WO2012142842A1

WO2012142842A1 - Content delivery and content acquisition method, content delivery apparatus and content transmission system

Info

Publication number: WO2012142842A1
Application number: PCT/CN2011/084004
Authority: WO
Inventors: 刘宇
Original assignee: 中兴通讯股份有限公司
Priority date: 2011-04-21
Filing date: 2011-12-14
Publication date: 2012-10-26
Also published as: CN102291620A

Abstract

Provided are a content delivery and content acquisition method, a content delivery apparatus and a content transmission system, applied to an IPTV system, comprising: a management server determining a management node according to the serial number of content waiting to be delivered, the management node is an edge node in a storage service domain consisting of a group of edge nodes; and the management server sending the content waiting to be delivered to the management node for storage. Using the method and system provided in the present invention can conserve edge node storage space, and reduce the pressure on the management server.

Description

Content distribution, content acquisition method, content distribution device, and content distribution system

The present invention relates to the field of data communications, and in particular, to a content distribution, content acquisition method, content distribution apparatus, and content distribution system applied to an IPTV (Internet Protocol Television) system. Background technique

With the rapid development of network and multimedia technologies, watching video programs online through the Internet has become a reality. IPTV can provide services such as live TV, VOD (Video On Demand) and TVOD (TV On Demand). Due to its convenience, intelligence and controllability, IPTV is being used by more and more users. accept.

CDN (Content Delivery Network) is an important part of the IPTV system. It mainly has functions such as content storage, content distribution and delivery, content scheduling control, live channel topology management, user stream service control and billing.

Existing CDNs typically employ a three-level node architecture, as shown in Figure 1, including a central node, a regional central node, and an edge node. The central node stores all the content; the regional central node stores all the content or part of the content as required, and when the edge node is unable to provide services for the user, the regional central node provides services for the user; the edge node stores part of the content as required, which is the set top box The user whose IP address belongs to the edge node (which may be referred to as the home node) provides services.

At present, the content acquisition method in the IPTV system is: when the edge node to which the IP address of the set top box belongs has the on-demand content, the edge node directly provides the on-demand content to the set-top box; when the edge node of the set-top box belongs to the edge node, the on-demand content does not exist. When the set top box issues a location request to the management server, the request management server provides an address of the node storing the multicast content; the management server receiving the location request determines the address of the node storing the multicast content and provides the address For the set top box, the set top box accesses the corresponding node according to the address of the received node, and the node provides the on-demand content to the set top box.

However, the current content acquisition method has the following disadvantages: 1. When more on-demand requests cannot be responded by the home node, each set-top box issues a location request to the management server, whereby the management server needs to process a large number of location requests, and the management server will It bears a lot of pressure, so that it can't respond to the positioning request in time, which affects the user's normal viewing of the program. 2. There is a certain amount of repetition in the content stored by each home node, but the stored content is not necessarily on-demand by the set-top box, which wastes storage capacity. Moreover, limited to the storage capacity of the edge node, even if some edge nodes are in the service idle period, since there is no content stored by the user on-demand, the corresponding service cannot be provided, and the service capability is reduced. Summary of the invention

In view of this, the present invention provides a method for content distribution and content acquisition and a content distribution system for saving storage space of edge nodes and reducing pressure on the management server.

The present invention provides a content publishing method, which is applied to an IPTV system, including: a management server determines a management node according to a serial number of a content to be released, and the management node is a storage service domain composed of a group of edge nodes. An edge node; the management server sends the to-be-published content to the management node for storage.

The present invention also provides a content acquisition method, including: the set top box sends an on-demand request to its home node; and the home node determines, according to the on-demand request, that the on-demand content is not stored locally, the home node determines according to the serial number of the on-demand content. a management node of the on-demand content; the home node notifying the set top box of the determined address of the management node; the set top box requests the on-demand content from the management node according to the received address; the management node determines according to the on-demand request When the on-demand content is stored therein, the on-demand content is provided by the management node to the set top.

