WO2017035789A1 - 一种数据传输方法及系统 - Google Patents
一种数据传输方法及系统 Download PDFInfo
- Publication number
- WO2017035789A1 WO2017035789A1 PCT/CN2015/088768 CN2015088768W WO2017035789A1 WO 2017035789 A1 WO2017035789 A1 WO 2017035789A1 CN 2015088768 W CN2015088768 W CN 2015088768W WO 2017035789 A1 WO2017035789 A1 WO 2017035789A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- node
- data
- child node
- designated
- root
- Prior art date
Links
Definitions
- the present invention belongs to the field of communication technologies, and in particular, to a data transmission method and system.
- streaming media communication technology In the field of streaming media related data transmission, the combination of streaming media communication technology and Internet communication technology is more and more widely used. Application scenarios often involve far-reaching data transmission, even multinational data transmission.
- Streaming media distribution based on distribution networks is an essential communication technology. In practical applications, due to the long propagation distance and the unstable network environment, there are some problems that need to be solved in data transmission.
- the node is deployed hierarchically, and the data processing function of each layer node is enhanced, or the number of nodes is increased to improve data transmission efficiency.
- the above prior art also increases the burden on the node to process data and prolongs. With the processing time, more nodes participate in the data processing transmission, which is easily limited by the network bandwidth, and is limited by the location of the node deployment, and causes cost and cost problems caused by the increase of bandwidth or equipment.
- the invention provides a data transmission method and device, which can participate in data transmission as a designated child node by specifying a small number of child nodes with better communication conditions in each hierarchical sub-node, avoid occupying excessive bandwidth, and improve transmission speed and save Network cost.
- the invention provides a data transmission method, comprising:
- the first designated child node is a next-level child node of the root node; requesting data from the root node, and sending the requested data to the second node Specifying a child node, wherein the second designated child node is a node of a next level of the first designated child node; wherein each level is set with a designated child node, and other nodes of the same level as the specified child node are set to The standby child node, the communication indicator parameter between the designated child node and the root node is better than the communication indicator parameter between the standby child node and the root node.
- the present invention provides a data transmission system comprising:
- a first designated child node configured to receive an acquisition data instruction sent by the distribution manager, request data from the root node, and send the requested data to a second designated child node, where the first designated child node is a root node a lower level child node, wherein the second designated child node is a node of a lower level of the first designated child node; wherein each level is provided with a designated child node, and other nodes of the same level as the specified child node All are set as spare child nodes, and the communication indicator parameter between the designated child node and the root node is better than the communication indicator parameter between the standby child node and the root node.
- the present invention selects a small number of designated sub-nodes with better communication status as nodes for requesting data from the upper-level parent node in each layer sub-node, while other nodes do not request the upper-level parent node. Data to save bandwidth, reduce costs, and increase data transfer speed.
- FIG. 1 is a schematic diagram of an application environment of a data transmission method and system provided by an embodiment of the present invention
- FIG. 2 is a schematic flowchart showing the implementation of a data transmission method provided by the first embodiment of the present invention
- Figure 3 is a block diagram showing the structure of a data transmission system provided by a second embodiment of the present invention.
- FIG. 1 is a schematic diagram of an application environment of a data transmission method and system according to an embodiment of the present invention.
- Figure 1 shows a system structure with two layers of child nodes, where g 1 node, g 2 node, g 3 node are first layer child nodes, l 1 node, and l 2 node are second layer child nodes, each Any two nodes between the hierarchical nodes can transfer data to each other.
- the g 1 node, the g 2 node, and the g 3 node are domestic nodes, and the l 1 node and the l 2 node are foreign nodes.
- the application scenario including 101, the root node (i.e. root node) 102, a first distribution manager node designated (i.e. g 1 node) 103, the child node 103 to a first designated alternate node with the node belonging to the same level (i.e. a g 2 node, a g 3 node) 104 , a second designated node (ie, a l 1 node) 105 , a spare child node (ie, a l 2 node) 106 that belongs to the same hierarchical node as the second designated node 105 , and finally receives data The destination node 107.
- the root node i.e. root node
- a first distribution manager node designated i.e. g 1 node
- the child node 103 to a first designated alternate node with the node belonging to the same level (i.e. a g 2 node, a g 3 node) 104
- the root node 102 is a data source and stores streaming media data. There are multiple roots in the system Nodes are networked in a certain way.
- the networking mode varies according to the actual network conditions. For example, a WAN is usually deployed as a mesh or tree topology, and a LAN is usually deployed as a ring or bus topology. structure.
- each roo The node has at least one alternate root node, the data stored on the alternate root node is the same as the data stored on the corresponding root node, and when root When a node or network fails, the corresponding alternate root node is enabled to take over the data transfer structure of the root node, that is, the alternate root node follows the root The way the node is connected to other nodes and the way the data is transmitted, to connect to other nodes and to transfer data in the same way.
- a lease mechanism can be used to ensure that only a specified number of root nodes are in use at the same time.
- the node of the lease gets the promise of the system, and the role of the node is valid and does not change during the validity period. Issue a lease that is assigned to its root node only for the node acting as the root node The node can maintain the identity of its root node for the lifetime.
- Distribution Manager 101 for managing each root node root
- Each child node of the node can adjust the relationship between the role of the node and the data acquired between the nodes according to the network condition. For example, add new channels to them, receive data for new channels, and tell you to delete channel information and resources. And manage each root Networking mode between nodes. It is also used to set the alternate root node. When the root node fails or the network fails, the alternate root node is enabled to take over the data transmission structure of the root node.
- the first node 103 designated for receiving the distribution manager 101 acquires the data transmission instruction, requesting data from the root node 102, and sends the data request to the specified child node of the second node 103 g 1.
- a first child node 103 specified as the next level of the child nodes of root node 102, the second node are designated l 1 node 105 is the next level node to the first child node specified.
- the designated child nodes are set in the child nodes of each level. To save bandwidth, the number of the designated child nodes is small. Preferably, only one designated child node is set in each level. . The other nodes at the same level as the specified child node are set as spare child nodes 104.
