省 1.1.1.1
www.test.com 中国移动 上海市 3.3.3.3
www.test.com 中国联通 北京市 2.2.2.1
需要说明的是,当所述判断策略不允许跨运营商进行网络访问时,则需要从排序后的服务节点中将与所述探测节点的网络服务运营商不一致的服务节点去除。然后可以从剩余的服务节点中筛选出最靠前的N个服务节点,并将所述N个服务节点确定为所述探测节点对应的最佳覆盖节点;其中,N为所述判断策略中各个探测节点对应的服务节点数量。
S4:所述调度中心系统将所述最佳覆盖记录转换成目标服务节点并反馈至本地DNS服务器。
具体地,在本实施方式中,所述调度中心系统可以将各个所述最佳覆盖记录转化为域名解析记录,并将所述域名解析记录推送至预设DNS服务器。
在本实施方式中,所述最佳覆盖记录通常不能被DNS服务器识别,因此所述调度中心系统可以将所述最佳覆盖记录转化为能够被DNS服务器识别的域名解析记录。该域名解析记录可以被推送至调度中心系统中的预设DNS服务器处,从而可以进行后续的流量调度过程。
此外,所述预设DNS服务器可以接收本地DNS服务器发来的域名解析请求,根据所述域名解析记录确定所述域名解析请求对应的目标服务节点地址,并将所述目标服务节点地址反馈给所述本地DNS服务器。
在本实施方式中,用户在访问所述预设域名时,可以将该域名发送至本地DNS服务器。本地DNS服务器在接收到用户发来的域名时,可以向所述预设DNS服务器发送域名解析请求,以确定所述预设域名所对应的最佳服务节点的IP地址。
具体地,请参阅图3,在本实施方式中,可以按照以下步骤来确定目标服务节点地址:
S51:所述预设DNS服务器从所述域名解析请求中获取所述本地DNS服务器的IP地址;
S52:所述预设DNS服务器根据所述本地DNS服务器的IP地址,确定所述本地DNS服务器的地理位置和网络服务运营商;
S53:所述预设DNS服务器从所述域名解析记录中,查询与所述本地DNS服务器的地理位置和网络服务运营商相匹配的目标记录,并将所述目标记录中的IP地址作为所述目标服务节点地址。
例如,所述本地DNS服务器的地理位置为广东省,ISP为中国电信,那么根据这两个信息,从表6中可以看出其对应的最佳服务节点为1.1.1.1,那么1.1.1.1便可以作为所述目标服务节点地址。通过该服务节点访问目标资源时,可以具备较流畅的网络体验。
在本实施方式中,各个所述探测节点生成自身的最佳覆盖记录,并将所述最佳覆盖记录上传至所述调度中心系统后,所述调度中心系统可以根据所述最佳覆盖记录刷新所述初始配置表,所述调度中心系统按刷新后的配置表识别目标服务节点。具体地,在所述调度中心系统中,可以将所述初始配置表删除,并在所述初始配置表的路径下写入所述最佳覆盖记录,从而用所述最佳覆盖记录更新所述初始配置表。
在本实施方式中,探测节点在生成最佳覆盖记录并将所述最佳覆盖记录上传至所述调度中心系统之后,各个所述探测节点可以按照所述判断策略中的探测时间间隔,再次获取各个所述服务节点的性能信息。然后,各个所述探测节点可以根据再次获取的性能信息以及所述判断策略,重新生成自身的最佳覆盖记录,并将重新生成的最佳覆盖记录上传至所述调度中心系统。这样,所述调度中心从而可以根据重新生成的最佳覆盖记录,更新本地之前存储的最佳覆盖记录。
这样,通过定期对最佳覆盖记录进行自我调整,可以使调度更加准确,使流量向空余的、响应质量好的服务节点倾斜,在保障服务节点使用率的同时还可以提高业务的访问速度及降低调度运维难度。
实施例二
请参阅图4,本申请还公开一种网络流量动态调度分配系统,所述系统包括调度中心系统100、至少一个探测节点200以及预设DNS服务器300,其中:
所述调度中心系统100,用于将域名初始配置表及判断策略下发至各个所述探测节点并接收各个所述探测节点上传的最佳覆盖记录;将各个所述最佳覆盖记录转化为域名解析记录,并将所述域名解析记录推送至所述预设DNS服务器;
各个所述探测节点200,用于针对所述域名初始配置表中的预设域名,发起对所述预设域名的各个服务节点的探测,以获取各个所述服务节点的性能信息;基于所述判断策略以及获取的所述性能信息,生成自身的最佳覆盖记录,并将所述最佳覆盖记录上传至所述调度中心系统;