The present invention also provides a content distribution apparatus, including a management server, configured to determine a corresponding management node according to the serial number of the content to be released;

a home node, configured to receive an on-demand request of the set-top box, and determine that the content of the request-on-demand is not stored therein, determine the management node according to the serial number of the on-demand content, and notify the set-top box of the determined address of the management node;

The management node is configured to store the publishing content of the management server, and the on-demand content requested by the set-top box, and when the content of the on-demand content is stored therein, provide the content to the set-top box;

The home node and the management node belong to the same group of edge nodes and form a storage service domain.

The invention also provides a content distribution system, including

a set top box, configured to send an on-demand request to its home node, and request the on-demand content from the management node according to the received address of the management node;

a management server, configured to determine a corresponding management node according to the serial number of the content to be released;

The technical solution of the present invention can realize that each content is only published in the storage service domain, thereby effectively saving the storage space of the edge node. And, when the on-demand request for the set-top box cannot be returned When the node responds, the address of the management node provided by the home node to the on-demand content of the set-top box, the set-top box sends an on-demand request to the management node, instead of issuing a positioning request to the management server by the set-top box, thereby greatly reducing the need for the management server to process. The location request reduces the pressure on the management server and speeds up the response to the user's request for on-demand. DRAWINGS

FIG. 1 is a schematic diagram of a three-level node architecture of an existing CDN;

2 is a schematic structural diagram of a content distribution system according to the present invention;

3 is a schematic diagram of deployment of edge nodes and virtual nodes in the present invention;

4 is a schematic flowchart of a method for acquiring content in the present invention;

Figure 5 is a schematic diagram of adding or deleting nodes in the present invention. detailed description

A group of edge nodes form a storage service domain, and edge nodes in the storage service domain form a ring structure, and each edge node corresponds to a matching number. Each edge node in the storage service domain may also be divided into a plurality of virtual nodes corresponding thereto, and the edge node and the virtual node form a ring structure, as shown in FIG. 3, each edge node and the virtual node correspond to each other. A matching number. Each virtual node occupies a point on the ring (indicated by 32-bit numbers), and is managed by a Hash algorithm (hash algorithm) to distribute the virtual nodes evenly over the ring structure. The above virtual nodes do not really exist. They only expand the number of nodes, which facilitates hashing, makes the distribution of hash results more balanced, and minimizes the pressure of copying content between nodes caused by physical node additions and deletions.

A set top box belonging to an edge node within the storage service domain can access all edge nodes within the domain. Each edge node is configured with a content table for storing content information and content management information. Each edge node in the storage service domain periodically notifies its other storage and service capabilities to other edge nodes in the domain; and each edge node in the storage service domain is connected The storage and service capability information notified by other edge nodes in the domain is stored in the table of contents. The communication between nodes in the storage service domain uses RTSP (Real Streaming Protocol).

2 is a schematic structural diagram of a content distribution system according to the present invention. As shown in FIG. 2, the content distribution system includes: a management server, a set top box, a home node, a management node, and an edge node. The management server is configured to determine a corresponding management node according to the serial number of the content to be released. Specifically, the management server hashes the sequence number of the content to be distributed by using a hash algorithm to obtain a hash code, determines a matching number matching the hash code, and uses an edge node corresponding to the matching number. Or an edge node corresponding to the virtual node corresponding to the matching number is determined as a management node.

The set top box is for issuing an on-demand request to its home node, and requesting the on-demand content from the management node based on the received management node's address.

The home node is configured to receive the on-demand request of the set-top box, and determine that the requested on-demand content is not stored, determine the management node according to the serial number of the on-demand content, and notify the set-top box of the determined address of the management node. Specifically, the home node hashing the sequence number of the on-demand content by using a hash algorithm to obtain a hash code, determining a matching number matching the hash code, and using an edge node corresponding to the matching number Or an edge node corresponding to the virtual node corresponding to the matching number is determined as a management node.

The management node is configured to store the publishing content of the management server, and the on-demand content requested by the set-top box, and provide the on-demand content to the set-top box when determining that the on-demand content is stored therein; wherein, the belonging A node and a management node belong to the same set of edge nodes and form a storage service domain. The content of the management node does not participate in the timing aging, and the content can only be deleted by the management server, so as to slow down the aging speed of the storage content in the node in the storage service domain, and reduce the number of times the management node cannot respond to the set-top box on-demand request. Thereby reducing the number of times the set top box issues a positioning request to the management server, thereby greatly reducing the management service The location request that needs to be processed reduces the pressure on the management server and speeds up the response to the user's request for on-demand.

When the management node counts that the number of requests for ordering a certain content reaches a certain set value, sending a command to generate a copy to other edge nodes in the storage service domain, and the edge node receiving the command copies the The content generates a copy, and the copy participates in timing aging. The participation of the secondary aging timer can reduce the number of duplicate storage contents in the storage service domain, effectively saving the storage space of the edge nodes.

When the content on the management node is deleted, and the management node counts that the number of requests for ordering a certain content reaches a certain set value, the management node sends a request for re-publishing the content to the management server. .

The process of content acquisition includes the following steps:

Step A: The set top box sends an on-demand request for the on-demand content to its home node.

Step B: When the home node determines, according to the on-demand request, that the on-demand content is not stored locally, the home node determines the management node of the on-demand content according to the sequence number of the on-demand content. Specifically, the home node hashing the sequence number of the on-demand content by using a hash algorithm to obtain a hash code; the home node determining a matching number matching the hash code, and the matching number The corresponding edge node or the edge node corresponding to the virtual node corresponding to the matching number is determined as the management node.

Step C: The home node notifies the set top box of the address of the management node determined by the home node. Step D: The set top box requests on-demand content from the management node according to the received address. Step E: When the management node determines, according to the on-demand request, that the on-demand content is stored, the management node provides the on-demand content to the set-top box, or the management node stores the content copy in the storage service domain. One of the nodes with the lightest load is selected, and the on-demand content is provided by the edge node to the set top box.

Specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings. The invention is illustrated below in a specific embodiment. Referring to FIG. 3, in this embodiment, four edge nodes node0, node1, node2, and node3 are combined into one storage service domain. Each edge node is divided into one virtual node (which can be divided into multiple according to needs, for example, 100), the virtual node V_nodeO corresponds to the edge node nodeO, the virtual node v_nodel corresponds to the edge node node 1, and the virtual node v-node2 Corresponding to the edge node node2, the virtual node v_node3 corresponds to the edge node node3, as shown in Figure 3. The nodes node0, node1, node2, node3, v_node0, v_nodel, v-node2, v-node3 are separated by another ¹ J corresponding to numbers 0, 1, 2, 3, 4, 5, 6, and 7.

For example, when the management server issues a content with a sequence number (ID) of "cid000015", hashing "000015" according to the hash algorithm yields a uniquely determined hash code (usually a 32 number), assuming the resulting hash The column code is 7, the matching number 7 matching the hash code is queried, the node corresponding to the matching number 7 is V_node3, and the edge node corresponding to v_node3 is node3, thereby determining that the edge node is the management node of the content "cid000015" . The management server posts the content on the node node3, and the node node3 becomes the management node of the content "cid000015".

Assume that the IP address of a user's set-top box belongs to the edge node node 1. When the user orders the content "cid000015", the set-top box requests the on-demand content from the edge node nodel, and the edge node nodel judges that there is no content according to the serial number "cid000015" of the on-demand content. The on-demand content is stored, and the edge node nodel hashes "000015" according to the hash algorithm to obtain a hash code 7, and the matching number 7 matching the hash code is queried, and the node corresponding to the matching number 7 is determined to be v_node3, thereby determining v—The edge node node3 corresponding to node3 is the management node of the content “cid000015”. Thus, the edge node node1 notifies the set-top box of the address of the edge node node3. The set top box requests the content of the on-demand content "cid000015" from the management node node3 according to the received address.

The management node node3 determines, according to the serial number "cid000015" of the on-demand content, whether the on-demand content is stored therein, and if the on-demand content is stored, the on-demand content is provided Give the set-top box.

