WO2009030080A1

WO2009030080A1 - A handover processing method in the cdn agent

Info

Publication number: WO2009030080A1
Application number: PCT/CN2007/003508
Authority: WO
Inventors: Yan Liu
Original assignee: Zte Corporation
Priority date: 2007-09-04
Filing date: 2007-12-10
Publication date: 2009-03-12
Also published as: CN100589452C; CN101119318A

Abstract

A handover processing method in the CDN agent is provided by this invention, including: setting an information table and/or a comparison table, wherein the information table notes the information of each agent in CDN, and the comparison table notes each node of CDN and/or the identifier information of the master CDN agent in the node (S202); traversing the information table and/or the comparison table, for the node that the master CDN agent with link status abnormity belongs to and/or the node that the identifier information of the master CDN agent is blank, initiating the master handover request (S204). Through this invention, the handover process of multi agents in IPTV is implemented.

Description

TECHNICAL FIELD The present invention relates to the field of telecommunications services IPTV (Internet Television), and in particular to a handover processing method for a streaming media node controller (Agent). Background technique

IPTV is an interactive network TV based on broadband high-speed IP network with network video resources as the main body and TV, computer and other display terminals. It integrates various technologies such as Internet, multimedia and communication, and can provide digital TV programs. Many services such as on-demand programming, online information consultation, and visual IP telephony. Since IPTV makes full use of the respective advantages and characteristics of telecommunications, television and internet, it effectively realizes the substantive interaction between media providers and media consumers. Therefore, it will become the next promising value-added service in the telecom industry. , with huge market potential. The Content Distribution Network (CDN Subsystem) is responsible for the distribution control of content and the push of content to edge service nodes to improve the quality of content services in IPTV. The CDN subsystem is mainly composed of a CDN controller (CDN Manager) and a streaming media node controller (CDN Agent). The CDN Manager is responsible for accessing the IPTV service management system, and the CDN Agent and the streaming media system together form a distributed streaming media node. Each agent of each node has only one agent at a certain time, and the agent acting as the scheduling agent is the main agent, and the other agent is the standby agent. The main agent or the standby agent has the label in the Manager. (EPG) and the streaming media system only maintain the heartbeat with the main agent, so only the main agent can perform operations such as redirection and other services for the user. When the main agent and the Manager are broken or the main agent itself is broken, the Manager looks for a status. A good Agent is used as the new main agent, so that it will not cause the user to provide services due to the Agent failure. For the normal communication of the message, the identity of the master/slave agent in the Manager must be consistent with the identity of the Agent itself. If the Agents coexist and cannot communicate with each other, the Manager must be able to perform accurate secondary planting and identification confirmation. There can be no '1·green conditions of multiple primary agents under the same node, nor can there be no primary agent under the node but Manager It does not trigger the search for the main Agent. However, it has not been proposed yet to achieve the above. Technical effects program.

1 P16849 SUMMARY OF THE INVENTION The present invention has been made in view of the above problems in the related art, and the present invention is directed to a switching processing method of a streaming media node controller (Agent). The streaming media node controller switching processing method according to the present invention is used for streaming media node controller switching in a content distribution network, wherein a streaming media node of the content distribution network has a plurality of streaming media node controllers, and at a certain moment Only one streaming node controller acts as a primary media node controller for scheduling purposes. The method includes: setting an information table and/or a comparison table, wherein the information table records information of each streaming node controller of the content distribution network, where the information includes a node to which the streaming node controller belongs, a working mode, and a link state. The comparison table records the identification information of each node of the content distribution network and/or the mainstream media node controller of the node; the traversal information table and/or the comparison table, for the mainstream media node controller whose link state is abnormal The node and/or the mainstream media node controller's identification information is an empty node, and a primary handover request is initiated. The process of initiating the main handover request by traversing the information table is specifically: setting a first timer, triggering each streaming node controller in the traversal information table, and determining whether the streaming node controller determined to be a link abnormality is The mainstream media node controller, if the determination result is yes, determines the node to which the mainstream media node controller belongs, and selects a streaming media node controller whose link state is normal to send a primary handover request, to request that the switch is Mainstream media node controller. In the above method, traversing is performed by transmitting a heartbeat request to each of the streaming node controllers in the information table, and a third timer is set, wherein, according to whether the streaming node controller returns to the heart before the third timer expires A mega-response is used to determine if the link is normal. Based on this, the streaming node controller that has no heartbeat response after the third timer expires for a predetermined predetermined number of times determines that it is a streaming node controller whose link is abnormal. Further, for the streaming node controller determined to be abnormal in the link, the link state corresponding to the link recorded in the information table is set to be abnormal. The process of initiating the main handover request by traversing the comparison table is specifically: setting a second timer, triggering each node in the traversal comparison table, and selecting a link of the node for the node whose identification information of the mainstream media node controller is empty The streaming node controller with the normal status sends a primary handover request to request that it switch to the mainstream media node controller. In the above method, the fourth timer is set when the main handover request is transmitted. It should be noted that the additional information of the third timer/fourth timer carries the base of the associated streaming media node controller.

2 P16849 This information is used to identify the streaming node controller. In this way, on the one hand, if the streaming node controller that is requested to switch returns a handover response before the fourth timer expires, the fourth timer is turned off, and it is determined whether the handover is successful; if the handover is successful, the streaming node is As a new mainstream media node controller, the controller sets its corresponding working mode in the information table as the main working mode, and saves its identifier to the comparison table; in the case of the switching failure, the streaming node controller is The corresponding working mode in the information table is set to the slave working mode, the corresponding mainstream media node controller identifier in the matching table is set to be empty, and the link state of the node to which the node belongs is controlled to be normal. The sender sends a master switch request. In addition, in the case that the handover is successful, the slave handover control request is further sent to all other media control pointcuts under the node to which the new mainstream media node controller belongs to request that the handover be from the streaming node controller. Then, for the case where the handover response is not returned or the handover fails, the slave slave switching request is continued; for the case where the handover succeeds, the corresponding working mode of the other stream media control pointcuts in the information table is set to the slave mode. On the other hand, if the streaming node controller that is requested to switch does not return the switching response before the fourth timer expires, the streaming node controller sets the corresponding working mode in the information table to the slave working mode, and belongs to it. The node sets the corresponding mainstream media node controller identifier in the comparison table to be empty, and reselects the streaming node controller whose link state is normal to send the primary handover request. The invention realizes the switching process of the multi-agent in the IPTV, effectively arbitrates the function of the mutual agent of the multiple agents under the node to provide the monthly service, and ensures that the situation of the unowned agent and the multiple main agents does not occur, so that each The system can stably provide various media services to users, thus ensuring system stability and post-maintenance quality, and reducing maintenance costs. Other features and advantages of the invention will be set forth in the description which follows, and The objectives and other advantages of the invention will be realized and attained by the <RTI The drawings are intended to provide a further understanding of the invention, and are intended to be a part of the description of the invention. In the drawing:

3 P 16849 1 is a system framework diagram of an IPT V CDN subsystem to which the present invention is applied; FIG. 2 is a schematic flowchart of an agent handover processing method according to an embodiment of the present invention; FIG. 3 is an agent handover processing method according to an embodiment of the present invention; FIG. 4 is a flowchart of a process after triggering a handover request in an agent handover processing method according to an embodiment of the present invention. FIG. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS The preferred embodiments of the present invention are described with reference to the accompanying drawings. The present invention aims to provide a switching processing method for a streaming media node controller (Agent).

Switching, wherein, as shown in FIG. 1, the CDN may include multiple streaming media nodes (hereinafter referred to as nodes), that is, one Manager may control multiple nodes, and each node may have multiple agents, and at some Only one agent acts as the primary agent at a time, and other agents are called standby agents or slave agents. Since each agent may become the primary agent, each agent and its own node are guaranteed to ensure that the system can provide services normally. All the streaming media under the link. The idea of the method according to the embodiment of the present invention is roughly as follows, that is, maintaining (storing) the information table and the comparison table, determining whether to initiate a handover request by traversing the information/reference table, and then following the response result to the handover request. deal with. Specifically, as shown in FIG. 2, the method may include the following processing: Step S202, setting an information table and/or a comparison table, where the information table records information of each agent of the CDN (including the node to which the agent belongs, the working mode Basic information such as link status, communication parameters, IP address, etc.; The comparison table records the identification information of each node of the CDN and/or the primary agent of the node (when a node does not have a primary agent, the corresponding primary agent identifier Empty, wherein the information stored in the information table and the comparison table can be obtained from the database synchronization after the system is restarted; Step S204, traversing the information table and/or the comparison table, for the mainstream media whose link status is abnormal