所述预设DNS服务器300,用于接收本地DNS服务器发来的域名解析请求,根据所述域名解析记录确定所述域名解析请求对应的目标服务节点地址,并将所述目标服务节点地址反馈给所述本地DNS服务器。
在本申请一个实施方式中,所述判断策略包括各个探测节点对应的服务节点数量、是否允许跨运营商进行网络访问、探测时间间隔、服务节点的期望带宽使用率以及服务节点的期望响应时间;各个所述服务节点的性能信息包括各个所述服务节点的实际响应时间、实际可用性以及实际带宽使用率。
在本申请一个实施方式中,各个所述探测节点200包括:
位置信息获取单元,用于根据自身的IP地址,获取自身的地理位置以及网络服务运营商;
优先级计算单元,用于基于服务节点的期望带宽使用率、服务节点的期望响应时间以及服务节点的性能信息,计算各个所述服务节点对应的优先级;
排序单元,用于根据优先级从高到低的顺序,对各个所述服务节点进行排序;
服务节点筛选单元,用于当所述判断策略允许跨运营商进行网络访问时,从排序后的服务节点中筛选出最靠前的N个服务节点,并将所述N个服务节点 确定为自身的最佳覆盖节点;其中,N为所述判断策略中各个探测节点对应的服务节点数量;
覆盖记录建立单元,用于建立自身的最佳覆盖记录,所述最佳覆盖记录中包括所述预设域名、所述探测节点的地理位置和网络服务运营商以及所述N个服务节点。
在本申请一个实施方式中,所述优先级计算单元按照下述公式计算各个所述服务节点对应的优先级:
Pi=(Ts/Ti)*Ai*(Bs/Bi)
其中,Pi表示第i个服务节点对应的优先级,Ts表示所述判断策略中的服务节点的期望响应时间,Ti表示所述第i个服务节点的实际响应时间,Ai表示所述第i个服务节点的实际可用性,若可用则取1,若不可用则取0,Bs表示所述判断策略中的服务节点的期望带宽使用率,Bi表示第i个服务节点的实际带宽使用率。
在本申请一个实施方式中,所述系统还包括:
服务节点去除单元,用于当所述判断策略不允许跨运营商进行网络访问时,从排序后的服务节点中将与所述探测节点的网络服务运营商不一致的服务节点去除;
剩余节点筛选单元,用于从剩余的服务节点中筛选出最靠前的N个服务节点,并将所述N个服务节点确定为所述探测节点对应的最佳覆盖节点;其中,N为所述判断策略中各个探测节点对应的服务节点数量。
在本申请一个实施方式中,所述预设DNS服务器包括:
本地DNS地址获取单元,用于从所述域名解析请求中获取所述本地DNS服务器的IP地址;
位置确定单元,用于根据所述本地DNS服务器的IP地址,确定所述本地DNS服务器的地理位置和网络服务运营商;
目标记录查询单元,用于从所述域名解析记录中,查询与所述本地DNS服务器的地理位置和网络服务运营商相匹配的目标记录,并将所述目标记录中的IP地址作为所述目标服务节点地址。
由上可见,本发明实施例提供的技术方案带来的有益效果是:本发明通过 建立一个覆盖广泛的探测网络来探测未分组的服务节点组,根据探测得到的数据来对这些服务节点进行筛选及动态分组,建立最佳DNS覆盖表。根据本发明的技术方案,配置管理员只需为域名建立一个简单的初始配置表即可,无需根据不同的网络服务运营商以及不同的地理位置建立庞大且复杂的覆盖表。此外,本发明打破服务节点IP地址及网络服务运营商的限制,一切以响应时间、资源使用率等因素为准,在服务质量可靠的情况下,还可以通过判断策略限定允许跨运营商进行网络访问。再者,本发明可以按照预设的周期,自动调整覆盖表,使覆盖表在实际使用中逐渐达到最佳,且这个过程无需有经验运维人员介入,从而能够降低运维成本。
上述本发明实施例序号仅仅为了描述,不代表实施例的优劣。
以上所描述的系统实施例仅仅是示意性的,其中所述作为分离部件说明的单元可以是或者也可以不是物理上分开的,作为单元显示的部件可以是或者也可以不是物理单元,即可以位于一个地方,或者也可以分布到多个网络单元上。可以根据实际的需要选择其中的部分或者全部模块来实现本实施例方案的目的。本领域普通技术人员在不付出创造性的劳动的情况下,即可以理解并实施。
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 (15)

