CN114513519B - Video peer-to-peer stream exchange method, system and device - Google Patents
Video peer-to-peer stream exchange method, system and device Download PDFInfo
- Publication number
- CN114513519B CN114513519B CN202210407106.8A CN202210407106A CN114513519B CN 114513519 B CN114513519 B CN 114513519B CN 202210407106 A CN202210407106 A CN 202210407106A CN 114513519 B CN114513519 B CN 114513519B
- Authority
- CN
- China
- Prior art keywords
- peer
- node
- video
- user
- server
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Images
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L67/00—Network arrangements or protocols for supporting network services or applications
- H04L67/01—Protocols
- H04L67/10—Protocols in which an application is distributed across nodes in the network
- H04L67/104—Peer-to-peer [P2P] networks
- H04L67/1074—Peer-to-peer [P2P] networks for supporting data block transmission mechanisms
- H04L67/1078—Resource delivery mechanisms
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L67/00—Network arrangements or protocols for supporting network services or applications
- H04L67/01—Protocols
- H04L67/10—Protocols in which an application is distributed across nodes in the network
- H04L67/104—Peer-to-peer [P2P] networks
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L67/00—Network arrangements or protocols for supporting network services or applications
- H04L67/01—Protocols
- H04L67/10—Protocols in which an application is distributed across nodes in the network
- H04L67/1097—Protocols in which an application is distributed across nodes in the network for distributed storage of data in networks, e.g. transport arrangements for network file system [NFS], storage area networks [SAN] or network attached storage [NAS]
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N21/00—Selective content distribution, e.g. interactive television or video on demand [VOD]
- H04N21/60—Network structure or processes for video distribution between server and client or between remote clients; Control signalling between clients, server and network components; Transmission of management data between server and client, e.g. sending from server to client commands for recording incoming content stream; Communication details between server and client
- H04N21/63—Control signaling related to video distribution between client, server and network components; Network processes for video distribution between server and clients or between remote clients, e.g. transmitting basic layer and enhancement layers over different transmission paths, setting up a peer-to-peer communication via Internet between remote STB's; Communication protocols; Addressing
- H04N21/643—Communication protocols
- H04N21/6437—Real-time Transport Protocol [RTP]
Abstract
The invention provides a method, a system and a device for exchanging video peer-to-peer streams, which are characterized in that each server node adopts a peer-to-peer network structure, does not distinguish the hierarchy, and gathers the local storage of each server node into a virtual shared storage to realize decentralization. And the full links between the server nodes and the user nodes are set to communicate by adopting a real-time audio protocol, so that the time delay and the calculation cost of a distribution network are reduced. By building a subscription pushing mechanism, when a user requests video data, a preset user perception guarantee algorithm is adopted to calculate a correlation score between a current user node and a user node which has obtained appointed video data, the user node with higher correlation is selected as a correlated user node to subscribe, and the correlated user node is directly enabled to push the video data to the current user node, so that network delay is greatly reduced, and user experience quality is guaranteed.
Description
Technical Field
The present invention relates to the field of network video transmission technologies, and in particular, to a method, a system, and an apparatus for exchanging video peer-to-peer streams.
Background
In recent years, the traffic scale of network video transmission is increasing unprecedentedly, videos are presented as visual and vivid contents, gradually occupy the mainstream of domestic network contents, and the traffic scale of Chinese video transmission is pushed to the top of the world. Facing the tremendous growth of video content data, network video transmission technology has experienced more serious challenges, especially for real-time video transmission systems that require low latency.
On the other hand, a network transmission system is developed rapidly in recent years, and the transmission architecture is gradually upgraded from a traditional tree-shaped centralized distribution network to a cloud network interconnection architecture deployed in a full cloud manner and then to a multi-platform binding architecture taking edge cloud as a core. However, the existing network video transmission architecture and commercial system still adopt the original tree-centered distribution mechanism. The architecture has a plurality of problems, which are mainly reflected in the aspects of protocol compatibility, system heterogeneity, software and hardware coupling and the like. Network video streaming technology based on a central distribution architecture of a CDN (Content Delivery Network) is difficult to meet the increasingly large Content scale. A multi-stage long path source returning mechanism exists, and larger network delay can be generated; the tree structure is limited to one-way forwarding and is difficult to support two-way video streams; the multi-stage transmission protocol is heterogeneous, has poor universality and is difficult to adapt to the rapid change of service requirements; the control platform is centralized in the center, and fine-grained control cannot be realized. Therefore, a new video data transmission method and system architecture are needed.
Disclosure of Invention
In view of this, the present invention provides a method, a system and a device for exchanging peer-to-peer video streams to eliminate or improve one or more defects in the prior art, and solve the problems of high transmission delay and high overhead of huge video data.
The technical scheme of the invention is as follows:
in one aspect, the present invention provides a method for exchanging video peer-to-peer streams, where the method operates in a peer-to-peer video stream exchange network formed by multiple server nodes, and the server nodes are connected by using a peer-to-peer architecture, and the method includes:
the local storage of each server node is gathered into a virtual shared storage, and video data sent to any server node by a video publishing user is received and stored;
receiving an acquisition request of a primary user node for appointed video data, and searching a plurality of secondary user nodes which have acquired the appointed video data;
calculating a correlation score between the primary user node and each secondary user node by adopting a preset user perception guarantee algorithm, and selecting one secondary user node as a correlation user node according to the correlation score, wherein the preset user perception guarantee algorithm is at least based on a content correlation index, a region proximity index and a plurality of network performance metrics to obtain the correlation score;
initiating a subscription to the associated user node to enable the associated user node to directly transmit the specified video data to the primary user node;
and performing resource loss constraint optimization by using a decision scheduler according to access resources, transmission resources and computing resources of each server node in the peer-to-peer video stream switching network;
and the video data is transmitted among all the server nodes, the primary user node and all the secondary user nodes by adopting a real-time audio protocol.