- the standby child node 104 acquires data from the root node 102 and sends it to the designated child node 103 as a supplemental acquisition method for the data.
- the first designated child node 103 determines that the difference between the data amount of the first designated child node itself and the data amount of the root node 102 exceeds the first preset threshold, the data is requested to the standby child node 104 of the same level.
- the alternate child node 104 requests data from the root node 102 and sends it to the first designated child node 103.
- g 1 node determines its own data amount of the difference between the amount of data exceeds a first root node preset threshold value, the data request to the same tier g 2 nodes and / or node g 3, g 2 nodes and / or
- the g 3 node requests data from the root node and sends it to the g 1 node.
- the communication indicator parameter between the first designated child node 103 and the root node 102 is better than the standby child node 104 and the root node 102.
- the communication indicator parameter between the first designated child node 103 and the node 102 is better than the communication status of the spare subsection 104 and the root node 102, and the distribution manager 101
- the node with good communication status is preferentially set to the specified child node.
- the communication indicator parameter is a parameter that identifies the network status, and mainly includes a packet loss rate, a bandwidth, etc., if it is a distribution that needs to access the disk, consider the disk io.
- the pressure can also be the bandwidth of the network line.
- the second designated child node 105 determines that the difference between the data amount of the second designated child node itself and the data amount of the first designated child node 103 exceeds the second preset threshold, then the first designated child node 103
- the spare child node 104 of the same level requests data, and after obtaining the data, it sends it to the destination node 107.
- a node determines the difference between L and g 1 node data amount exceeds the amount of data of its own second preset threshold value, the node requesting data from the same level g 1 g 2 nodes 1 and / or g 3 nodes.
- the second preset threshold may be the same as the first preset threshold.
- the alternate child nodes of each level request data from the designated child nodes of the same level, respectively, without requesting data from any of the parent nodes of the previous level. Only when the relevant instruction is received, or when the specified sub-node of the layer is detected to be faulty, the preset parent node requests data from the upper level, and the requested data is sent to the next layer of the child node according to the preset setting. . There is a list of data relationships between nodes in each node, through which the nodes can learn the data transmission rules.
- the g 2 node and the g 3 node request data from the g 1 node of the same level, and do not request data from the root node of the first level, if the data request instruction of the g 1 node is received, or the g 1 node is detected to be faulty. , request data from the root node.
- l 2 node requests to the same level as the node data l 1, without a level up node g 1, g 2 nodes, g 3 is any node in a data request. If the node receiving the data l 1 request command, or l 1 detected node fails, then, g 2 node, a node g 3 or more requests for data to nodes according to a preset g 1.
- a proxy server is further disposed in the system.
- the distribution manager and each node have communication obstacles due to restrictions such as access, the distribution manager reconnects with each node through the proxy server.
- the above system configuration can reverse operation, the destination node as a root node 107, l 1 node, l 2 is the child node of the next level node to the destination node 107, g 1 node, g 2 nodes, g 3 node is a node l 1, l 2 nodes of the next level of nodes, said data transmission process can flow back from the destination node 107 root node 102, the same principle will not be repeated here.
- the data transmission method is particularly applicable to the transmission of remote streaming media data.
- the method mainly includes the following steps S201 to S203:
- the first designated child node receives an acquiring data instruction sent by the distribution manager.
- Distribution Manager manages each root node, root Each child node of the node can adjust the relationship between the role of the node and the data acquired between the nodes according to the network condition. For example, add new channels to them, receive data for new channels, and tell you to delete channel information and resources. And manage each root Networking mode between nodes. It is also used to set the alternate root node. When the root node fails or the network fails, the alternate root node is enabled to take over the data transmission structure of the root node.
- the first designated child node receives an acquisition data instruction sent by the distribution manager, where the first designated child node is a lower level child node of the root node.
- the designated child nodes are set in the child nodes of each level. To save bandwidth, the number of the designated child nodes is small. Preferably, only one designated child node is set in each level.
- the other nodes at the same level as the specified child node are set as spare child nodes.
- the communication indicator parameter between the designated child node 103 and the root node 102 is better than the alternate subsection 104 and the root node 102.
- the communication indicator parameter between, that is, the communication status of the designated child node 103 and the node 102 is better than the communication status of the spare child node 104 and the root node 102, and the distribution manager 101
- the node with good communication status is preferentially set to the specified child node.
- the communication indicator parameter is a parameter that identifies the network status, and mainly includes a packet loss rate, a bandwidth, etc., if it is a distribution that needs to access the disk, consider the disk io.
- the pressure can also be the bandwidth of the network line.
- the first designated child node requests data from the root node.
- the root node is a data source that stores streaming media data. There are multiple roots in the system Nodes are networked in a certain way.
- the networking mode varies according to the actual network conditions. For example, a WAN is usually deployed as a mesh or tree topology, and a LAN is usually deployed as a ring or bus topology. structure.
- each roo has at least one alternate root node.
- the data of the alternate root node is the same as the corresponding root node.
- the corresponding standby is enabled.
- the root node takes over the data transfer structure of the root node, that is, the alternate root node follows the root The way the node is connected to other nodes and the way the data is transmitted, to connect to other nodes and to transfer data in the same way.
- the lease mechanism can be used to ensure that only a specified number of root nodes are in use at the same time, that is, the lease is obtained.
- the node gets the promise of the system, and the node role is valid and does not change during the validity period. Issue a lease that is assigned to its root node only for the node acting as the root node
- the node can maintain the identity of its root node for the lifetime.
- the first designated child node obtains data from the root node, and the standby child node of the same level as the first designated child node does not
- the root node gets the data. Only when the amount of data acquired by the first designated child node is insufficient, data is requested to the standby child node of the same level, and the standby child node is rooted. The node obtains data and sends it to the first designated child node.
- the first designated child node determines that the difference between the data amount of the first designated child node itself and the data amount of the root node exceeds the first preset threshold
- the data is requested to the standby child node of the same level, and the standby child
- the node requests data from the root node and sends it to the first designated child node.