  1. 一种网络流量动态调度分配方法,其特征在于,所述方法包括:
    调度中心系统将域名初始配置表及判断策略下发至各个探测节点;
    针对所述域名初始配置表中的预设域名,各个所述探测节点发起对所述预设域名的各个服务节点的探测,以获取各个所述服务节点的性能信息;
    基于所述判断策略以及获取的所述性能信息,各个所述探测节点生成自身的最佳覆盖记录,并将所述最佳覆盖记录上传至所述调度中心系统;
    所述调度中心系统将所述最佳覆盖记录转换成目标服务节点并反馈至本地DNS服务器。
  2. 根据权利要求1所述的网络流量动态调度分配方法,其特征在于,所述判断策略包括各个探测节点对应的服务节点数量、是否允许跨运营商进行网络访问、探测时间间隔、服务节点的期望带宽使用率以及服务节点的期望响应时间;
    各个所述服务节点的性能信息包括各个所述服务节点的实际响应时间、实际可用性以及实际带宽使用率。
  3. 根据权利要求2所述的网络流量动态调度分配方法,其特征在于,基于所述判断策略以及获取的所述性能信息,各个所述探测节点生成自身的最佳覆盖记录的具体步骤包括:
    根据探测节点的IP地址,获取所述探测节点的地理位置以及网络服务运营商;
    基于服务节点的期望带宽使用率、服务节点的期望响应时间以及服务节点的性能信息,所述探测节点计算各个所述服务节点对应的优先级;
    根据优先级从高到低的顺序,对各个所述服务节点进行排序;
    当所述判断策略允许跨运营商进行网络访问时,从排序后的服务节点中筛选出最靠前的N个服务节点,并将所述N个服务节点确定为所述探测节点对应的最佳覆盖节点;其中,N为所述判断策略中各个探测节点对应的服务节点数量;
    所述探测节点建立自身的最佳覆盖记录,所述最佳覆盖记录中包括所述预设域名、所述探测节点的地理位置和网络服务运营商以及所述N个服务节点。
  4. 根据权利要求3所述的网络流量动态调度分配方法,其特征在于,按照下述公式计算各个所述服务节点对应的优先级:
    Pi=(Ts/Ti)*Ai*(Bs/Bi)
    其中,Pi表示第i个服务节点对应的优先级,Ts表示所述判断策略中的服务节点的期望响应时间,Ti表示所述第i个服务节点的实际响应时间,Ai表示所述第i个服务节点的实际可用性,若可用则取1,若不可用则取0,Bs表示所述判断策略中的服务节点的期望带宽使用率,Bi表示第i个服务节点的实际带宽使用率。
  5. 根据权利要求3所述的网络流量动态调度分配方法,其特征在于,所述方法还包括:
    当所述判断策略不允许跨运营商进行网络访问时,从排序后的服务节点中将与所述探测节点的网络服务运营商不一致的服务节点去除;
    从剩余的服务节点中筛选出最靠前的N个服务节点,并将所述N个服务节点确定为所述探测节点对应的最佳覆盖节点;其中,N为所述判断策略中各个探测节点对应的服务节点数量。
  6. 根据权利要求1所述的网络流量动态调度分配方法,其特征在于,根据所述域名解析记录确定所述域名解析请求对应的目标服务节点地址的具体步骤包括:
    所述预设DNS服务器从所述域名解析请求中获取所述本地DNS服务器的IP地址;
    所述预设DNS服务器根据所述本地DNS服务器的IP地址,确定所述本地DNS服务器的地理位置和网络服务运营商;
    所述预设DNS服务器从所述域名解析记录中,查询与所述本地DNS服务器的地理位置和网络服务运营商相匹配的目标记录,并将所述目标记录中的IP地址作为所述目标服务节点地址。
  7. 根据权利要求1所述的网络流量动态调度分配方法,其特征在于,各个所述探测节点在将所述最佳覆盖记录上传至所述调度中心系统之后,所述方法还包括:
    各个所述探测节点按照所述判断策略中的探测时间间隔,再次获取各个所述服务节点的性能信息;
    各个所述探测节点根据再次获取的性能信息以及所述判断策略,重新生成自身的最佳覆盖记录,并将重新生成的最佳覆盖记录上传至所述调度中心系统。
  8. 根据权利要求1所述的网络流量动态调度分配方法,其特征在于,所述各个所述探测节点生成自身的最佳覆盖记录,并将所述最佳覆盖记录上传至所述调度中心系统还包括以下步骤:所述调度中心系统根据所述最佳覆盖记录刷新所述初始配置表,所述调度中心系统按刷新后的配置表识别目标服务节点。
  9. 根据权利要求1所述的网络流量动态调度分配方法,其特征在于,所述调度中心系统将所述最佳覆盖记录转换成目标服务节点并反馈至本地DNS服务器包括:
    所述调度中心系统将各个所述最佳覆盖记录转化为域名解析记录,并将所述域名解析记录推送至预设DNS服务器;
    所述预设DNS服务器接收本地DNS服务器发来的域名解析请求,根据所述域名解析记录确定所述域名解析请求对应的目标服务节点地址,并将所述目标服务节点地址反馈给所述本地DNS服务器。
  10. 一种网络流量动态调度分配系统,其特征在于,所述系统包括调度中心系统、至少一个探测节点以及预设DNS服务器,其中:
    所述调度中心系统,用于将域名初始配置表及判断策略下发至各个所述探测节点并接收各个所述探测节点上传的最佳覆盖记录;将各个所述最佳覆盖记录转化为域名解析记录,并将所述域名解析记录推送至所述预设DNS服务器;
    各个所述探测节点,用于针对所述域名初始配置表中的预设域名,发起对所述预设域名的各个服务节点的探测,以获取各个所述服务节点的性能信息; 基于所述判断策略以及获取的所述性能信息,生成自身的最佳覆盖记录,并将所述最佳覆盖记录上传至所述调度中心系统;
    所述预设DNS服务器,用于接收本地DNS服务器发来的域名解析请求,根据所述域名解析记录确定所述域名解析请求对应的目标服务节点地址,并将所述目标服务节点地址反馈给所述本地DNS服务器。
  11. 根据权利要求10所述的网络流量动态调度分配系统,其特征在于,所述判断策略包括各个探测节点对应的服务节点数量、是否允许跨运营商进行网络访问、探测时间间隔、服务节点的期望带宽使用率以及服务节点的期望响应时间;各个所述服务节点的性能信息包括各个所述服务节点的实际响应时间、实际可用性以及实际带宽使用率。
  12. 根据权利要求11所述的网络流量动态调度分配系统,其特征在于,各个所述探测节点包括:
    位置信息获取单元,用于根据自身的IP地址,获取自身的地理位置以及网络服务运营商;
    优先级计算单元,用于基于服务节点的期望带宽使用率、服务节点的期望响应时间以及服务节点的性能信息,计算各个所述服务节点对应的优先级;
    排序单元,用于根据优先级从高到低的顺序,对各个所述服务节点进行排序;
    服务节点筛选单元,用于当所述判断策略允许跨运营商进行网络访问时,从排序后的服务节点中筛选出最靠前的N个服务节点,并将所述N个服务节点确定为自身的最佳覆盖节点;其中,N为所述判断策略中各个探测节点对应的服务节点数量;
    覆盖记录建立单元,用于建立自身的最佳覆盖记录,所述最佳覆盖记录中包括所述预设域名、所述探测节点的地理位置和网络服务运营商以及所述N个服务节点。
  13. 根据权利要求12所述的网络流量动态调度分配系统,其特征在于,所述优先级计算单元按照下述公式计算各个所述服务节点对应的优先级:
    Pi=(Ts/Ti)*Ai*(Bs/Bi)
    其中,Pi表示第i个服务节点对应的优先级,Ts表示所述判断策略中的服务节点的期望响应时间,Ti表示所述第i个服务节点的实际响应时间,Ai表示所述第i个服务节点的实际可用性,若可用则取1,若不可用则取0,Bs表示所述判断策略中的服务节点的期望带宽使用率,Bi表示第i个服务节点的实际带宽使用率。
  14. 根据权利要求12所述的网络流量动态调度分配系统,其特征在于,所述系统还包括:
    服务节点去除单元,用于当所述判断策略不允许跨运营商进行网络访问时,从排序后的服务节点中将与所述探测节点的网络服务运营商不一致的服务节点去除;
    剩余节点筛选单元,用于从剩余的服务节点中筛选出最靠前的N个服务节点,并将所述N个服务节点确定为所述探测节点对应的最佳覆盖节点;其中,N为所述判断策略中各个探测节点对应的服务节点数量。
  15. 根据权利要求10所述的网络流量动态调度分配系统,其特征在于,所述预设DNS服务器包括:
    本地DNS地址获取单元,用于从所述域名解析请求中获取所述本地DNS服务器的IP地址;
    位置确定单元,用于根据所述本地DNS服务器的IP地址,确定所述本地DNS服务器的地理位置和网络服务运营商;
    目标记录查询单元,用于从所述域名解析记录中,查询与所述本地DNS服务器的地理位置和网络服务运营商相匹配的目标记录,并将所述目标记录中的IP地址作为所述目标服务节点地址。
PCT/CN2017/086805 2017-05-15 2017-06-01 一种网络流量动态调度分配方法及系统 Ceased WO2018209734A1 (zh)