In some embodiments, aggregating the local storage of the server nodes into a virtual shared storage comprises: and each server node stores the video data by adopting a VSEN architecture.
In some embodiments, a preset user perception assurance algorithm is used to calculate a correlation score between the primary user node and each secondary user node, and the calculation formula is:
wherein S represents a relevance score, V is a content relevance index, the value of V is 0 when relevant, and the value of V is 1 when irrelevant;
is a regional proximity index;
representing a vector matrix composed of a plurality of network performance metrics;
a coefficient which is a regional proximity index;
is composed of
The coefficient matrix of (2).
In some embodiments, the network performance metrics include at least: round trip delay, throughput, and packet loss rate.
In some embodiments, performing resource loss constraint optimization based on access resources, transmission resources, and computational resources of server nodes in the peer-to-peer video stream switching network using a decision scheduler comprises: and maximizing the use amount of global access resources, transmission resources and computing resources of all server nodes in the peer-to-peer video stream switching network, and performing matching scheduling on the access resources, the transmission resources and the computing resources of the video data in each server node.
In some embodiments, access resource, transmission resource, and computational resource usage are maximized globally for all server nodes in the peer-to-peer video stream switching network, where the expression is:
wherein R represents a set of access resources, computational resources and transmission resources, and N is the total number of server nodes;
the theoretical maximum benefit of the nth server node is the dimensionless addition result of the theoretical access resource, the calculation resource and the transmission resource;
representing the resource loss of the nth server node, wherein the resource loss is a dimensionless sum result of the access resource loss, the computing resource loss and the transmission resource loss;
representing the actual resource usage of the nth server node.
In another aspect, the present invention further provides a video peer-to-peer stream switching system, which at least includes: a plurality of server nodes connected by adopting a peer-to-peer architecture, wherein each server node is used for accessing a user node; the local storage of each server node is gathered into a virtual shared storage, and the video data sent to any server node by a video publishing user is received and stored; transmitting the video data between each server node and each user node by adopting a real-time audio protocol;
each server node is also used for receiving an acquisition request of a primary user node for the designated video data and searching a plurality of secondary user nodes which have acquired the designated video data; calculating a correlation score between the primary user node and each secondary user node by adopting a preset user perception guarantee algorithm, and selecting one secondary user node as a correlation user node according to the correlation score, wherein the preset user perception guarantee algorithm is at least based on a content correlation index, a region proximity index and a plurality of network performance metrics to obtain the correlation score; and initiating a subscription to the associated user node to enable the associated user node to directly transmit the specified video data to the primary user node.
In some embodiments, each server node calculates the relevance score between the primary user node and each secondary user node by using a preset user perception guarantee algorithm, where the calculation formula is:
wherein S represents a relevance score, V is a content relevance index, the value of V is 0 when relevant, and the value of V is 1 when irrelevant;
is a regional proximity index;
representing a vector matrix composed of a plurality of network performance metrics;
a coefficient which is a regional proximity index;
is composed of
A coefficient matrix of (a);
each server node utilizes a decision scheduler to perform resource loss constraint optimization according to access resources, transmission resources and computing resources of each server node in the peer-to-peer video stream switching network, and the resource loss constraint optimization method comprises the following steps: maximizing the use amount of global access resources, transmission resources and computing resources of all server nodes in the peer-to-peer video stream switching network, and performing matching scheduling on the access resources, the transmission resources and the computing resources of video data in each server node, wherein the expression is as follows:
wherein R represents a set of access resources, computing resources and transmission resources, and N is the total number of server nodes;
the theoretical maximum benefit of the nth server node is the dimensionless addition result of the theoretical access resource, the calculation resource and the transmission resource;
representing the resource loss of the nth server node, wherein the resource loss is a dimensionless sum result of the access resource loss, the calculation resource loss and the transmission resource loss;
representing the actual resource usage of the nth server node.
In another aspect, the present invention also provides an electronic device, which includes a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the steps of the method.
In another aspect, the present invention also provides a computer readable storage medium having stored thereon a computer program which, when executed by a processor, performs the steps of the above method.
The invention has the beneficial effects that:
in the method, the system and the device for exchanging the video peer-to-peer streams, the network structure of the peer-to-peer type is adopted by each server node, the hierarchy is not distinguished, and the local storage of each server node is gathered into a virtual shared storage, so that the decentralization is realized. And the full links between the server nodes and the user nodes are set to communicate by adopting a real-time audio protocol, so that the time delay and the calculation cost of a distribution network are reduced. By building a subscription pushing mechanism, when a user requests video data, a preset user perception guarantee algorithm is adopted to calculate a relevance score between a current user node and a user node which has obtained appointed video data, the user node with higher relevance is selected as a relevant user node to subscribe, and the relevant user node is directly enabled to push the video data to the current user node, so that network delay is greatly reduced, and user experience quality is guaranteed.
Additional advantages, objects, and features of the invention will be set forth in part in the description which follows and in part will become apparent to those having ordinary skill in the art upon examination of the following or may be learned from practice of the invention. The objectives and other advantages of the invention will be realized and attained by the structure particularly pointed out in the written description and claims hereof as well as the appended drawings.
It will be appreciated by those skilled in the art that the objects and advantages that can be achieved with the present invention are not limited to the specific details set forth above, and that these and other objects that can be achieved with the present invention will be more clearly understood from the detailed description that follows.