- FIG. 1 g 1 node determines its own data amount of the difference between the amount of data exceeds a first root node preset threshold value, the data request to the same tier g 2 nodes and / or node g 3, g2 node and
- the / or g3 node requests data from the root node and sends it to the g1 node.
- the first designated child node sends the requested data to the second designated child node, which is the node of the next level of the first designated child node.
- the data of the second designated child node comes from the first designated child node, and does not request data from the spare child node of the same level as the first designated child node.
- the amount of data acquired by the second designated child node is insufficient, data is requested to the standby child node of the same level as the first designated child node, and is sent to the destination node.
- the second designated child node 105 determines that the difference between the data amount of the second designated child node itself and the data amount of the first designated child node 103 exceeds the second preset threshold, the same is performed to the first designated child node 103.
- the alternate child node 104 of the hierarchy requests data. For example, a node determines the difference between L and g 1 node data amount exceeds the amount of data of its own second preset threshold value, the node requesting data from the same level g 1 g 2 nodes 1 and / or g 3 nodes.
- the second preset threshold may be the same as the first preset threshold.
- the alternate child nodes of each level request data from the designated child nodes of the same level, respectively, without requesting data from any of the parent nodes of the previous level. Only when the relevant instruction is received, or when the specified sub-node of the layer is detected to be faulty, the preset parent node requests data from the upper level, and the requested data is sent to the next layer of the child node according to the preset setting. . There is a list of data relationships between nodes in each node, through which the nodes can learn the data transmission rules.
- the g 2 node and the g 3 node request data from the g 1 node of the same level, and do not request data from the root node of the first level, if the data request instruction of the g 1 node is received, or the g 1 node is detected to be faulty. , request data from the root node.
- l 2 node requests to the same level as the node data l 1, without a level up node g 1, g 2 nodes, g 3 is any node in a data request. If the node receiving the data l 1 request command, or l 1 detected node fails, then, g 2 node, a node g 3 or more requests for data to nodes according to a preset g 1.
- FIG. 3 is a schematic structural diagram of a data transmission system according to an embodiment of the present invention, where the system includes:
- a first designated child node 301 a distribution manager 302, a root node 303, a second designated child node 304, and a standby child node 305 , and an alternate root node 306 ;
- the first designated child node 301 is configured to receive the data acquisition instruction sent by the distribution manager 302 to the root node 303. The data is requested and the requested data is sent to the second designated child node 304.
- the first designated child node 301 is the next level child node of the root node 303
- the second designated child node 304 is the first designated child node.
- the communication indicator parameter between the designated child node 103 and the root node 102 is better than the communication indicator parameter between the spare subsection 104 and the root node 102, that is, the designated child node
- the communication status with the node 102 is better than the communication status of the spare subsection 104 with the root node 102, and the distribution manager 101
- the node with good communication status is preferentially set to the specified child node.
- the communication indicator parameter is a parameter that identifies the status of the network.
- the root node 303 is a data source and stores streaming media data. There are multiple roots in the system Nodes are networked in a certain way. The networking mode varies according to the actual network conditions.
- Each level is set with a designated child node. To save bandwidth, the number of designated child nodes is small. Preferably, only one designated child node is set for each level. Other nodes at the same level as the specified child node are set as alternate child nodes 305 When the specified child node works, the standby child node does not work, and the standby child node is started only when the specified child node does not work.
- the communication indicator parameter between the designated child node and the root node is better than the standby child node 305 and the root node. Communication indicator parameters between 303.
- the first designated child node 301 is further configured to: when the first designated child node 301 determines its own data amount and the root node 303 When the difference in the amount of data exceeds the first preset threshold, data is requested to the alternate child node 305 of the same level. Under normal circumstances, the first specified child node is rooted The node obtains data, and the alternate child node of the same level as the first specified child node does not go to the root The node gets the data. Only when the amount of data acquired by the first designated child node is insufficient, data is requested to the standby child node of the same level, and the standby child node is rooted. The node obtains data and sends it to the first designated child node. When judging whether the amount of acquired data is insufficient, by determining its own data amount and the root node 303 Whether the difference in the amount of data exceeds the first preset threshold, and if so, it is determined that the amount of data acquired is insufficient.
- the standby child node 305 is configured to request data from the designated child nodes of the same level respectively.
- the alternate child nodes of each level request data from the designated child nodes of the same level, respectively, without requesting data from any of the parent nodes of the previous level. Only when the relevant instruction is received, or when the specified sub-node of the layer is detected to be faulty, the preset parent node requests data from the upper level, and the requested data is sent to the next layer of the child node according to the preset setting. .
- system further includes:
- a second designated child node 304 configured to: when the second designated child node 304 determines its own data amount and the first designated child node 301 When the difference in the amount of data exceeds the second preset threshold, data is requested to the spare child node of the same level as the first designated child node 301. Under normal circumstances, the data of the second designated child node comes from the first designated child node, and does not request data from the spare child node of the same level as the first designated child node. When the amount of data acquired by the second designated child node is insufficient, data is requested to the standby child node of the same level as the first designated child node, and is sent to the destination node.
- the second preset threshold may be the same as the first preset threshold.
- the system also includes:
- a distribution manager 302 for setting an alternate root node 306 and, when the root node 303 When a failure or network failure occurs, the alternate root node 306 is enabled to take over the data transfer structure of the root node 303.
- the system cannot transmit data.
- At least one spare is set for each roo node.
- the root node, the data of the alternate root node is the same as the corresponding root node, and when the root node or the network fails, the corresponding alternate root node is enabled to take over the
- the data transfer structure of the root node, that is, the alternate root node will follow the root The way the node is connected to other nodes and the way the data is transmitted, to connect to other nodes and to transfer data in the same way.
- the lease mechanism can be used to ensure that only a specified number of roots are available at a time. The node is in use.