Priority Applications (2)

Application Number Priority Date Filing Date Title
US16/321,870 US10757037B2 (en) 2017-05-15 2017-06-01 Method and system for dynamic scheduling and allocation of network traffic
EP17909729.0A EP3522466B1 (en) 2017-05-15 2017-06-01 Dynamic scheduling and allocation method and system for network traffic

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201710337462.6 2017-05-15
CN201710337462.6A CN106992938B (zh) 2017-05-15 2017-05-15 一种网络流量动态调度分配方法及系统

Publications (1)

Publication Number Publication Date
WO2018209734A1 true WO2018209734A1 (zh) 2018-11-22

Family

ID=59419460

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2017/086805 Ceased WO2018209734A1 (zh) 2017-05-15 2017-06-01 一种网络流量动态调度分配方法及系统

Country Status (4)

Country Link
US (1) US10757037B2 (zh)
EP (1) EP3522466B1 (zh)
CN (1) CN106992938B (zh)
WO (1) WO2018209734A1 (zh)

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN114448893A (zh) * 2021-12-24 2022-05-06 天翼云科技有限公司 一种cdn节点任务下发的聚合方法、装置及计算机设备
CN114615705A (zh) * 2022-03-11 2022-06-10 广东技术师范大学 一种基于5g网络下单用户资源分配策略方法