Drawings
The accompanying drawings, which are included to provide a further understanding of the invention and are incorporated in and constitute a part of this application, illustrate embodiment(s) of the invention and together with the description serve to explain the principles of the invention. In the drawings:
fig. 1 is a flowchart illustrating a video peer-to-peer stream switching method according to an embodiment of the invention;
fig. 2 is a schematic diagram of a network framework of a video peer-to-peer stream switching system according to an embodiment of the present invention;
fig. 3 is a schematic flowchart of resource loss constraint optimization in the video peer-to-peer stream switching method according to an embodiment of the present invention.
Detailed Description
In order to make the objects, technical solutions and advantages of the present invention more apparent, the present invention will be described in further detail with reference to the following embodiments and accompanying drawings. The exemplary embodiments and descriptions of the present invention are provided to explain the present invention, but not to limit the present invention.
It should be noted that, in order to avoid obscuring the present invention with unnecessary details, only the structures and/or processing steps closely related to the solution according to the present invention are shown in the drawings, and other details not so related to the present invention are omitted.
It should be emphasized that the term "comprises/comprising" when used herein, is taken to specify the presence of stated features, elements, steps or components, but does not preclude the presence or addition of one or more other features, elements, steps or components.
It is also noted herein that the term "coupled," if not specifically stated, may refer herein to not only a direct connection, but also an indirect connection in which an intermediate is present.
The existing real-time video streaming technology is mainly based on a central delivery mechanism of a content delivery network CDN. The current main architecture of the mechanism is the traditional tree-shaped centralized distribution network. In a traditional tree-shaped centralized distribution network, video streams are recorded by a producer, are uploaded to a central cloud storage platform through a Real-Time Messaging Protocol (RTMP) after terminal codes are divided into video blocks, and are distributed to edge cloud platforms in various places through RTMP multicast or unicast, and a user directly obtains videos from the edge platforms through other video formats such as FLV (flash video) and the like to form a tree-shaped structure distributed in the center. When a user side requests a network for a related video stream, a control platform located in the center distributes the user to the nearest edge cloud storage platform, if the current edge does not have a needed video block, the video block flows back to the center cloud, is transmitted to the edge through an RTMP protocol, and is decoded and played by the client side. This approach has several major disadvantages: first, there are multiple stages of long path back-to-source mechanisms, which can generate large network delay. The edge nodes need to request the video blocks from the central source, and when the uplink network from the video producer to the central network shakes, the source returning request consumes a large amount of time, so that the client is very easy to jam, and the user experience is influenced. In addition, the access hot spots of the users are dynamically changed, which causes the cache hit rate of the edge nodes to be low, and causes frequent back-source requests. Second, the tree structure is limited to unidirectional forwarding, which makes it difficult to support bidirectional video streams. In the whole transmission architecture, only a video unidirectional transmission path from the center to the edge and then to the client exists, and the common client cannot push the local video stream to the center and backtrack to the terminal of a video producer. Thirdly, the multi-stage transmission protocol is heterogeneous and has poor universality, which is difficult to adapt to the rapid change of service requirements. At present, an RTMP protocol based on TCP is generally used, network dynamic sensing capability is lacked, transmission fine control cannot be realized, and optimization space is limited. Fourthly, the control platform is centralized in the center, and fine-grained control cannot be realized. In a centralized distribution structure, the control platform is only arranged on the central cloud platform, so that the transmission dynamics from the edge to the client cannot be sensed, and the regulation and control are more difficult to achieve.
The invention provides a video peer-to-peer stream exchange method, which operates in a peer-to-peer video stream exchange network formed by a plurality of server nodes, wherein the server nodes are connected by adopting a peer-to-peer architecture, as shown in figure 1, the method comprises the following steps of S101-S105:
step S101: and aggregating the local storage of each server node into a virtual shared storage, and receiving and storing the video data sent to any server node by the video publishing user.
Step S102: and receiving an acquisition request of the primary user node for the designated video data, and searching a plurality of secondary user nodes which have acquired the designated video data.
Step S103: and calculating a correlation score between the primary user node and each secondary user node by adopting a preset user perception guarantee algorithm, selecting one secondary user node as a correlation user node according to the correlation score, and calculating the correlation score by adopting the preset user perception guarantee algorithm at least based on a content correlation index, a region proximity index and a plurality of network performance metrics.
Step S104: and initiating subscription to the associated user node so that the associated user node directly transmits the specified video data to the primary user node.
And, step S105: and utilizing the decision scheduler to perform resource loss constraint optimization according to the access resources, transmission resources and computing resources of each server node in the peer-to-peer video stream switching network.
And video data are transmitted among all the server nodes, the primary user nodes and all the secondary user nodes by adopting a real-time audio protocol.
In step S101, based on a plurality of server nodes connected in a peer-to-peer configuration, without distinguishing a hierarchy, hosts of all the server nodes perform transmission of storage data through a private network. Specifically, in this embodiment, each server node stores the video data by using a VSEN architecture. VSEN constructs software-defined server cluster-based distributed storage. The stored management program is realized by software, only the software can be developed, flexibly and quickly, various requirements of different scenes on storage can be met, data and storage reading access can be distributed to a plurality of nodes in a distributed mode, and the capacity and the performance of the whole storage system are linearly increased along with the increase of the nodes. The underlying storage technology of the VSEN architecture is represented as an abstracted pool of storage space with multiple functions and is exposed to an administrator for deployment. In the embodiment, the VSEN architecture is adopted, so that the control of the video data stream can be directly sunk to each server node and user node of the peer-to-peer network, and the direct interactive management of 'storage-forwarding-control' is carried out without delaying back to the source through the original lengthy multi-stage tree-shaped long path.