Landscapes
- Data Exchanges In Wide-Area Networks (AREA)
Abstract
本发明公开了一种数据传输方法及系统，该方法包括：第一指定子节点接收分发管理器发送的获取数据指令，第一指定子节点为根节点的下一层级子节点，向根节点请求数据并将请求到的数据发送给第二指定子节点，第二指定子节点为第一指定子节点的下一层级的节点，每一层级均设置有指定子节点，与指定子节点同一层级的其他节点均设置为备用子节点，该指定子节点与该根节点之间的通信指标参数优于该备用子节点与该根节点之间的通信指标参数。本发明通过在每一层级子节点中指定数量较少通信状况较好的子节点，作为指定子节点参与数据传输，避免占用过多带宽，提高传输速度，节约网络成本。
Description
本发明属于通信技术领域，尤其涉及一种数据传输方法及系统。
在流媒体相关的数据传输领域，流媒体传播技术与互联网通信技术的结合应用越来越广泛。应用场景常常涉及数据传播距离远，甚至是跨国数据传输，基于分发网络的流媒体传播是必不可少的传播技术。在实际应用中，由于传播距离长，网络环境不够稳定，造成数据传输中存在一些急需解决的难题。
现有技术中，通常采用分层次部署节点，并增强各层节点的数据处理功能，或增加节点数量，来提高数据传输效率，但是上述现有技术中往往也增加了节点处理数据的负担，延长了处理时间，更多节点参与到数据处理传输中，容易受网络带宽的限制，以及受节点部署位置的限制，且导致了由于带宽或设备增加引起的费用成本问题。
本发明提供一种数据传输方法及装置，通过在每一层级子节点中指定数量较少通信状况较好的子节点，作为指定子节点参与数据传输，避免占用过多带宽，提高传输速度，节约网络成本。
本发明提供 一种数据传输方法 ，包括：
第一指定子节点接收分发管理器发送的获取数据指令，所述第一指定子节点为根节点的下一层级子节点；向根节点请求数据，并将请求到的所述数据发送给第二指定子节点，所述第二指定子节点为所述第一指定子节点的下一层级的节点；其中，每一层级均设置有指定子节点，与指定子节点同一层级的其他节点均设置为备用子节点，所述指定子节点与所述根节点之间的通信指标参数优于所述备用子节点与所述根节点之间的通信指标参数。
本发明提供 一种数据传输系统 ，包括：
第一指定子节点，用于接收分发管理器发送的获取数据指令，向根节点请求数据，并将请求到的所述数据发送给第二指定子节点，所述第一指定子节点为根节点的下一层级子节点，所述第二指定子节点为所述第一指定子节点的下一层级的节点；其中，每一层级均设置有指定子节点，与指定子节点同一层级的其他节点均设置为备用子节点，所述指定子节点与所述根节点之间的通信指标参数优于所述备用子节点与所述根节点之间的通信指标参数。
从上述本发明实施例可知，本发明通过在各层子节点中，选择少量通信状况较好的指定子节点作为向上一层级父节点请求数据的节点，而同时其他节点不向上一层级父节点请求数据，以节省带宽，降低成本，提高数据传输速度。
为了更清楚地说明本发明实施例或现有技术中的技术方案，下面将对实施例或现有技术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，对于本领域技术人员来讲，在不付出创造性劳动性的前提下，还可以根据这些附图获得其他的附图
。
图 1 是本发明实施例提 供的 数据传输 方法及系统 的应用环境示意图；
图 2 是本发明第一实施例提 供的 数据传输 方法 的实现流程示意图；
图 3 是本发明第二实施例提 供的 数据传输 系统 的结构示意图。
为使得本发明的发明目的、特征、优点能够更加的明显和易懂，下面将结合本发明实施例中的附图，对本发明实施例中的技术方案进行清楚、完整地描述，显然，所描述的实施例仅仅是本发明一部分实施例，而非全部实施例。基于本发明中的实施例，本领域技术人员在没有做出创造性劳动前提下所获得的所有其他实施例，都属于本发明保护的范围。
请参阅图 1 ，图 1 为本发明实施例提供的数据传输方法及系统的应用环境示意图。
图 1 中示出了具有两层子节点的系统结构，其中， g1 节点、
g2 节点、 g3 节点为第一层子节点， l1 节点、 l2
节点为第二层子节点，各层子节点之间任意两个节点都可以相互传输数据。若在跨国传输的应用场景下， g1 节点、 g2
节点、 g3 节点为国内节点， l1 节点、 l2 节点为国外节点。
该应用场景中，具体包括分发管理器 101 ，根节点（即 root 节点） 102 ，第一指定节点（即
g1 节点） 103 ，与第一指定节点 103 同属于同一层级节点的备用子节点（即 g2 节点、
g3 节点） 104 ，第二指定节点（即 l1 节点） 105 ，与第二指定节点 105
同属于同一层级节点的备用子节点（即 l2 节点） 106 ，以及，最终接收数据的目的节点 107 。
其中， root 节点 102 为数据源，存储有流媒体数据。在系统中有多个 root
节点，以一定的方式组网，组网方式会根据实际网络情况的不同而不同，例如，广域网中通常部署为网状或树型拓扑结构，而在局域网中通常部署为环型或总线型拓扑结构。
进一步地，为防止 root 节点发生故障或者网络故障导致系统无法传输数据，每个 roo
节点都至少设置一个备用 root 节点，该备用 root 节点上存储的数据与对应的 root 节点上存储的数据相同，且当 root
节点或网络发生故障时，会启用对应的备用 root 节点接管该 root 节点的数据传输结构，即，该备用 root 节点会按照该 root
节点与其他节点的连接方式以及数据传输方式，去连接其他节点以及按照相同的方式进行数据传输。
具体可采用 lease （租约）机制保证同一时刻只有指定数量的 root 节点在使用中，获得
lease 的节点得到系统的承诺，在有效期内节点角色是有效的、不会变化的。只为充当 root 节点的节点颁发赋予其 root 节点身份的 lease
，该节点便可在有效期内维持其 root 节点的身份。
分发管理器 101 ，用于管理各 root 节点、 root
节点的各子节点，可根据网络状况调整节点的角色和节点之间获取数据的关系。例如为它们添加新频道、接收新频道的数据、告知删除频道信息和资源。并且管理各 root
节点间的组网方式。还用于设置备用根节点，当根节点发生故障或网络发生故障时，启用该备用根节点接管该根节点的数据传输结构。
第一指定节点 103 ，用于接收分发管理器 101 发送的获取数据指令，向 root 节点 102
请求数据，并将请求到的数据发送给第二指定子节点 g1 节点 103 。第一指定子节点 103 为 root 节点 102