Families Citing this family (21)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN107071089B (zh) * 2017-05-18 2020-09-15 腾讯科技(深圳)有限公司 调度控制方法、装置和系统
CN108377245A (zh) * 2018-02-26 2018-08-07 湖南科技学院 一种网络接入请求的优化认证方法及系统
CN110198267B (zh) * 2018-04-26 2022-01-14 腾讯科技(深圳)有限公司 一种流量调度方法、系统及服务器
CN109167674B (zh) * 2018-07-24 2021-05-18 网宿科技股份有限公司 服务节点的评分方法、域名系统dns调度方法及服务器
CN110768852A (zh) * 2018-07-25 2020-02-07 贵州白山云科技股份有限公司 一种云网络节点的测试方法及装置
CN111371826B (zh) * 2018-12-26 2024-04-09 三六零科技集团有限公司 Cdn节点的性能探测方法、装置及系统
US11240049B2 (en) 2019-06-05 2022-02-01 International Business Machines Corporation Automatic recharging of data quota for successful completion of transaction
CN110830604B (zh) * 2019-10-28 2021-09-10 腾讯科技(深圳)有限公司 Dns调度方法、装置
CN113067714B (zh) * 2020-01-02 2022-12-13 中国移动通信有限公司研究院 一种内容分发网络调度处理方法、装置及设备
CN113315719B (zh) * 2020-02-27 2024-09-13 阿里巴巴集团控股有限公司 流量调度方法、设备、系统及存储介质
CN111756731B (zh) * 2020-06-23 2022-06-28 全球能源互联网研究院有限公司 一种专用网络的可信测度方法及系统
CN114095443B (zh) * 2020-08-24 2023-11-03 百度在线网络技术(北京)有限公司 用于管控流量的方法、装置、电子设备及可读存储介质
CN114257634B (zh) * 2020-09-11 2024-03-01 中国电信股份有限公司 服务器发现方法、装置和存储介质
CN114629737B (zh) * 2020-12-14 2024-03-12 深圳Tcl新技术有限公司 一种带宽调整方法、装置、网关设备及存储介质
CN112910798B (zh) * 2021-01-20 2023-07-04 北京洛塔信息技术有限公司 一种流量自动调度方法、系统、设备及存储介质
CN112887224B (zh) * 2021-01-25 2023-01-20 北京金山云网络技术有限公司 流量调度处理方法、装置、电子设备及存储介质
CN113099261B (zh) * 2021-04-27 2023-04-18 上海哔哩哔哩科技有限公司 节点处理方法及装置、节点处理系统
CN114884820A (zh) * 2022-06-02 2022-08-09 北京亿安天下科技股份有限公司 一种网络信息主动配对检测的方法、装置、电子设备及存储介质
CN115914043B (zh) * 2022-11-30 2025-02-11 中盈优创资讯科技有限公司 一种基于bgp路由和dns的客户调度行为分析方法及装置
CN116633864B (zh) * 2023-07-19 2023-11-03 国家计算机网络与信息安全管理中心江西分中心 一种基于云计算平台的流量调度方法
CN117240707A (zh) 2023-09-27 2023-12-15 北京火山引擎科技有限公司 一种网络线路的配置方法、网络请求的调度方法及装置

Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20030079027A1 (en) * 2001-10-18 2003-04-24 Michael Slocombe Content request routing and load balancing for content distribution networks
CN101764747A (zh) * 2009-12-24 2010-06-30 北京世纪互联宽带数据中心有限公司 内容分发网络流量智能调度方法、装置及系统
CN102055818A (zh) * 2010-12-30 2011-05-11 北京世纪互联工程技术服务有限公司 分布式智能dns库系统
CN102136969A (zh) * 2011-02-17 2011-07-27 北京蓝汛通信技术有限责任公司 探测链路质量的方法、装置和系统
CN102148752A (zh) * 2010-12-22 2011-08-10 华为技术有限公司 基于内容分发网络的路由实现方法及相关设备、系统
US20130290466A1 (en) * 2012-04-30 2013-10-31 Sk Telecom Co., Ltd. Method of providing content during hand-over and appartus therefor

Family Cites Families (9)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US7725602B2 (en) 2000-07-19 2010-05-25 Akamai Technologies, Inc. Domain name resolution using a distributed DNS network
EP3567881B1 (en) * 2008-11-17 2021-02-24 Amazon Technologies, Inc. Request routing and updating routing information utilizing client location information
CN102932451B (zh) * 2012-10-31 2015-11-18 北龙中网(北京)科技有限责任公司 一种针对内容分发网络的解决系统
CN104734869A (zh) * 2013-12-19 2015-06-24 华耀(中国)科技有限公司 基于动态探测的智能dns域名系统及方法
TW201728124A (zh) * 2014-09-16 2017-08-01 科勞簡尼克斯股份有限公司 以彈性地定義之通信網路控制器為基礎之網路控制、操作及管理
CN104363313B (zh) * 2014-12-02 2018-09-18 网宿科技股份有限公司 使用内容分发网络的网站的资源使用率保障方法和系统
CN104994123B (zh) * 2015-05-12 2018-10-02 段利平 一种cdn云平台及cdn云平台的流量调度方法
US10091057B2 (en) * 2015-12-28 2018-10-02 Ca, Inc. Configuring distributed monitoring systems
US10601704B2 (en) * 2017-07-31 2020-03-24 Cisco Technology, Inc. Asymmetric routing minimization

Patent Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20030079027A1 (en) * 2001-10-18 2003-04-24 Michael Slocombe Content request routing and load balancing for content distribution networks
CN101764747A (zh) * 2009-12-24 2010-06-30 北京世纪互联宽带数据中心有限公司 内容分发网络流量智能调度方法、装置及系统
CN102148752A (zh) * 2010-12-22 2011-08-10 华为技术有限公司 基于内容分发网络的路由实现方法及相关设备、系统
CN102055818A (zh) * 2010-12-30 2011-05-11 北京世纪互联工程技术服务有限公司 分布式智能dns库系统
CN102136969A (zh) * 2011-02-17 2011-07-27 北京蓝汛通信技术有限责任公司 探测链路质量的方法、装置和系统
US20130290466A1 (en) * 2012-04-30 2013-10-31 Sk Telecom Co., Ltd. Method of providing content during hand-over and appartus therefor

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN114448893A (zh) * 2021-12-24 2022-05-06 天翼云科技有限公司 一种cdn节点任务下发的聚合方法、装置及计算机设备
CN114615705A (zh) * 2022-03-11 2022-06-10 广东技术师范大学 一种基于5g网络下单用户资源分配策略方法
CN114615705B (zh) * 2022-03-11 2022-12-20 广东技术师范大学 一种基于5g网络下单用户资源分配策略方法

Also Published As

Publication number Publication date
CN106992938B (zh) 2020-03-31
US20200186479A1 (en) 2020-06-11
EP3522466A4 (en) 2019-12-11
EP3522466A1 (en) 2019-08-07
CN106992938A (zh) 2017-07-28
EP3522466B1 (en) 2021-04-14
US10757037B2 (en) 2020-08-25

Similar Documents

Publication Publication Date Title
WO2018209734A1 (zh) 一种网络流量动态调度分配方法及系统
US8745128B2 (en) Optimized content distribution based on metrics derived from the end user
US9172632B2 (en) Optimized content distribution based on metrics derived from the end user
CN104320487B (zh) 内容分发网络的http调度系统和方法
CN102148752B (zh) 基于内容分发网络的路由实现方法及相关设备、系统
US10361902B2 (en) Method and system for guaranteeing resource utilization rate of website using content distribution network
CN109951338B (zh) Cdn网络配置方法、配置装置、电子设备及存储介质
CA2827572C (en) Analytics management
CN103841045B (zh) 一种互联网缓存处理方法、内容探测子系统和Cache系统
CN109167674B (zh) 服务节点的评分方法、域名系统dns调度方法及服务器
CN107317879B (zh) 一种用户请求的分发方法及系统
CN101540775B (zh) 内容分发方法、装置与内容分发网络系统
US20150100691A1 (en) Systems and Methods for Performing Localized Server-Side Monitoring in a Content Delivery Network
US12200047B2 (en) Service unit switching method, system, and device for disaster tolerance switching and capacity dispatching
US20140188801A1 (en) Method and system for intelligent load balancing
WO2021169294A1 (zh) 更新应用识别模型的方法、装置及存储介质
WO2016145753A1 (zh) 基于内容分发网络的移动终端流量处理方法、装置及系统
CN107395683A (zh) 一种回源路径的选择方法及服务器
CN103347068A (zh) 一种基于代理集群网络缓存加速方法
CN104092620A (zh) 一种实现网络带宽调整的方法及装置
HK1253571A1 (zh) 服务器集群的自动发现
US12107750B1 (en) Smart cluster load balancing
CN104270371A (zh) 一种基于模糊逻辑的cdn缓存服务器选择方法
AU2024201299B2 (en) Smart cluster load balancing
Xue et al. Dissecting persistent instability of web service: A joint perspective of server schedule dynamics and path latency

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: 17909729

Country of ref document: EP

Kind code of ref document: A1

ENP Entry into the national phase

Ref document number: 2017909729

Country of ref document: EP

Effective date: 20190429

NENP Non-entry into the national phase

Ref country code: DE