In step S102, the primary user node refers to a user node that initiates a request to specify video data. In the peer-to-peer video stream switching network constructed in this embodiment, each server node and/or user node may store not only its unique hot data, but also concurrent collaborative data associated with other nodes in a peer-to-peer manner. The peer-to-peer video stream switching network is managed and operated by a video service provider, when a user puts forward an acquisition request for specified video data to the video service provider, the peer-to-peer video stream switching network does not directly send the video data to a first-level user node, but carries out transmission on the basis of a point-to-point technology, and preferentially selects a second-level user node which has already obtained the specified video data to carry out point-to-point transmission on the first-level user node.
In step S103, by analyzing and evaluating the relevance between the user nodes, the secondary user node with the highest relevance score is selected as the relevant user node, so as to perform point-to-point video data transmission to the primary user node. The preset user perception guarantee algorithm is used for evaluating the incidence relation among all nodes and is mainly evaluated based on a content incidence index, a region proximity index and a plurality of network performance metrics, wherein the content incidence index is used for evaluating the content incidence of video data required to be acquired by a primary user node and video data acquired and stored by a secondary user node. And the region proximity index is used for evaluating the region proximity degree between the primary user node and the secondary user node. The multiple network performance metrics are used for evaluating transmission performance indexes of transmission links between the primary user node and the secondary user node, and may include round-trip delay, throughput, packet loss rate, and the like.
In some embodiments, in step S103, a preset user perception assurance algorithm is used to calculate a correlation score between the primary user node and each secondary user node, where the calculation formula is:
wherein S represents a relevance score, V is a content relevance index, the value of V is 0 when relevant, and the value of V is 1 when irrelevant;
is a regional proximity index;
representing a vector matrix composed of a plurality of network performance metrics;
a coefficient which is a regional proximity index;
is composed of
The coefficient matrix of (2). The network performance metrics include at least: round Trip Time (RTT), throughput, and packet loss rate.
After the relevance score calculated by the calculation formula 1 is higher, the higher the relevance score is, the higher the relevance is, and a secondary user node with the highest relevance to the primary user node can be directly selected for video data point-to-point transmission. In order to improve the fault tolerance, a threshold value can be set, and one of a plurality of secondary user nodes with the correlation scores higher than the threshold value is randomly selected to carry out point-to-point transmission of the video data. Furthermore, global or local optimization can be performed, so that the sum of the relevance scores corresponding to the relevant user nodes selected by the plurality of primary user nodes in the global or local mode is the highest.
In step S104, by constructing a push form, the relevant user node continuously pushes the video data to the first-level user node without the first-level user node continuously requesting the video data, which can greatly reduce the time delay and ensure the QoE of the user experience score.
In step S105, based on the increase of the user scale of the video network, a peer-to-peer video stream switching network needs to deploy a large number of nodes, and in the conventional CDN network, because certain hierarchical management and distributed storage are required among the nodes, while the QoE is approximately ensured, the interaction overhead among the nodes shows a polynomial increase. In this embodiment, in order to ensure linear increase of node overhead, access resources, transmission resources, and computation resources of each server node are scheduled globally, and resource loss constraint optimization is performed.
In some embodiments, in step S105, as shown in fig. 3, performing resource loss constraint optimization by using a decision scheduler according to access resources, transmission resources and computation resources of each server node in the peer-to-peer video stream switching network includes: and maximizing the use amount of global access resources, transmission resources and computing resources of all server nodes in the peer-to-peer video stream switching network, and performing matching scheduling on the access resources, the transmission resources and the computing resources of the video data in each server node.
The method maximizes the use amount of access resources, transmission resources and computing resources of all server nodes in the peer-to-peer video stream switching network, and comprises the following steps:
wherein R represents a set of access resources, computing resources and transmission resources, and N is the total number of server nodes;
the theoretical maximum benefit of the nth server node is the dimensionless sum result of the theoretical access resource, the calculation resource and the transmission resource;
expressing the resource loss of the nth server node, wherein the resource loss is a dimensionless sum result of the access resource loss, the calculation resource loss and the transmission resource loss;
representing the actual resource usage of the nth server node.
Furthermore, RTC bidirectional transmission protocol is adopted among all server nodes, primary user nodes and all secondary user nodes to push video data. The RTC bidirectional transmission protocol can realize high-speed transmission of packet loss resistance and delay jitter resistance; inside the distribution network, the RTC protocol is used for realizing the transmission control of the network packet level, thereby reducing the cost of returning the source; when the network is distributed to the user terminal, the RTC protocol is used to avoid frequent transcoding of the video stream, thereby reducing the time delay of the distribution network and the cost of computing resources.
On the other hand, as shown in fig. 2, the present invention also provides a video peer-to-peer stream switching system, which at least includes: a plurality of server nodes connected by adopting a peer-to-peer architecture, wherein each server node is used for accessing a user node; the local storage of each server node is gathered into a virtual shared storage, and the video data sent to any server node by a video publishing user is received and stored; and transmitting the video data between each server node and each user node by adopting a real-time audio protocol.
Each server node is also used for receiving an acquisition request of the first-level user node for the appointed video data and searching a plurality of second-level user nodes which have acquired the appointed video data; calculating a correlation score between a primary user node and each secondary user node by adopting a preset user perception guarantee algorithm, selecting one secondary user node as a correlation user node according to the correlation score, and calculating the correlation score by adopting the preset user perception guarantee algorithm at least based on a content correlation index, a region proximity index and a plurality of network performance metrics; and initiating subscription to the associated user node so that the associated user node directly transmits the specified video data to the primary user node.
In some embodiments, each server node calculates, by using a preset user perception assurance algorithm, a correlation score between the primary user node and each secondary user node, where the calculation formula is:
wherein S represents a relevance score, V is a content relevance index, the value of V is 0 when relevant, and the value of V is 1 when irrelevant;
is a regional proximity index;
representing a vector matrix composed of a plurality of network performance metrics;
a coefficient which is a regional proximity index;
is composed of
The coefficient matrix of (2). The network performance metrics include at least: round Trip Time (RTT), throughput, and packet loss rate.