的下一层级子节点，该第二指定子节点 l1 节点 105 为第一指定子节点的下一层级的节点。
在本实施例的数据传输系统中，每一层级的子节点中均设置有指定子节点，为节省带宽，该指定子节点的数量较少，优选地，每一层级只设置一个指定子节点 103
。与该指定子节点同一层级的其他节点均设置为备用子节点 104 。
在正常情况下，只有第一指定子节点 103 向 root 节点 102 获取数据，其备用子节点 104
不向 root 节点 102 获取数据，只有当第一指定子节点 103 获取的数据量不足时，备用子节点 104 才向根节点 102 获取数据并发送给指定子节点
103 作为数据的补充获取方式。具体地，当第一指定子节点 103 判断第一指定子节点自身的数据量与 root 节点 102
的数据量的差值超过第一预置临界值时，向同一层级的备用子节点 104 请求数据，备用子节点 104 向 root 节点 102 请求数据，并发送给第一指定子节点
103 。例如， g1 节点判断它自身的数据量与 root 节点的数据量的差值超过第一预置临界值，则向同一层级
g2 节点和 / 或 g3 节点请求数据， g2 节点和 / 或 g3
节点向 root 节点请求数据，并发送给 g1 节点。
第一指定子节点 103 与根节点 102 之间的通信指标参数，优于备用子节点 104 与根节点 102
之间的通信指标参数，即，第一指定子节点 103 与跟节点 102 的通信状况要好于备用子节 104 与根节点 102 的通信状况，分发管理器 101
会优先选择通信状况好的节点设置为指定子节点。该通信指标参数是标识网络状况的参数，主要包括丢包率、带宽等，如果是需要访问磁盘的分发，要考虑磁盘 io
的压力，还可以是网络线路的带宽资费。
进一步地，当第二指定子节点 105 判断第二指定子节点自身的数据量与第一指定子节点 103
的数据量的差值超过第二预置临界值时，则与第一指定子节点 103 同一层级的备用子节点 104 请求数据，获取到数据后发送给目的节点 107 。例如，
l1 节点判断它自身的数据量与 g1 节点的数据量的差值超过第二预置临界值，则向 g1 节点
1 的同一层级 g2 节点和 / 或 g3 节点请求数据。该第二预置临界值可以与该第一预置临界值相同。
在正常通信状况下，各层级的备用子节点分别向同一层级的指定子节点请求数据，而不与上一层级的任何一个父节点请求数据。只有当接收到相关指令，或者，检测到本层指定子节点发生故障时，则向上一层级的预置的父节点请求数据，并将请求到的数据按照预先的设置发送给下一层子节点。在各节点中均有一个节点之间传输数据关系的列表，通过该列表，节点可获知数据传输规则。例如，
g2 节点、 g3 节点向同一层级的 g1 节点请求数据，而不向上一层级的 root
节点请求数据，若接收到 g1 节点的数据请求指令，或者，检测到 g1 节点发生故障，则向 root
节点请求数据。再例如， l2 节点向同一层级的 l1 节点请求数据，而不向上一层级的 g1
节点、 g2 节点、 g3 节点中的任何一个请求数据。若接收到 l1
节点的数据请求指令，或者，检测到 l1 节点发生故障，则按照预先设置向 g1 节点、 g2
节点、 g3 节点中的一个或多个请求数据。
另外，在该应用场景下，系统中还设置有代理服务器，当分发管理器与各节点间因为访问受限制等原因造成通信障碍时，分发管理器通过代理服务器与各节点重新连接通信。
需要说明的是，以上系统结构可以反向操作，将目的节点 107 当作 root 节点，
l1 节点、 l2 节点为目的节点 107 下一层级的子节点， g1 节点、
g2 节点、 g3 节点为 l1 节点、 l2
节点的下一层级的节点，上述数据传输过程可从目的节点 107 流回 root 节点 102 ，原理相同，此处不再赘述。
请参阅图 2 ，图 2
为本发明实施例提供的数据传输方法的实现流程示意图，该数据传输方法尤其适用于远程流媒体数据的传输，该方法主要包括以下步骤 S201 至步骤 S203 ：
S201 、第一指定子节点接收分发管理器发送的获取数据指令；
分发管理器管理各 root 节点、 root
节点的各子节点，可根据网络状况调整节点的角色和节点之间获取数据的关系。例如为它们添加新频道、接收新频道的数据、告知删除频道信息和资源。并且管理各 root
节点间的组网方式。还用于设置备用根节点，当根节点发生故障或网络发生故障时，启用该备用根节点接管该根节点的数据传输结构。
第一指定子节点接收分发管理器发送的获取数据指令，该第一指定子节点为根节点的下一层级子节点。每一层级的子节点中均设置有指定子节点，为节省带宽，该指定子节点的数量较少，优选地，每一层级只设置一个指定子节点。与该指定子节点同一层级的其他节点均设置为备用子节点，在指定子节点工作时，备用子节点是不工作的，只有指定子节点不工作时，才启动备用子节点。
指定子节点 103 与根节点 102 之间的通信指标参数，优于备用子节 104 与根节点 102
之间的通信指标参数，即，指定子节点 103 与跟节点 102 的通信状况要好于备用子节 104 与根节点 102 的通信状况，分发管理器 101
会优先选择通信状况好的节点设置为指定子节点。该通信指标参数是标识网络状况的参数，主要包括丢包率、带宽等，如果是需要访问磁盘的分发，要考虑磁盘 io
的压力，还可以是网络线路的带宽资费。
S202 、向根节点请求数据；
第一指定子节点向根节点请求数据。
root 节点为数据源，存储有流媒体数据。在系统中有多个 root
节点，以一定的方式组网，组网方式会根据实际网络情况的不同而不同，例如，广域网中通常部署为网状或树型拓扑结构，而在局域网中通常部署为环型或总线型拓扑结构。
进一步地，为防止 root 节点发生故障或者网络故障导致系统无法传输数据，每个 roo
节点都至少设置一个备用 root 节点，该备用 root 节点的数据与对应的 root 节点相同，且当 root 节点或网络发生故障时，会启用对应的备用
root 节点接管该 root 节点的数据传输结构，即，该备用 root 节点会按照该 root
节点与其他节点的连接方式以及数据传输方式，去连接其他节点以及按照相同的方式进行数据传输。
具体可采用 lease 机制保证同一时刻只有指定数量的 root 节点在使用中，即获得 lease
的节点得到系统的承诺，在有效期内节点角色是有效的、不会变化的。只为充当 root 节点的节点颁发赋予其 root 节点身份的 lease
，该节点便可在有效期内维持其 root 节点的身份。
在正常情况下，第一指定子节点向 root 节点获取数据，而与第一指定子节点同一层级的备用子节点不向
root 节点获取数据。只有当第一指定子节点获取的数据量不足时，才向其同一层级的备用子节点请求数据，该备用子节点向 root
节点获取数据并发送给该第一指定子节点。
具体地，当第一指定子节点判断第一指定子节点自身的数据量与 root
节点的数据量的差值超过第一预置临界值时，向同一层级的备用子节点请求数据，该备用子节点向 root 节点请求数据，并发送给第一指定子节点。例如，图 1 中
g1 节点判断它自身的数据量与 root 节点的数据量的差值超过第一预置临界值，则向同一层级 g2 节点和 / 或
g3 节点请求数据， g2 节点和 / 或 g3 节点向 root 节点请求数据，并发送给 g1 节点。
S203 、将请求到的数据发送给第二指定子节点。
第一指定子节点将请求到的数据发送给第二指定子节点，该第二指定子节点为该第一指定子节点的下一层级的节点。
在正常情况下，第二指定子节点的数据来自于第一指定子节点，而不会向与第一指定子节点同一层级的备用子节点请求数据。当第二指定子节点获取的数据量不足时，向与第一指定子节点同一层级的备用子节点请求数据，并发送给目的节点。
具体地，当第二指定子节点 105 判断第二指定子节点自身的数据量与第一指定子节点 103
的数据量的差值超过第二预置临界值时，向第一指定子节点 103 同一层级的备用子节点 104 请求数据。例如， l1
节点判断它自身的数据量与 g1 节点的数据量的差值超过第二预置临界值，则向 g1 节点 1 的同一层级
g2 节点和 / 或 g3 节点请求数据。该第二预置临界值可以与该第一预置临界值相同。
在正常通信状况下，各层级的备用子节点分别向同一层级的指定子节点请求数据，而不与上一层级的任何一个父节点请求数据。只有当接收到相关指令，或者，检测到本层指定子节点发生故障时，则向上一层级的预置的父节点请求数据，并将请求到的数据按照预先的设置发送给下一层子节点。在各节点中均有一个节点之间传输数据关系的列表，通过该列表，节点可获知数据传输规则。例如，
g2 节点、 g3 节点向同一层级的 g1 节点请求数据，而不向上一层级的 root
节点请求数据，若接收到 g1 节点的数据请求指令，或者，检测到 g1 节点发生故障，则向 root
节点请求数据。再例如， l2 节点向同一层级的 l1 节点请求数据，而不向上一层级的 g1
节点、 g2 节点、 g3 节点中的任何一个请求数据。若接收到 l1
节点的数据请求指令，或者，检测到 l1 节点发生故障，则按照预先设置向 g1 节点、 g2
节点、 g3 节点中的一个或多个请求数据。
本实施例中，通过在各层子节点中，选择少量通信状况较好的指定子节点作为向上一层级父节点请求数据的节点，而同时其他节点不向上一层级父节点请求数据，以节省带宽，降低成本，提高数据传输速度。
请参阅图 3 ，图 3 为本发明实施例提供的数据传输系统的结构示意图，该系统包括：
第一指定子节点 301 、分发管理器 302 、根节点 303 、第二指定子节点 304 、备用子节点
305 ，以及备用根节点 306 ；
其中，第一指定子节点 301 ，用于接收分发管理器 302 发送的获取数据指令，向根节 303
请求数据，并将请求到的数据发送给第二指定子节点 304 。
第一指定子节点 301 为根节点 303 的下一层级子节点，第二指定子节点 304 为第一指定子节点
301 的下一层级的节点。指定子节点 103 与根节点 102 之间的通信指标参数，优于备用子节 104 与根节点 102 之间的通信指标参数，即，指定子节点
103 与跟节点 102 的通信状况要好于备用子节 104 与根节点 102 的通信状况，分发管理器 101
会优先选择通信状况好的节点设置为指定子节点。该通信指标参数是标识网络状况的参数。
根节点 303 为数据源，存储有流媒体数据。在系统中有多个 root
节点，以一定的方式组网，组网方式会根据实际网络情况的不同而不同。
每一层级均设置有指定子节点，为节省带宽，该指定子节点的数量较少，优选地，每一层级只设置一个指定子节点。与指定子节点同一层级的其他节点均设置为备用子节点 305
，在指定子节点工作时，备用子节点是不工作的，只有指定子节点不工作时，才启动备用子节点。该指定子节点与该根节点之间的通信指标参数优于备用子节点 305 与根节点
303 之间的通信指标参数。
进一步地，第一指定子节点 301 ，还用于当第一指定子节点 301 判断其自身的数据量与根节点 303
的数据量的差值超过第一预置临界值时，向同一层级的备用子节点 305 请求数据。 在正常情况下，第一指定子节点向 root
节点获取数据，而与第一指定子节点同一层级的备用子节点不向 root
节点获取数据。只有当第一指定子节点获取的数据量不足时，才向其同一层级的备用子节点请求数据，该备用子节点向 root
节点获取数据并发送给该第一指定子节点。判断获取到的数据量是否不足时，通过判断其自身的数据量与根节点 303
的数据量的差值是否超过第一预置临界值，若是，则确定获取到的数据量不足。
进一步地，备用子节点 305 ，用于分别向同一层级的指定子节点请求数据。
在正常通信状况下，各层级的备用子节点分别向同一层级的指定子节点请求数据，而不与上一层级的任何一个父节点请求数据。只有当接收到相关指令，或者，检测到本层指定子节点发生故障时，则向上一层级的预置的父节点请求数据，并将请求到的数据按照预先的设置发送给下一层子节点。
进一步地，该系统还包括：
第二指定子节点 304 ，用于当第二指定子节点 304 判断其自身的数据量与第一指定子节点 301
的数据量的差值超过第二预置临界值时，向与第一指定子节点 301 同一层级的备用子节点请求数据。
在正常情况下，第二指定子节点的数据来自于第一指定子节点，而不会向与第一指定子节点同一层级的备用子节点请求数据。当第二指定子节点获取的数据量不足时，向与第一指定子节点同一层级的备用子节点请求数据，并发送给目的节点。判断获取到的数据量是否不足时，通过判断其自身的数据量与第一指定子节点
301 的数据量的差值是否超过第二预置临界值，若是，则确定获取到的数据量不足。该第二预置临界值可以与该第一预置临界值相同。
该系统还包括：
分发管理器 302 ，用于设置备用根节点 306 ，以及，当根节点 303
发生故障或网络发生故障时，启用备用根节点 306 接管根节点 303 的数据传输结构。
为防止 root 节点发生故障或者网络故障导致系统无法传输数据，每个 roo 节点都至少设置一个备用
root 节点，该备用 root 节点的数据与对应的 root 节点相同，且当 root 节点或网络发生故障时，会启用对应的备用 root 节点接管该