4 P16849 The node to which the node controller belongs and/or the identifier information of the mainstream media node controller is an empty node, and a primary handover request is initiated. On the one hand, in step S204, the process of initiating the main handover request by traversing the information table is specifically: setting a first timer, triggering each agent in the traversal information table, that is, according to the communication parameters of the Agent in the information table The agent sends a heartbeat request and sets a third timer. If an agent returns a heart/signal response before the third timer expires, the agent is considered to be normal and the timer is turned off; if an agent is scheduled for a predetermined number of times (may Pre-set) If there is no 3 mega-response after the third timer expires, the agent link is considered abnormal; for the agent determined to be abnormal, the link status recorded in the information table is set to abnormal. And determine whether it is the primary agent. If the judgment result is yes, determine the node to which the primary agent belongs, clear the corresponding primary agent identifier of the node in the comparison table, and select one of the node's link states to be sent by the normal agent. Master switching request to request that it switch to the primary agent, and set a fourth timer; on the other hand, in step S204, by traversing the comparison table The processing of the primary handover request is specifically as follows: setting a second timer, triggering each node in the traversal comparison table, and selecting a node with a normal link status of the node to be a normal node. The request is switched to request that it be switched to the primary agent, and the fourth timer is set. It should be noted that the durations of the foregoing first to fourth timers are flexibly configured according to requirements, to adapt to different application environments or to set an optimal value according to actual conditions, for the third timer/fourth timer, The agent, the set of the heart and the timeout timer are all the third timer/fourth timer. Therefore, in order to distinguish which agent is, it is necessary to carry the related agent in the additional information of setting the timer. Basic information to identify or distinguish Agents.

Described above is the process of transmitting a handover request and transmitting a handover request. The above processing can be better understood by the detailed processing flowchart shown in FIG. As shown in Figure 3, and as mentioned above, there are two situations that may trigger the Agent switchover, that is, the primary Agent is broken or has no primary agent: Case 1: The primary Agent is broken, you can perform the following processing: Step 1: Set The timer A (301), the timer A (ie, the first timer) triggers to traverse the Agent information table (302) stored in the memory, where the information table includes the section to which the Agent belongs.

5 P 16849 Basic information such as point, communication parameters, IP address, working mode, link status, etc. Step 2: According to the communication parameters, the Manager initiates a heartbeat request to all Agents (303), and then sets a timeout timer C (ie, the third Timer) (304); Step 3: Determine whether the requested Agent returns the heart 5 response (305). If the response is returned before the timer C times out, the timer C is closed (306), and it is determined whether the working mode reported by the agent is consistent with the saved in the Manager memory. If they are consistent, the process ends (317); if not, the manager The Agent initiates a switch to a consistent handover request (308) and sets a timer D (316) awaiting the handover result. If the agent does not return the heartbeat response after the timer C expires, it is determined that the agent is broken, and the link state of the agent in the memory is set to abnormal (309), and it is determined whether the broken chain agent is The current master agent (310), if not, the process ends (317); if yes, the node (314) to which the agent belongs is obtained, and an agent with a normal state under the node is searched as the master agent, and the switch is sent to The master agent's handover request (315), and sets a timer D (316) waiting for the handover result. Case 2, without the main Agent, the following processing can be performed: Step 1: Set the timer B ( 311 ), the timer B triggers to traverse the node in the memory and the master agent comparison table ( 312 ); Step 2: Determine the node in the comparison Whether there is a corresponding main agent (313) in the table, if yes, the process ends (317); if not, the node (314) to which the agent belongs is obtained, and an agent with a normal state under the node is searched as the main agent, It transmits a handover request (315) that is switched to the primary agent, and sets a timer D (316) waiting for the handover result. Based on the above description, the subsequent processing of the embodiment of the present invention after the handover request is transmitted will be further described. On the one hand, if the agent that is requested to switch returns a handover response before the fourth timer expires, the fourth timer is closed, and the handover is successful; if the handover is successful, the agent is used as the new primary agent, and The corresponding working mode in the information table is set as the main working mode, and its identifier is saved to the comparison table; in the case of the switching failure, the corresponding working mode of the agent in the information table is set to the working mode, and the node to which it belongs is set. In the comparison table, the corresponding primary agent ID is set to null, and the link whose status of the node to which the node belongs is normal is sent.

6 P 16849 begging. In addition, in the case that the handover is successful, the slave handover control request is further sent to all other streaming media control points under the node to which the new primary agent belongs to request that it switch to the secondary agent. Then, for the case where the handover response is not returned or the handover fails, the slave slave switching request is continued; for the case where the handover succeeds, the corresponding working mode of the other stream media control pointcuts in the information table is set to the slave mode. On the other hand, if the agent requested to switch does not return the handover response before the fourth timer expires, the corresponding working mode of the Agent in the information table is set to the slave working mode, and the node to which the node belongs is in the corresponding table in the comparison table. The Agent ID is set to null, and the link of the node to which it belongs is reselected ^! The large state sends a primary handover request to the normal Agent. In fact, when receiving the handover response or the handover result, the Manager needs to determine whether the response responds to the request to switch to the primary agent, or the response to the request to switch from the agent, and then perform subsequent processing according to the determination result. The content can be further understood as shown in Figure 4. 4 is a flowchart showing the processing flow after receiving the handover response in detail, and the present invention will be further described below with reference to FIG. 4. As shown in FIG. 4, the processing flow after receiving the handover response also includes two cases: Case 1: The agent returns the handover result (401) before the timer D (ie, the fourth timer described above) times out, and then the processing step As follows: Step 1: Turn off timer D (402); Step 2: Determine whether the working mode that the agent wants to switch to is the primary agent (403); If you want to switch to the secondary agent, determine whether the returned switching result is successful ( 404), if it is unsuccessful, continue to send a handover request to the agent to switch from the agent (405), and set a timer D to wait for the handover result (420); if the handover is successful, update the agent in the in-memory agent information table The working mode (406), and synchronously update the information of the agent in the database (407); to switch to the main agent, determine whether the returned switching result is successful (408), if successful, update the in-memory agent information table The working mode of the agent (409), and synchronously updating the information (410) of the agent in the database, and saving the agent identification information (for example, number) to the in-memory node and the main agent comparison table (4) 11), in order to ensure that there is only one main agent under the same node, you must also send all the other agents under the node to switch from the agent.

7 P16849 If the method is unsuccessful, the working mode of the agent in the memory is set to the slave working mode (415), and the information of the agent in the database is updated synchronously (417), and the node in the memory and the master agent are compared. The primary agent corresponding to the node is cleared (416). The node is then searched for a new available primary agent (418), a handover request is sent to the primary agent (419), and a timer D is set to wait for the handover result (420). Case 2: After the timer D has not returned the handover result (413), the agent has the following steps: Determine whether the working mode that the agent is to switch to is the primary agent (414); to switch to the secondary agent, continue to The Agent sends a handover request to the slave agent (421), and sets a timer D to wait for the handover result (420). To switch to the primary agent, the working mode of the agent in the memory is set to the slave working mode (415), and the information of the agent in the database is updated synchronously (417), and the node in the memory and the master agent are compared in the table. The primary agent corresponding to the node is cleared ( 416 ). The node is then sought for a new available primary agent (418), a handover request is sent to the primary agent (419), and a timer D is set to wait for the handover result (420). At this point, the process is all over. In this way, through the invention, it is ensured that under the arbitration and identification confirmation of the Manager, the situation of the main agent and the multi-master agent does not occur, so that the whole system can stably provide various media services for the user. The above is only the preferred embodiment of the present invention, and is not intended to limit the present invention, and various modifications and changes can be made to the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention are intended to be included within the scope of the present invention.

8 P 16849

Claims

Claim

A switching processing method for a streaming media node controller, configured for switching of a streaming media node controller in a content distribution network, wherein a streaming media node of the content distribution network has multiple streaming media node controllers, and Only one streaming media node controller acts as a mainstream media node controller at a time, and is characterized in that it includes:

Setting a information table and/or a comparison table, wherein the information table records information of each streaming media node controller of the content distribution network, where the information includes a node to which the streaming media node controller belongs, an operating mode, and a link state; The table records identification information of each node of the content distribution network and/or the mainstream media node controller of the node;

The information table and/or the comparison table is traversed, and a node for which the node of the mainstream media node controller whose link state is abnormal and/or a node whose identification information of the mainstream media node controller is empty is traversed, and a master handover request is initiated.

2. The method of claim 1, wherein the process of initiating a master handover request by traversing the information table is specifically:

Setting a first timer, triggering to traverse each streaming media node controller in the information table, and determining, by the streaming node controller determined to be a link abnormality, whether it is a mainstream media node controller, if the determination result is yes Then, the node to which the mainstream media node controller belongs is determined, and a streaming node controller whose link state is normal is sent to send a primary handover request to request that it be switched to the mainstream media node controller.

3. The method according to claim 2, wherein traversing is performed by sending a heartbeat request to each of the streaming node controllers in the information table, and setting a third timer, wherein, according to the streaming node Whether the controller returns a heartbeat response before the third timer expires to determine whether the link is normal.

The method according to claim 3, wherein the additional information of the third timer carries basic information of an associated streaming media node controller to identify a streaming media node controller.

The method according to claim 3, wherein the streaming node controller that determines that it is a link abnormality is determined by the streaming node controller that does not return a heartbeat response after the third timer expires for a predetermined predetermined number of times Device.

9 P 16849

The method according to claim 5, wherein, for the streaming node controller determined to be abnormal in the link, the link state corresponding to the information recorded in the information table is set to be abnormal.

The method according to claim 6, wherein the mainstream media node controller determined to be a link abnormality clears the corresponding mainstream media node controller identification information of the node to which the node belongs.

8. The method according to claim 1, wherein the process of initiating a master handover request by traversing the lookup table is specifically:

Setting a second timer, triggering to traverse each node in the comparison table, and selecting a node of the node that has a link state of normal, the node of the node is a normal node of the streaming media node controller Switch the request to request it to switch to the mainstream media node controller.

The method according to any one of claims 2 to 8, characterized in that, when the main handover request is transmitted, a fourth timer is set.

10. The method according to claim 9, wherein the additional information of the fourth timer carries basic information of an associated streaming media node controller to identify a streaming media node controller.

The method according to claim 9, further comprising:

If the streaming media node controller that is requested to switch returns a handover response before the fourth timer expires, the fourth timer is closed, and it is determined whether the handover is successful.

In the case that the handover is successful, the streaming media node controller is used as a new mainstream media node controller, and the corresponding working mode in the information table is set as the main working mode, and the identifier is saved to the Chart;

In the case that the handover fails, the corresponding working mode of the streaming media node controller in the information table is set to the slave working mode, and the node to which the node belongs is set in the corresponding mainstream media node controller identifier in the comparison table. It is empty, and reselects the streaming node controller whose link state of the node to which it belongs is normal to send the primary switching request.

10 P 16849

The method according to claim 11, wherein, in the case that the handover is successful, the handover request is further sent to all other streaming media control points under the node to which the new mainstream media node controller belongs, To request it to switch to the slave streaming node controller.

13. The method according to claim 12, wherein, for the case that no handover response or handover failure is returned, the slave handover request is continued to be sent; for the handover success, the other streaming media control is switched to The corresponding working mode in the information table is set to the slave working mode.

The method according to claim 9, further comprising:

If the streaming media node controller requested to switch does not return a handover response before the fourth timer expires, setting the corresponding working mode of the streaming media node controller in the information table to the slave working mode And setting the corresponding mainstream media node controller identifier of the node to which the node belongs to be empty, and reselecting the streaming node controller whose link state of the node to which the node belongs is normal to send the primary handover request.

11 P16849