Each server node utilizes a decision scheduler to perform resource loss constraint optimization according to access resources, transmission resources and computing resources of each server node in a peer-to-peer video stream switching network, and the resource loss constraint optimization method comprises the following steps: the method comprises the steps of maximizing the use amount of global access resources, transmission resources and computing resources of all server nodes in a peer-to-peer video stream switching network, and performing matching scheduling on the access resources, the transmission resources and the computing resources of video data in each server node, wherein the expression is as follows:
wherein R represents a set of access resources, computing resources and transmission resources, and N is the total number of server nodes;
the theoretical maximum benefit of the nth server node is the dimensionless addition result of the theoretical access resource, the calculation resource and the transmission resource;
indicating resource loss of the nth server nodeThe resource loss is a dimensionless sum result of the access resource loss, the calculation resource loss and the transmission resource loss;
representing the actual resource usage of the nth server node.
Specifically, the operation of the video peer-to-peer stream switching system described in this embodiment may refer to the description of the video peer-to-peer stream switching method described above.
In another aspect, the present invention also provides an electronic device, which includes a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the steps of the method.
In another aspect, the present invention also provides a computer readable storage medium, on which a computer program is stored, which program, when executed by a processor, performs the steps of the above method.
The invention is illustrated below with reference to a specific example:
a flexible and extensible video transmission architecture and an update of a video transmission protocol are urgently needed in the current video transmission field. The present embodiment proposes a peer-to-peer interconnect transport structure based on the full link RTC protocol,
the technical problem to be solved is mainly in two aspects, namely firstly optimizing a tree distribution structure of a central distribution mechanism and secondly optimizing and replacing a traditional transmission protocol. The key problems with conventional distribution network architectures are: firstly, controlling rigidity and failing to process fine-grained information; second, the lack of effective large-scale, multiple concurrent control; thirdly, the strategy issuing time delay is larger. To solve the above problems, the present embodiment mainly adopts two parts of innovation points: firstly, decentralized intelligent control of a non-hierarchical peer-to-peer network; second, the "producer-terminal" employs the full link RTC bidirectional protocol.
Specifically, this embodiment provides a method and a system for exchanging video peer-to-peer streams, where the system forms a peer-to-peer video stream exchange network based on multiple server nodes, and each server node adopts a peer-to-peer network structure to implement decentralization. The end devices of the users are used as user nodes to access the peer-to-peer video stream switching network.
Therefore, in the embodiment, intelligent control is realized through a decentralized non-hierarchical peer-to-peer network. The network architecture of the equality is adopted to replace a tree-shaped centralized distribution network architecture, so that decentralization is realized. The so-called equation pair, that is, all distribution nodes are no longer hierarchical, in this embodiment, a transmission framework based on a point-to-point (P2P) technology is introduced, and all nodes are all deployed with control platforms to dynamically monitor a network of a transmission link and dynamically analyze a user QoE index in a return request, thereby implementing intelligent and fine-grained distribution control. Further, the full link from the video producer to the user terminal adopts an RTC bidirectional transmission protocol to carry out video data transmission. The present embodiment proposes to use RTC protocol in the distribution network full link to replace the conventional RTMP and other various streaming video encapsulation protocols. Specifically, RTC components are embedded in a server side and a client side, and a subscription-push transmission framework is built by using a related message mechanism. The RTC protocol can be used for realizing high-speed transmission of packet loss resistance and delay jitter resistance from a video producer to a distribution network; inside the distribution network, the RTC protocol is used for realizing the transmission control of the network packet level, thereby reducing the cost of returning the source; when the network is distributed to the user terminal, the RTC protocol is used to avoid frequent transcoding of the video stream, thereby reducing the time delay of the distribution network and the cost of computing resources.
In the peer-to-peer video stream switching network of this embodiment, each server node and each user node are associated with each other equally, each user node includes a main broadcaster and a viewer as a video publisher, and each server node and each user node not only stores its unique "hot data", but also can store concurrent cooperative data in association with other peer nodes, which is a "store-and-forward" manner of a non-full network image that reduces redundant storage. Under this network, key modules of the network node need to be considered, including: capacity management, routing back to source, elastic scaling, mirror image construction, information management and the like. As shown in fig. 2, the system and algorithm design is performed from 3 levels in this embodiment, including: and the method controls sinking, data exchange and management arrangement, and provides a key technology for constructing a peer-to-peer video stream exchange network.
Regarding the control sink part, the existing video streams build a video data transmission network based on the CDN, but the static tree structure of the CDN results in a weak control management capability of the video network. In the peer-to-peer video stream switching network provided in this embodiment, a traditional tree CDN network architecture is converted into a peer-to-peer architecture with mesh-type nodes, and a control plane originally located at a high layer of a tree network is sunk, so that a video provider can monitor and allocate each node of the network flexibly. The innate advantage of the VSEN architecture can directly sink the control plane of the video stream to each node of the peer-to-peer network for direct interactive management of "store-forward-control", without delaying back to the source through the original lengthy multi-level tree-like long path.
Specifically, the core innovation of "VSEN-control sinking" is "decentralized" intelligent control, a video provider does not need a central control node any more, each node in a peer-to-peer network has a complete video flow control strategy, and fine-grained user quality of service (QoE) feedback can be directly obtained from a direct terminal source return message of the node, so that intelligent data analysis and low-delay fine-grained regulation and control are performed. In order to implement "VSEN-control convergence", the control plane of the peer-to-peer video network is optimized in three aspects, specifically: firstly, uniformly arranging slicing and serving cloud network resources; secondly, virtualizing and atomically packaging the cloud network function; third, a generalized and standardized hardware configuration. Overall, the VSEN designed by this embodiment implements a fine-grained, low-delay video stream QoE control plane.