When the number of requests for the on-demand content "cid000015" reaches a certain number on the management node node3 within a certain period of time (for example, 1000 times per hour), the management node node3 uses RTSP to store other edge nodes (such as edge nodes) in the service domain. Node2) sends a command to generate a copy, node node2 downloads a copy of the content from node node3 via FTP (File Transfer Protocol) to generate a copy. The copied content is marked in the content table of the node node2; at the same time, the node node2 in the content table of the node node3 issues a copy of the content. The content with the replica mark participates in the timing aging on the node node2. When the content copy on the node node2 is aged out, the node node2 notifies the management node node3, and the record node node2 in the content table of the node node3 deletes a copy of the content. When the content on the management node node3 is deleted by the management server, if the content copy on the node node2 is not aged out, the management node node3 notifies the node node2 to delete the content copy through the RTSP.

When another user requests the content of the order number "cid000015", if the other node stores the content copy of "cid000015", the management node node3 can assign the user to other nodes according to the load condition of each node, so as to achieve equalization. The purpose of the node service is to prevent an edge node from being overburdened, while other nodes are idle.

Similarly, when the number of requests for the on-demand content "cid000015" reaches a certain number on the management node node3 within a certain period of time (for example, 10,000 times per hour), the management node node3 copies another serial number "cid000015". A copy of the content is stored to other edge nodes within the storage service domain (eg, edge node node0). By analogy, when the number of requests for the on-demand content "cid000015" reaches a certain number (for example, 100,000 on-demand per hour) on the management node node3 within a certain period of time, all edge nodes in the storage service domain store a copy of the content.

When the on-demand content is deleted by the management server due to aging or other reasons, and the management node node3 has not stored the bit-cast content "cid000015", the management node node3 issues a location request to the management server, requesting the management server to provide the storage service extra-domain storage of the broadcast content. The address of the node of "cid000015". When the number of requests for the on-demand content "cid000015" reaches a certain number on the management node node3 within a certain period of time (for example, 100 times per hour), the management node node3 notifies the management server to reissue the content with the serial number "cid000015" to the management. On the node node3, the number of times the management node node3 issues a positioning request to the management server is reduced, the burden on the management server is alleviated, and the request for responding to the user's on-demand is accelerated.

The content acquisition method of the present invention will be described in detail below with reference to FIG. As shown in FIG. 4, node A, node M, and node S form a storage service domain, where node A is a home node of a set top box, node M is a management node of on-demand content, and node S is a node that stores a copy of the content of the broadcast. The node with the lightest load. Node N is a node that stores a copy of the content of the broadcast content outside the storage service domain. The method includes the following steps:

Step 301: The set top box sends a VOD request to the node A through the RTSP.

Step 302: The node A searches for the content stored in the node according to the serial number of the on-demand content, and determines whether the on-demand content is stored in the node.

Step 303: If the on-demand content corresponding to the request is stored in the node A, the node A directly provides the on-demand content to the set-top box, completes the on-demand request, and ends the current process.

Step 304: If the on-demand content corresponding to the request is not stored in the node A, the node A determines the management node of the on-demand content as the node M according to the sequence number of the on-demand content.

Step 305, the node A notifies the set top box of the address of the node M determined by the node A.

Step 306: The set top box requests the on-demand content from the node M according to the received address. Step 307: The node M searches for the content stored in the node according to the serial number of the on-demand content, and determines whether the on-demand content is stored therein. If yes, the process proceeds to step 308; if not, the process proceeds to step 311. .

Step 308, if the on-demand content corresponding to the request is stored in the node M, the node M selects a node S with the lightest load among the nodes storing the content copy according to the number of requests to be responded by the nodes in the storage service domain. And provide the address of node S to the set top box. Step 309, the set top box sends a VOD request to the node S according to the received address of the node S. Step 310: The node S provides the on-demand content to the set-top box, completes the on-demand request, and ends the current process.

Step 311: If the content corresponding to the on-demand request is not stored in the node M, the node M notifies the set-top box node M that the on-demand content corresponding to the request is not stored.

Step 312: The set top box sends a location request to the management server, and the request management server provides an address of a node that stores the content of the broadcast content outside the storage service domain.

Step 313: The management server searches for the content stored in each node according to the serial number of the on-demand content, and determines that the node N outside the storage service domain stores the content of the broadcast.

Step 314, the management server notifies the set top box of the address of the node N.

Step 315: The set top box requests the on-demand content from the node N according to the received address. Step 316, the node N provides the on-demand content to the set-top box, and completes the on-demand request.

Nodes in the storage service domain can be added or deleted according to the amount of content to be published, which satisfies the needs of data storage. FIG. 5 is a schematic diagram of adding or deleting a node in the present invention. Referring to FIG. 5, when a node N needs to be added between the node A and the node B, according to the hash algorithm principle, the original storage between the node A and the node N is stored. The content on the Node B is copied to the node N, and the information of the copy content managed by the Node B is also copied to the node N.

When the node N needs to be deleted, according to the principle of the hash algorithm, the node B copies the content stored by the node N to the node B through FTP, and the information of the copy content managed by the node N is copied to the node B through the RTSP between the nodes, and the node B after the copying is completed. The copy is written to its content table and the management server is notified of the result of the content copy.

The technical solution of the present invention can be implemented: only one content is distributed in the storage service domain, which effectively saves the storage space of the edge node. When the on-demand request of the set-top box cannot be responded by the home node, the home node provides the address of the set-top box management node, and the set-top box sends an on-demand request to the management node of the on-demand content in the domain, instead of being sent by the set-top box to the management server. The positioning request is greatly reduced, thereby greatly reducing the positioning request that the management server needs to process, reducing the pressure on the management server, and improving the speed of responding to the user's request for on-demand.

In addition, the present invention can also perform hash calculation using other hash algorithms. Communication between nodes in the storage service domain can also use other communication protocols.

When publishing content, you can also divide content into popular content and unpopular content according to certain criteria. Popular content is published on all nodes of the storage service domain; unpopular content is published according to the content publishing method described above. When a lot of content to be released has a high on-demand rate, the above technical solution can avoid the number of times the nodes in the storage service domain copy the content copy, thereby reducing the burden on the node and avoiding waste of service resources.

The creation and playback of TVOD and channels can also be handled using the present invention.

Those skilled in the art will appreciate that embodiments of the present invention can be provided as a method, system, or computer program product. Accordingly, the present invention may take the form of an entirely hardware embodiment, an entirely software embodiment, or a combination of software and hardware. Moreover, the invention can be embodied in the form of a computer program product embodied on one or more computer-usable storage media (including but not limited to disk storage and optical storage, etc.) including computer usable program code.

The present invention has been described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (system), and computer program products according to embodiments of the invention. It will be understood that each flow and/or block of the flowchart and/or block diagrams, and combinations of flows and/or blocks in the flowcharts and/or block diagrams can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing device to produce a machine for the execution of instructions for execution by a processor of a computer or other programmable data processing device. Means for implementing the functions specified in one or more of the flow or in a block or blocks of the flow chart.

The computer program instructions can also be stored in a computer readable memory that can direct a computer or other programmable data processing device to operate in a particular manner, such that stored in the computer readable memory The instructions in the reservoir produce an article of manufacture comprising an instruction device that implements the functions specified in one or more blocks of the flow or in a flow or block diagram of the flowchart.

These computer program instructions can also be loaded onto a computer or other programmable data processing device such that a series of operational steps are performed on a computer or other programmable device to produce computer-implemented processing for execution on a computer or other programmable device. The instructions provide steps for implementing the functions specified in one or more of the flow or in a block or blocks of a flow diagram. The spirit and scope of the invention. Thus, it is intended that the present invention cover the modifications and modifications of the invention

Claims

Claim

A content distribution method, applied to an IPTV system, comprising: a management server determining a management node according to a serial number of a content to be released, wherein the management node is a storage service domain composed of a group of edge nodes An edge node;

The management server sends the to-be-published content to the management node for storage.

The content distribution method according to claim 1, wherein the management server determines a management node according to the serial number of the content to be distributed, and specifically includes:

进行 Hashing the sequence number of the published content with a hash algorithm to obtain a hash code; determining the management node of the posted content according to the hash code.

The content distribution method according to claim 2, wherein the determining, by the hash code, the management node that issues the content, specifically includes:

The edge nodes in the storage service domain form a ring structure, and each edge node corresponds to a matching number;

The management server determines a matching number that matches the hash code, and determines an edge node corresponding to the matching number as a management node.

Each edge node in the storage service domain is divided into a plurality of virtual nodes corresponding thereto, and the edge node and the virtual node form a ring structure, and each edge node and the virtual node correspond to a matching number;

The management server determines a matching number that matches the hash code, and determines an edge node corresponding to the matching number or an edge node corresponding to the virtual node corresponding to the matching number as a management node.

The content distribution method according to claim 1, wherein the content on the management node does not participate in timing aging.

The content distribution method according to claim 1, further comprising: when the management node counts that the number of requests for the on-demand content reaches the first set value, sending and generating to the other edge nodes in the storage service domain. The command of the copy, the edge node receiving the command copies the content generation copy from the management node.

7. The content distribution method according to claim 6, wherein the copy participates in aging.

8. The content distribution method according to claim 1, further comprising: each edge node in the storage service domain periodically notifying other edge nodes in the domain of its own storage and service capabilities;

Each edge node in the storage service domain accepts storage and service capability information notified by other edge nodes in the domain and saves it.

The content distribution method according to claim 1, wherein when the content on the management node is deleted, and the management node counts that the number of requests for on-demand content reaches a second set value, The management node issues a request to the management server to republish the content.

10. A content acquisition method, comprising:

The set top box sends an on-demand request to its home node;

When the home node determines that the on-demand content is not stored locally according to the on-demand request, the home node determines the management node of the on-demand content according to the sequence number of the on-demand content;

The home node notifies the set top box of the address of the management node determined by the home node;

The set top box requests the management node to request the content according to the received address; when the management node determines, according to the on-demand request, that the on-demand content is stored, the management node provides the on-demand content to the set top box.

The content acquisition method according to claim 10, wherein the determining, by the home node, the management node of the on-demand content according to the serial number of the on-demand content, specifically includes:

The home node uses a hash algorithm to hash the serial number of the on-demand content to obtain a hash Column coding

A management node that determines the on-demand content based on the hash code.

The content acquisition method according to claim 11, wherein the determining the management node of the on-demand content according to the hash coding comprises:

The home node determines a matching number that matches the hash code, and determines an edge node corresponding to the matching number or an edge node corresponding to the virtual node corresponding to the matching number as a management node.

The content acquisition method according to claim 10, further comprising: the management node selecting a lightest edge node among the nodes storing the content copy in the storage service domain, The edge node provides the on-demand content to the set top box.

14. A content distribution apparatus, comprising:

15. The content distribution apparatus according to claim 14, wherein:

The management server is specifically configured to perform a hash calculation on a sequence number of the content to be distributed by using a hash algorithm to obtain a hash code, and determine a management node that issues the content according to the hash code; and/or, the home node, Specifically used to hash the serial number of the on-demand content by using a hash algorithm A management node that computes a hash code and determines the on-demand content based on the hash code.

The content distribution apparatus according to claim 15, wherein the edge nodes in the storage service domain form a ring structure, each edge node corresponds to a matching number; or each edge in the storage service domain The node is divided into a plurality of virtual nodes corresponding thereto, and the edge node and the virtual node form a ring structure, and each edge node and the virtual node correspond to a matching number;

The management server is specifically configured to determine a matching number that matches the hash code, and determine an edge node corresponding to the matching number or an edge node corresponding to the virtual node corresponding to the matching number as a management node; and / or,

The home node is specifically configured to determine a matching number that matches the hash code, and determine an edge node corresponding to the matching number or an edge node corresponding to the virtual node corresponding to the matching number as a management node.

17. A content dissemination system, characterized in that