root 节点的数据传输结构，即，该备用 root 节点会按照该 root
节点与其他节点的连接方式以及数据传输方式，去连接其他节点以及按照相同的方式进行数据传输。可采用 lease 机制保证同一时刻只有指定数量的 root
节点在使用。
本实施例中其他未尽细节，请参见前述图 2 所示实施例中的相关描述。
本实施例中，通过在各层子节点中，选择少量通信状况较好的指定子节点作为向上一层级父节点请求数据的节点，而同时其他节点不向上一层级父节点请求数据，以节省带宽，降低成本，提高数据传输速度。
需要说明的是，对于前述的各方法实施例，为了简便描述，故将其都表述为一系列的动作组合，但是本领域技术人员应该知悉，本发明并不受所描述的动作顺序的限制，因为依据本发明，某些步骤可以采用其它顺序或者同时进行。其次，本领域技术人员也应该知悉，说明书中所描述的实施例均属于优选实施例，所涉及的动作和模块并不一定都是本发明所必须的。
在上述实施例中，对各个实施例的描述都各有侧重，某个实施例中没有详述的部分，可以参见其它实施例的相关描述。
以上为对本发明所提供的一种数据传输方法及系统的描述，对于本领域的技术人员，依据本发明实施例的思想，在具体实施方式及应用范围上均会有改变之处，综上，本说明书内容不应理解为对本发明的限制。
Claims (10)
- 一种数据传输方法，其特征在于，所述方法包括：第一指定子节点接收分发管理器发送的获取数据指令，所述第一指定子节点为根节点的下一层级子节点；向根节点请求数据，并将请求到的所述数据发送给第二指定子节点，所述第二指定子节点为所述第一指定子节点的下一层级的节点；其中，每一层级均设置有指定子节点，与指定子节点同一层级的其他节点均设置为备用子节点，所述指定子节点与所述根节点之间的通信指标参数优于所述备用子节点与所述根节点之间的通信指标参数。
- 根据权利要求 1 所述的方法，其特征在于，所述方法还包括：当所述第一指定子节点判断其自身的数据量与所述根节点的数据量的差值超过第一预置临界值时，向同一层级的备用子节点请求数据。
- 根据权利要求 1 或 2 所述的方法，其特征在于，所述方法还包括：各层级的所述备用子节点分别向同一层级的指定子节点请求数据。
- 根据权利要求 3 所述的方法，其特征在于，所述方法还包括：当所述第二指定子节点判断其自身的数据量与所述第一指定子节点的数据量的差值超过第二预置临界值时，向与所述第一指定子节点同一层级的备用子节点请求数据。
- 根据权利要求 1 所述的方法，其特征在于，所述方法还包括：分发管理器设置备用根节点；当所述根节点发生故障或网络发生故障时，启用所述备用根节点接管所述根节点的数据传输结构。
- 一种数据传输系统，其特征在于，所述系统包括：第一指定子节点，用于接收分发管理器发送的获取数据指令，向根节点请求数据，并将请求到的所述数据发送给第二指定子节点，所述第一指定子节点为根节点的下一层级子节点，所述第二指定子节点为所述第一指定子节点的下一层级的节点；其中，每一层级均设置有指定子节点，与指定子节点同一层级的其他节点均设置为备用子节点，所述指定子节点与所述根节点之间的通信指标参数优于所述备用子节点与所述根节点之间的通信指标参数。
- 根据权利要求 6 所述的系统，其特征在于，所述第一指定子节点，还用于当所述第一指定子节点判断其自身的数据量与所述根节点的数据量的差值超过第一预置临界值时，向同一层级的备用子节点请求数据。
- 根据权利要求 6 或 7 所述的系统，其特征在于，所述系统还包括：备用子节点，用于分别向同一层级的指定子节点请求数据。
- 根据权利要求 8 所述的系统，其特征在于，所述系统还包括：第二指定子节点，用于当所述第二指定子节点判断其自身的数据量与所述第一指定子节点的数据量的差值超过第二预置临界值时，向与所述第一指定子节点同一层级的备用子节点请求数据。
- 根据权利要求 9 所述的系统，其特征在于，所述系统还包括：分发管理器，用于设置备用根节点，当所述根节点发生故障或网络发生故障时，启用所述备用根节点接管所述根节点的数据传输结构。
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| PCT/CN2015/088768 WO2017035789A1 (zh) | 2015-09-01 | 2015-09-01 | 一种数据传输方法及系统 |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| PCT/CN2015/088768 WO2017035789A1 (zh) | 2015-09-01 | 2015-09-01 | 一种数据传输方法及系统 |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| WO2017035789A1 true WO2017035789A1 (zh) | 2017-03-09 |
Family
ID=58186484
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| PCT/CN2015/088768 WO2017035789A1 (zh) | 2015-09-01 | 2015-09-01 | 一种数据传输方法及系统 |
Country Status (1)
| Country | Link |
|---|---|
| WO (1) | WO2017035789A1 (zh) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111182036A (zh) * | 2019-12-12 | 2020-05-19 | 腾讯云计算（北京）有限责任公司 | 数据分流方法及网络构建方法、装置、设备、存储介质 |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101146021A (zh) * | 2006-09-11 | 2008-03-19 | 思华科技（上海）有限公司 | 内容分发网络的路由系统及方法 |
| CN101399713A (zh) * | 2008-09-28 | 2009-04-01 | 腾讯科技（深圳）有限公司 | 测量网络节点邻近度的方法 |
| CN101414949A (zh) * | 2008-11-28 | 2009-04-22 | 阿里巴巴集团控股有限公司 | 一种链式数据传输方法、节点及系统 |
| CN101499914A (zh) * | 2008-01-28 | 2009-08-05 | 华为技术有限公司 | 一种组播系统的父节点选择方法、系统及节点 |
| CN101583161A (zh) * | 2009-05-26 | 2009-11-18 | 北京邮电大学 | 协作节点单元选择方法及装置 |
| CN101594292A (zh) * | 2008-05-30 | 2009-12-02 | 中兴通讯股份有限公司 | 内容发布方法、服务重定向方法及系统、节点设备 |
-
2015
- 2015-09-01 WO PCT/CN2015/088768 patent/WO2017035789A1/zh active Application Filing
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101146021A (zh) * | 2006-09-11 | 2008-03-19 | 思华科技（上海）有限公司 | 内容分发网络的路由系统及方法 |
| CN101499914A (zh) * | 2008-01-28 | 2009-08-05 | 华为技术有限公司 | 一种组播系统的父节点选择方法、系统及节点 |
| CN101594292A (zh) * | 2008-05-30 | 2009-12-02 | 中兴通讯股份有限公司 | 内容发布方法、服务重定向方法及系统、节点设备 |
| CN101399713A (zh) * | 2008-09-28 | 2009-04-01 | 腾讯科技（深圳）有限公司 | 测量网络节点邻近度的方法 |
| CN101414949A (zh) * | 2008-11-28 | 2009-04-22 | 阿里巴巴集团控股有限公司 | 一种链式数据传输方法、节点及系统 |
| CN101583161A (zh) * | 2009-05-26 | 2009-11-18 | 北京邮电大学 | 协作节点单元选择方法及装置 |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111182036A (zh) * | 2019-12-12 | 2020-05-19 | 腾讯云计算（北京）有限责任公司 | 数据分流方法及网络构建方法、装置、设备、存储介质 |
| CN111182036B (zh) * | 2019-12-12 | 2023-07-25 | 腾讯云计算（北京）有限责任公司 | 数据分流方法及网络构建方法、装置、设备、存储介质 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| KR101341272B1 (ko) | 다지점 링크를 포함하는 지점간 통합 논리 데이터 링크 | |
| US9614788B2 (en) | All delivered network switch | |
| CN108199897B (zh) | 一种支持缓存管理的opc ua多服务器聚合方法 | |
| EP3958509A1 (en) | Method, apparatus and system for communication between controllers in tsn | |
| US10374895B2 (en) | Configuration of communication interfaces for link aggregation | |
| CN109391614B (zh) | 一种基于视联网调取监控数据的方法和装置 | |
| WO2020063991A1 (zh) | Pon网络，用于pon网络的方法及装置，以及机器人系统 | |
| CN107615721B (zh) | 传输软件定义网络-逻辑链路聚合成员信令的系统和方法 | |
| JP7144539B2 (ja) | ネットワークトポロジ生成方法および装置 | |
| WO2015013896A1 (zh) | 一种网络控制方法及装置 | |
| US20140226525A1 (en) | Safe Multicast Distribution with Predictable Topology Changes | |
| US20160277279A1 (en) | Link discovery method, system, and device | |
| JP2010161501A (ja) | ネットワークの管理システム及び管理方法 | |
| CN109067633B (zh) | 基于以太网菊花链通讯网络拓扑的电源管理系统及方法 | |
| EP3588859A1 (en) | Network device configuration versioning | |
| JP4559512B2 (ja) | パケット転送システムおよびパケット転送方法 | |
| CN109802879B (zh) | 一种数据流路由方法及装置 | |
| WO2019170085A1 (zh) | 一种故障诊断方法及其装置 | |
| WO2017035789A1 (zh) | 一种数据传输方法及系统 | |
| CN111585791A (zh) | 一种数据同步配置方法、系统及存储介质 | |
| TW201817196A (zh) | 電腦叢集系統 | |
| US20190268235A1 (en) | Method for managing network nodes and communication control method thereof | |
| CN115514651A (zh) | 基于软件定义层叠网的云边数据传输路径规划方法及系统 | |
| Li et al. | Towards centralized and semi‐automatic VLAN management | |
| CN109194503B (zh) | 一种利用ospf报文分配站点ip的方法 |
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: 15902596 Country of ref document: EP Kind code of ref document: A1 |
|
| NENP | Non-entry into the national phase |
Ref country code: DE |
|
| 122 | Ep: pct application non-entry in european phase |
Ref document number: 15902596 Country of ref document: EP Kind code of ref document: A1 |