Regarding the data exchange part, in a conventional video network, video data is sent from a video producer (anchor/video provider video source) to a server node (the node may be in a tree structure of a CDN or a peer structure of a peer-to-peer network), and then forwarded to a terminal side device (a smartphone, a computer, etc.). Macroscopically, the video stream is transmitted in series in a segmented form in the network transmission process, and in the original video network, the video stream can be roughly divided into 3 segments of transmission paths: the video producer reaches the CDN, the CDN is sent down by hierarchical nodes, and a CDN end node reaches the end user. Since the existing video network is a CDN transmission method (designed for file transmission) and does not optimize video streaming specifically, there is often a large difference in the transmission protocols between the above 3 segments of transmission paths, and the inefficiency of any one segment of link results in the degradation of the end user qoe (quality of experience). For example, a video between CDN hierarchical nodes issues a multi-purpose RTMP protocol (existing protocols for multiplexing CDNs are low in development difficulty and initial deployment cost), but complex node source return information and low-efficiency RTMP directly cause significant link delay and packet loss delay, and are more disadvantageous for feeding back a true fine-grained QoE. Second, the "CDN end node to end user" link often uses non-RTMP protocols, such as FLV, which results in more transcoding delay and low scalability. On the whole, the video stream transmitted from the video producer to the terminal lacks dynamic sensing capability, the source return cost is high, the cache mode causes serious problems of high delay, low control efficiency and the like, and a novel protocol with customization, low delay and high expandability is urgently needed.
In this embodiment, a full-link RTC (Real-time Communication) protocol is designed for the peer-to-peer video stream switching network, so as to achieve centralized storage without additional transcoding of intermediate nodes. In particular, any server node, even a user node, in the VSEN network can do the upstream or downstream of the video stream.
In this embodiment, users in the peer-to-peer video stream switching network are also expanded to be peer nodes, and real-time data exchange of a full-link RTC can be performed between users. For example, a user a holds video frame data, and a user B in the peer-to-peer network can continuously obtain video data from the user a in a "subscription-push" manner to the user a without consuming network bandwidth of the CDN node. In order to ensure the QoE of the peer-to-peer user nodes under the mechanism, the present embodiment designs a long and short-term QoE guarantee algorithm between peer-to-peer user nodes, and performs modeling and scoring according to information such as network similarity, content association, geographical proximity, network load, and short-term network mutation between peer-to-peer user nodes, where a specific calculation method may refer to calculation formula 1.
For the management and arrangement part, along with the rapid increase of the user scale of the video network, the video network needs to deploy a large number of nodes to ensure the good QoE of the user, and in the CDN network, because certain hierarchical management and distributed storage are needed among the nodes, the interaction overhead among the nodes presents polynomial increase O (N) while the QoE is approximately ensured 2 ) Forming a big bottleneck of service capability expansion, which is also a big problem in the field, namely: how to ensure the linear increase of node overhead-keep linear while enlarging the network scale.
The embodiment provides a special node interaction framework for a peer-to-peer video stream switching network and provides a cloud-network storage-transmission-computation resource unified packaging, arranging and scheduling method. As shown in fig. 3, the usage amount of global access resources, transmission resources, and computational resources of all server nodes in the peer-to-peer video stream switching network is maximized, and the access resources, transmission resources, and computational resources of video data in each server node are subjected to matching scheduling, which may specifically refer to calculation formula 2.
The core innovation of the cloud-network storage-transmission-computation framework provided by the embodiment lies in that a node mesh structure is formed, the back source rate and the redundancy concurrency are effectively reduced, the resource synchronization and sharing are facilitated, and the characteristic can be used for realizing the interaction overhead O (N) of the original video network based on the CDN 2 ) The video network video platform system reduces the video network video platform system to O (N) based on a VSEN network, improves the expandability of the network scale, and ensures the linear growth of the scale-service capability of the platform, thereby providing a very solid foundation for further enriching and enlarging the existing video network.
In summary, the embodiment provides a peer-to-peer video stream switching network VSEN designed for customizing video stream application, and a novel and unique algorithm and framework design are performed from three aspects of control sinking, data exchange and management arrangement, so that a novel video network with high reliability, low delay, fine-grained regulation and control and low interaction overhead is finally provided. The method has the advantages that great innovation is realized in the field of real-time video transmission and distribution, the provided decentralized intelligent control realizes intelligent data analysis, means for accurate distribution control are provided, an RTC (real time clock) transmission protocol is used in a full link of a distribution network, transmission time delay is optimized, bidirectional node perception is realized, fine granularity control can be realized, and the development of a novel video transmission architecture is promoted.
The embodiment provides a decentralized peer-to-peer video streaming transmission network architecture, and sinks a control strategy to a terminal node of a peer-to-peer network, and the novel transmission network architecture is a first protection point of the patent. The embodiment designs and realizes a transmission system for monitoring network dynamics and user side service quality in a fine granularity manner, realizing intelligent data analysis and performing fine scheduling decision, and the customized fine regulation and control is the second protection point of the embodiment. The embodiment provides a video transmission framework of a full-link RTC bidirectional transmission protocol based on a 'producer-video terminal', all nodes have uplink and downlink capabilities, transcoding consumption is not needed, real-time transmission and service quality monitoring at a network packet or video frame level can be performed, and the full-link RTC transmission protocol suitable for real-time video streaming is a third protection point of the embodiment. The patent proposes a storage-transmission-calculation integrated node interaction framework, which guarantees linear increase of node overhead, and is called a "linearity preserving" method, which is a fourth protection point of the embodiment.
In summary, the video peer-to-peer stream switching method, system and device of the present invention set up that each server node adopts an equality network structure, do not differentiate hierarchies, and aggregate the local storage of each server node into a virtual shared storage, thereby implementing decentralized. The full links between the server nodes and the user nodes are set to communicate by adopting a real-time audio protocol, so that the time delay and the calculation cost of a distribution network are reduced. By building a subscription pushing mechanism, when a user requests video data, a preset user perception guarantee algorithm is adopted to calculate a correlation score between a current user node and a user node which has obtained appointed video data, the user node with higher correlation is selected as a correlated user node to subscribe, and the correlated user node is directly enabled to push the video data to the current user node, so that network delay is greatly reduced, and user experience quality is guaranteed.
Those of ordinary skill in the art will appreciate that the various illustrative components, systems, and methods described in connection with the embodiments disclosed herein may be implemented as hardware, software, or combinations of both. Whether this is done in hardware or software depends upon the particular application and design constraints imposed on the solution. Skilled artisans may implement the described functionality in varying ways for each particular application, but such implementation decisions should not be interpreted as causing a departure from the scope of the present invention. When implemented in hardware, it may be, for example, an electronic circuit, an Application Specific Integrated Circuit (ASIC), suitable firmware, plug-in, function card, or the like. When implemented in software, the elements of the invention are the programs or code segments used to perform the required tasks. The program or code segments may be stored in a machine-readable medium or transmitted by a data signal carried in a carrier wave over a transmission medium or a communication link. A "machine-readable medium" may include any medium that can store or transfer information. Examples of a machine-readable medium include electronic circuits, semiconductor memory devices, ROM, flash memory, Erasable ROM (EROM), floppy disks, CD-ROMs, optical disks, hard disks, fiber optic media, Radio Frequency (RF) links, and so forth. The code segments may be downloaded via computer networks such as the internet, intranet, etc.
It should also be noted that the exemplary embodiments mentioned in this patent describe some methods or systems based on a series of steps or devices. However, the present invention is not limited to the order of the above-described steps, that is, the steps may be performed in the order mentioned in the embodiments, may be performed in an order different from the order in the embodiments, or may be performed simultaneously.
Features that are described and/or illustrated with respect to one embodiment may be used in the same way or in a similar way in one or more other embodiments and/or in combination with or instead of the features of the other embodiments in the present invention.
The above description is only a preferred embodiment of the present invention, and is not intended to limit the present invention, and various modifications and changes may be made to the embodiment of the present invention by those skilled in the art. Any modification, equivalent replacement, or improvement made within the spirit and principle of the present invention should be included in the protection scope of the present invention.
Claims (8)
1. A method of video peer-to-peer stream switching operating in a peer-to-peer video stream switching network comprising a plurality of server nodes, each server node connected using a peer-to-peer architecture, the method comprising:
the local storage of each server node is gathered into a virtual shared storage, and video data sent to any server node by a video publishing user is received and stored;
receiving an acquisition request of a primary user node for appointed video data, and searching a plurality of secondary user nodes which have acquired the appointed video data;
calculating the correlation scores between the primary user nodes and the secondary user nodes by adopting a preset user perception guarantee algorithm, wherein the calculation formula is as follows:
wherein S represents a relevance score, V is a content relevance index, the value of V is 0 when relevant, and the value of V is 1 when irrelevant;
is a regional proximity index;
representing a vector matrix composed of a plurality of network performance metrics;
a coefficient which is a regional proximity index;
is composed of
A coefficient matrix of (a);
selecting a secondary user node as a related user node according to the relevance score, wherein the preset user perception guarantee algorithm is at least based on a content relevance index, a region proximity index and a plurality of network performance metrics to calculate the relevance score;
initiating a subscription to the associated user node to enable the associated user node to directly transmit the specified video data to the primary user node;
and performing resource loss constraint optimization by using a decision scheduler according to access resources, transmission resources and computing resources of each server node in the peer-to-peer video stream switching network;
and the video data is transmitted among all the server nodes, the primary user node and all the secondary user nodes by adopting a real-time audio protocol.
2. The method of claim 1, wherein aggregating the local storage of each server node into a virtual shared storage comprises: and each server node stores the video data by adopting a VSEN architecture.
3. The video peer-to-peer stream exchange method of claim 1, wherein the network performance metrics comprise at least: round trip delay, throughput, and packet loss rate.
4. The video peer-to-peer stream switching method of claim 1, wherein performing resource loss constraint optimization based on access resources, transmission resources and computational resources of each server node in said peer-to-peer video stream switching network using a decision scheduler comprises: and maximizing the overall access resources, transmission resources and computing resource usage of all server nodes in the peer-to-peer video stream switching network, and performing matching scheduling on the access resources, the transmission resources and the computing resources of the video data in each server node.
5. The method of claim 4, wherein the access resource, transmission resource and computing resource usage of all server nodes in the peer-to-peer video stream switching network are maximized globally, and the expression is:
wherein R represents a set of access resources, computing resources and transmission resources, and N is the total number of server nodes;
the theoretical maximum benefit of the nth server node is the dimensionless addition result of the theoretical access resource, the calculation resource and the transmission resource;
representing the resource loss of the nth server node, wherein the resource loss is a dimensionless sum result of the access resource loss, the computing resource loss and the transmission resource loss;
representing the actual resource usage of the nth server node.
6. A video peer-to-peer stream switching system, comprising at least: a plurality of server nodes connected by adopting a peer-to-peer architecture, wherein each server node is used for accessing a user node; the local storage of each server node is gathered into a virtual shared storage, and the video data sent to any server node by a video publishing user is received and stored; transmitting the video data between each server node and each user node by adopting a real-time audio protocol;
each server node is also used for receiving an acquisition request of a primary user node for the appointed video data and searching a plurality of secondary user nodes which have acquired the appointed video data; calculating a correlation score between the primary user node and each secondary user node by adopting a preset user perception guarantee algorithm, and selecting one secondary user node as a correlation user node according to the correlation score, wherein the preset user perception guarantee algorithm is at least based on a content correlation index, a region proximity index and a plurality of network performance metrics to obtain the correlation score; initiating a subscription to the associated user node to enable the associated user node to directly transmit the specified video data to the primary user node;
wherein, each server node adopts a preset user perception guarantee algorithm to calculate the relevance score between the primary user node and each secondary user node, and the calculation formula is as follows:
wherein S represents a relevance score, V is a content relevance index, the value of V is 0 when relevant, and the value of V is 1 when irrelevant;
is a regional proximity index;
representing a vector matrix composed of a plurality of network performance metrics;
a coefficient which is a regional proximity index;
is composed of
A coefficient matrix of (a);
each server node utilizes a decision scheduler to perform resource loss constraint optimization according to access resources, transmission resources and computing resources of each server node in the peer-to-peer video stream switching network, and the resource loss constraint optimization method comprises the following steps: maximizing the use amount of global access resources, transmission resources and computing resources of all server nodes in the peer-to-peer video stream switching network, and performing matching scheduling on the access resources, the transmission resources and the computing resources of video data in each server node, wherein the expression is as follows:
wherein R represents a set of access resources, computing resources and transmission resources, and N is the total number of server nodes;
the theoretical maximum benefit of the nth server node is the dimensionless addition result of the theoretical access resource, the calculation resource and the transmission resource;
representing the resource loss of the nth server node, wherein the resource loss is a dimensionless sum result of the access resource loss, the calculation resource loss and the transmission resource loss;
representing the actual resource usage by the nth server node.
7. An electronic device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, characterized in that the processor implements the steps of the method according to any of claims 1 to 5 when executing the program.
8. A computer-readable storage medium, on which a computer program is stored which, when being executed by a processor, carries out the steps of the method according to any one of claims 1 to 5.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210407106.8A CN114513519B (en) | 2022-04-19 | 2022-04-19 | Video peer-to-peer stream exchange method, system and device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210407106.8A CN114513519B (en) | 2022-04-19 | 2022-04-19 | Video peer-to-peer stream exchange method, system and device |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN114513519A CN114513519A (en) | 2022-05-17 |
| CN114513519B true CN114513519B (en) | 2022-09-09 |
Family
ID=81555059
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210407106.8A Active CN114513519B (en) | 2022-04-19 | 2022-04-19 | Video peer-to-peer stream exchange method, system and device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114513519B (en) |
Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108521577A (en) * | 2018-04-11 | 2018-09-11 | 武汉斗鱼网络科技有限公司 | A kind of video broadcasting method, device, equipment and storage medium |
Family Cites Families (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US10979674B2 (en) * | 2013-07-22 | 2021-04-13 | Intellivision | Cloud-based segregated video storage and retrieval for improved network scalability and throughput |
| US20190342626A1 (en) * | 2018-05-02 | 2019-11-07 | Andres Hernandez | Method of Implementing A Peer-To-Peer On-Demand Live Video Streaming Service |
| CN111556344A (en) * | 2020-04-04 | 2020-08-18 | 网宿科技股份有限公司 | Method, device, equipment and system for downloading video stream |
-
2022
- 2022-04-19 CN CN202210407106.8A patent/CN114513519B/en active Active
Patent Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108521577A (en) * | 2018-04-11 | 2018-09-11 | 武汉斗鱼网络科技有限公司 | A kind of video broadcasting method, device, equipment and storage medium |
Also Published As
| Publication number | Publication date |
|---|---|
| CN114513519A (en) | 2022-05-17 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| USRE49943E1 (en) | System and method for a context layer switch | |
| Shen et al. | Peer-to-peer media streaming: Insights and new developments | |
| US10986387B1 (en) | Content management for a distributed cache of a wireless mesh network | |
| CN110336843B (en) | Content distribution method for crowdsourcing, central node and edge node | |
| Thouin et al. | Video-on-demand networks: design approaches and future challenges | |
| US7818445B2 (en) | Methods and devices for obtaining a broadcast-like streaming content | |
| US6970939B2 (en) | Method and apparatus for large payload distribution in a network | |
| CN102326161A (en) | The buffer memory cooperation that topology is responsive | |
| JP2015503146A (en) | Distributed mapping function for large-scale media cloud | |
| Fan et al. | Video delivery networks: Challenges, solutions and future directions | |
| Tuncer et al. | Scalable cache management for ISP-operated content delivery services | |
| Al-Abbasi et al. | Fasttrack: Minimizing stalls for cdn-based over-the-top video streaming systems | |
| US11089103B1 (en) | Content management in a distributed cache of a wireless mesh network | |
| Llorca et al. | Energy efficient dynamic content distribution | |
| CN102158767A (en) | Scalable-coding-based peer to peer live media streaming system | |
| Pal et al. | Priority-based scheduling scheme for live video streaming in peer-to-peer network | |
| CN114513519B (en) | Video peer-to-peer stream exchange method, system and device | |
| Ghandeharizadeh et al. | Cooperative caching techniques for continuous media in wireless home networks | |
| Jiang et al. | A replica placement algorithm for hybrid CDN-P2P architecture | |
| Yue et al. | Survey on scheduling technologies of P2P media streaming | |
| CN109246487B (en) | Intelligent scheduling system | |
| Alaya et al. | QoS enhancement In VoD systems: load management and replication policy optimization perspectives | |
| CN101540884B (en) | Construction method of equivalent VoD system based on jump graph | |
| Fekih et al. | Quality of experience aware replication framework for video streaming in content-centric mobile networks based on SDN architecture | |
| Riad et al. | A framework for cloud P2P VoD system based on user's behavior analysis |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |