WO2016061898A1 - 直播间的频道访问方法和系统 - Google Patents

直播间的频道访问方法和系统 Download PDF

Info

Publication number
WO2016061898A1
WO2016061898A1 PCT/CN2014/096019 CN2014096019W WO2016061898A1 WO 2016061898 A1 WO2016061898 A1 WO 2016061898A1 CN 2014096019 W CN2014096019 W CN 2014096019W WO 2016061898 A1 WO2016061898 A1 WO 2016061898A1
Authority
WO
WIPO (PCT)
Prior art keywords
client
channel
clients
live broadcast
access
Prior art date
Application number
PCT/CN2014/096019
Other languages
English (en)
French (fr)
Inventor
王树
李锦�
郭锦洋
古培昌
Original Assignee
广州华多网络科技有限公司
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 广州华多网络科技有限公司 filed Critical 广州华多网络科技有限公司
Publication of WO2016061898A1 publication Critical patent/WO2016061898A1/zh

Links

Images

Classifications

    • 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]
    • H04N21/40—Client devices specifically adapted for the reception of or interaction with content, e.g. set-top-box [STB]; Operations thereof
    • H04N21/47—End-user applications
    • H04N21/472—End-user interface for requesting content, additional data or services; End-user interface for interacting with content, e.g. for content reservation or setting reminders, for requesting event notification, for manipulating displayed content

Definitions

  • the present invention relates to the field of Internet technologies, and in particular, to a channel access method and system between live broadcasts.
  • online live broadcast has achieved rapid development, and a large number of online live broadcast systems, such as voice live broadcast system, video on demand live broadcast system and education training live broadcast system, have appeared.
  • the online live broadcast system is mainly composed of live broadcast rooms, and each live broadcast room has multiple different types of channels, such as entertainment channels or music channels. After the client logs in to the live broadcast room, you can enter the live content of the anchor in the channel.
  • the number of clients that log in to these online live broadcast systems is large, and higher requirements are imposed on network bandwidth and server processing capabilities.
  • the online video playing system broadcasts the live content to the client through the broadcast server.
  • the processing pressure on the broadcast server continues to increase.
  • the broadcast server sets the upper limit of the number of online viewers.
  • the number of clients that the broadcast server allows to view the live content online does not exceed the upper limit, thereby alleviating the processing pressure of the broadcast server.
  • the upper limit of the number of online viewers at the same time is limited.
  • the unconnected client may continuously try to enter the channel and repeatedly send the incoming channel request.
  • the number of incoming channel requests that the broadcast server needs to process is multiplied, causing the processing pressure of the broadcast server to be too large, and even causing the broadcast server to crash, affecting The stability of the live broadcast system, and the live server cannot effectively manage the client.
  • a channel access method for a live broadcast includes the following steps:
  • the service information of the channel is sent to the first client
  • the first client is added to the queuing list for caching.
  • a channel access system for a live broadcast comprising:
  • connection module configured to acquire an incoming channel request of the first client, and establish a connection of the first client to a channel corresponding to the incoming channel request;
  • the determining module determines whether the number of clients that the channel is currently online accessing reaches a predetermined access threshold
  • a live broadcast module configured to send the service information of the channel to the first client if the access threshold is not reached;
  • the queuing module is configured to add the first client to the queuing list for caching if the access threshold has been reached.
  • a channel access system between live broadcasts comprising: a receiver, a processor and a transmitter;
  • the processor is configured to acquire, by the receiver, an incoming channel request of the first client, and establish a connection of the first client to a channel corresponding to the incoming channel request;
  • the processor is configured to determine whether the number of clients that the channel is currently online accessing reaches a predetermined access threshold
  • the transmitter is configured to send the service information of the channel to the access threshold if the access threshold is not reached.
  • the processor is configured to add the first client to the queuing list to perform caching if the access threshold has been reached.
  • the channel access method in the live broadcast of the above embodiment by receiving and responding to the client entering the channel request, and establishing a connection with the channel corresponding to the incoming channel request; using the predetermined access threshold to determine the client that the channel is currently online accessing
  • the number of terminals if the access threshold is not reached, controls the client to watch the live content; if the predetermined access threshold is reached, the client is added to the queuing list for caching.
  • the number of clients in the live broadcast channel is simultaneously watched, and the client exceeding the access threshold is added to the queuing list for caching, thereby effectively preventing the client exceeding the access threshold from repeatedly sending the channel request.
  • the number of incoming channel requests that need to be processed by the server in the live broadcast is reduced, the processing pressure of the live broadcast server is alleviated, the stability of the live broadcast system is improved, and the client access of the live broadcast channel is also effectively managed.
  • FIG. 1 is a schematic structural diagram of an application system of an embodiment
  • FIG. 2 is a flowchart of a channel access method between live rooms in an embodiment
  • FIG. 3 is a flowchart of processing of a client entering a channel in an embodiment
  • Figure 4 is a schematic diagram of an application UI
  • Figure 5a is a first schematic diagram of notifying a client by using a sliding window
  • Figure 5b is a second schematic diagram of notifying a client by using a sliding window
  • FIG. 6 is a system structural diagram of a server of an application example
  • FIG. 7 is a timing diagram of a client entering a channel of an example
  • FIG. 8 is a timing diagram of channel client data repair of an example
  • FIG. 9 is a schematic structural diagram of a channel access system between live broadcasts according to an embodiment
  • FIG. 10 is a schematic structural diagram of a channel access system between live broadcasts according to another embodiment
  • FIG. 11 is a schematic structural diagram of a channel access system between live broadcasts according to an embodiment.
  • FIG. 1 is a For a schematic diagram of the application system structure of the embodiment, refer to FIG. 1 , which includes the terminal device 02 and the server terminal 04.
  • the terminal device 02 may be a device such as a PC (Personal Computer), a Smart Phone, a Tablet PC (Patal Personal Computer), or a PDA (Personal Digital Assistant), but is not limited. herein.
  • PC Personal Computer
  • PDA Personal Digital Assistant
  • the terminal device 02 can access the public network by means of wired or wireless, and communicate with the server 04.
  • the public network can be a GPRS (General Packet Radio Service) or an Internet (Internet).
  • a client can be run on the terminal device 02, and the client can access the live channel.
  • the client accesses the server 04 through the public network and enters the live content required for the corresponding channel request.
  • the client presents the live content returned by the server 04 to the user.
  • the terminal device 02 may refer to one device or multiple devices.
  • the server 04 includes a server device that implements various background functions, and may be a server device or a server group composed of multiple server devices.
  • FIG. 2 is a flowchart of a channel access method between live rooms in an embodiment, including the following steps:
  • Step S10 Acquire an incoming channel request of the first client, and establish a connection between the first client and a channel corresponding to the incoming channel request.
  • the entry channel request carries at least the identifier of the first client and the channel identifier of the channel to be entered.
  • the channel corresponding to the incoming channel request is the channel corresponding to the channel identifier carried by the incoming channel request.
  • the live broadcast server responds to the incoming channel request, according to the identifier of the first client carried in the incoming channel request.
  • the channel identifier establishes a communication connection between the first client and the channel corresponding to the channel identifier.
  • step S20 it is determined whether the number of clients currently accessed by the channel has reached a predetermined access threshold. If yes, step S40 is performed, and if no, step S30 is performed.
  • the access threshold of the client that is accessed online in the channel of the live broadcast is preset, and the access threshold is used to control the number of clients viewing the live content. After the new client enters the channel, it is determined whether the number of clients currently accessed online reaches a predetermined access threshold, thereby determining whether the new client needs to be queued.
  • Step S30 If the access threshold is not reached, the service information of the channel is sent to the first client.
  • the service information of the channel is obtained, and the service information of the channel is sent to the first client, so that A user corresponding to a client can watch the live content of the channel.
  • Step S40 If the access threshold has been reached, the first client is added to the queuing list for caching.
  • the first client that enters the channel is added to the queue list corresponding to the channel for caching, thereby realizing Effective control and management of clients entering the channel.
  • the ID of the client may be the data element of the queuing list
  • the ID of the first client may be the login account of the user corresponding to the first client.
  • the logical order of the data elements is arranged in the queuing list with the time the first client entered the channel. The client in the queue list is temporarily not allowed to view the live content of the channel.
  • the channel access method of the live broadcast provided by the above steps, by receiving and responding to the incoming channel request of the first client, and establishing a connection between the first client and the channel corresponding to the incoming channel request; determining the using the predetermined access threshold Whether the number of clients that the channel currently has online access reaches the access The threshold value is used to control the first client to watch the live content if the access threshold is not reached; if the predetermined access threshold is reached, the first client is added to the queue list corresponding to the channel for caching.
  • the number of clients watching live content in the channel of the live broadcast is limited, and the excessive number of clients watching the live content at the same time is prevented from causing excessive processing pressure on the live broadcast server of the channel, thereby eliminating the live broadcast.
  • the system crash occurs, which improves the stability of the live broadcast system, and also enables effective management of clients accessing the channels of the live broadcast.
  • Redis is a key-value (distributed storage system) storage system that supports storing a variety of value types, including string (string), list (linked list), set (set), zset (sorted set, ordered Collection) and hash (hash type) and so on. These data types support rich operations such as push/pop, add/remove, take intersection, fetch and combine. And Redis also supports sorting in a variety of different ways.
  • a linked list such as a skiplist can be used to implement an efficient queuing algorithm.
  • the jump table is a type of linked list, and the jump table is mainly composed of the following parts:
  • head a node pointer responsible for maintaining the jump table
  • node the actual value of the element is saved, each node has one or more layers;
  • tail all consist of null (empty).
  • the jump table adds a jump function to the linked list, which makes it easy to implement Log N calculations.
  • the jump table is a randomized data structure. Based on the parallel linked list, it is possible to perform operations such as finding, deleting or inserting in a simple and intuitive manner within a logarithmic desired time. With a jump table, the traversal process always starts at the upper level and then decreases as the range of element values decreases until it reaches the lower level.
  • the channel access method of the live broadcast room of the present invention may further include the following steps:
  • the exit channel request carries at least the identifier of the second client and the channel identifier of the channel to be exited.
  • the channel corresponding to the exit channel request is the channel corresponding to the channel identifier carried by the exit channel request.
  • Sorting if the second client is deleted in the currently online client, then one client in the queue list can be moved into the client of the online access, so that the order in the queue list changes, and the update is performed.
  • the logical order of the data elements in the queue list if the second client is deleted in the queue list, the corresponding data elements are deleted, the order in the queue list changes, and the logical order of the data elements in the queue list is updated.
  • the channel access method of the live broadcast of the present invention may further include the following steps before the step of determining whether the number of clients currently accessed by the channel reaches the predetermined access threshold:
  • the solution of the above steps realizes more effective management after the client enters the channel, and sets the corresponding service level to the client through the technical means based on the management purpose, which can not only realize the value added of the service, but also can be used as the performance evaluation of the live room. index.
  • FIG. 3 is a flowchart of processing a client entering a channel according to an embodiment; and the method includes the following steps:
  • S304 Put the first client into the client set that is currently online access, and notify the user to watch the live broadcast;
  • S307 The first client is placed in a client set that is currently online accessed, and is notified to watch the live broadcast;
  • the process of the above-mentioned client entering the channel sets a queuing-free client by setting a corresponding service level to the client, and the queuing-free client can directly access the channel to watch the live content, and prompt the client to select a higher service level.
  • the value added of the service can be realized, and the service level of the client entering the channel can also be used as an evaluation indicator of the performance of the live broadcast corresponding to the channel.
  • FIG. 4 is a schematic diagram of an application UI (User Interface).
  • the VIP client queue-free client
  • the VIP client can quickly enter the channel through the corresponding channel, and the ordinary client. Then you need to wait in line.
  • the channel access method of the live broadcast room of the present invention may further include the following steps:
  • the solution of the above steps after the first client enters the channel, if the first client needs to row
  • the client of the team can notify the first client of the current queued information in real time, such as the ranking information, that is, the number of clients in front of the first client, thereby improving the application experience of the first client.
  • the present invention provides an improved notification scheme, periodically. Notify all queued clients of their queue status.
  • the step of separately sending the ranking information of each first client to each first client may include the following:
  • the times when the first client ID1, ID2, and ID3 enter the channel are t0, t1, and t2, respectively, and the times when the first client 1, 2, and 3 enter the channel are t0, t1, and t2, respectively.
  • Establishing a sliding time axis as shown in FIG. 5a sliding the sliding window along the sliding time axis.
  • the first client ID1 and ID2 are located in the sliding window, and the first client ID1 and ID2 are arranged.
  • the bit information is sent to the first client ID1 and ID2 respectively.
  • the second time is swiped, the first client ID3 is located in the sliding window, and the ranking information of the first client ID3 is sent to the first client ID3.
  • FIG. 5b is a schematic diagram of notifying a client by using a sliding window.
  • the sliding window moves along the sliding time axis on the queuing client, and only the client located in the sliding window is notified each time sliding, in general,
  • the sliding window can be set to slide the entire sliding time axis at a preset time to notify all the queuing clients once every preset time, and the preset time can be 10 seconds or 20 seconds.
  • the number of the first clients entering the sliding window is the same each time, and enters the first client in the sliding window when sliding twice in succession. There is no duplicate first client. For example, if the number of the first clients entering the sliding window is 5 each time, when the sliding window is slid twice, the first two clients entering the sliding window have a total of 10, which is The same first client does not exist in the 10 first clients.
  • the solution of the foregoing embodiment can improve the utilization efficiency of the server resources, reduce the impact of the client on the operation of the server between the live broadcasts when the number of queued clients is large, and reduce the burden on the server between the live broadcasts.
  • a scheme for notifying the client of the information currently queued by the client after entering the channel is provided, but in the actual application, the notification message needs to be forwarded in multiple nodes, and if a single point of failure occurs, the retransmission may occur.
  • Some notification messages For example, a client has entered the channel but has repeatedly sent an incoming channel request to the server due to a failure of the client. Accordingly, in order to avoid processing the client that has entered the channel as a client of the new request due to the retransmission of the message, the duplicate message checking mechanism may be further set.
  • the repeated message checking mechanism may specifically include the following steps:
  • determining whether the incoming first client exists in the currently accessed client or in the queuing list if yes, determining the entered A client is an existing client, and sends the last sent ranking information to the first client; if not, it determines that the incoming client is a newly entered client, and the newly generated ranking information is Send to the first client.
  • Embodiments of the channel access method between the live broadcasts may also provide a data integrity check mechanism by which the mutual verification data is verified to be complete.
  • the channel access method of the live broadcast room of the present invention may further include an inspection scheme for the lost client exit channel message, the check scheme is mainly for some clients that have exited the channel, because the channel message is exited. Lost during the forwarding process, causing the client that exits the channel to still be cached in the currently online client or queued list, while still occupying the service resources.
  • the inspection mechanism proposed by this includes the following steps:
  • Get the current number of clients on the channel, and the current number of clients and the currently online clients The number of the end is compared with the sum of the number of the first client in the queuing list. If the current number of clients is inconsistent with the sum, the client that is not currently in the channel is found, the found client is deleted, and the updated client is updated. Queue the linked list.
  • the service node corresponding to the client accessing the channel maintains a client list. Through this client list, the data accuracy of the current channel can be obtained, because the client list is served by the service.
  • the node maintains, so the number of clients obtained through the client list is highly accurate, and the number of clients can be considered equal to the actual number of all clients currently entering the channel.
  • the anchor of the channel starts broadcasting, the anchor periodically records (such as every minute) the number of clients of the current channel, and uploads a heartbeat message to the server of the live broadcast, the heartbeat message carries the number of the client, and records the client of the current channel.
  • the live broadcast server triggers a full check once to check whether there is a client that is not on the channel in the client that has established the connection.
  • the checking process may be triggered in a timed manner, as follows:
  • the step of obtaining the current number of clients of the channel is not performed.
  • the step of obtaining the current number of clients of the channel is performed, and then the flag is rewritten in the cache.
  • the above-mentioned timing triggering strategy can be implemented by setting a flag bit that is invalidated within a set time (for example, 1 minute) in the cache.
  • a client enters the channel, if there is no such flag in the cache, the trigger is triggered.
  • this flag is present in the cache, and the check process is not triggered.
  • the method is triggered by the shortest period of time within the set time, so that when no client enters or exits the channel, the inspection process is not triggered, the repeated frequent triggering of the inspection process is avoided, and a better flow control function is realized.
  • the repair process is as follows:
  • the client in the front of the queue list is moved to the client set that is currently online access, and these are moved to The client sends a notification message to enter the channel; after the operation of the mobile client, the ranking of the queued list is updated by the number of remaining clients in the queue list, and the remaining ranking information of each client is generated, and the ranking information of each client is respectively Send to each client.
  • the processing scheme of the foregoing embodiment updates the currently accessed client and the queuing list in time, and can notify the client of the status of the client in time, thereby ensuring the real-time performance of the currently cached data.
  • FIG. 6 is a system structural diagram of a server of an application example, which mainly includes a service node, a session, a live broadcast server, a cache proxy, and a query process.
  • the service node maintains a list of clients on all clients within the channel and communicates with the client. Each channel corresponds to a session process, and the live broadcast server processes the incoming channel request or the exit channel request of all channels in the entire live broadcast.
  • the caching proxy caches client data in a linked list.
  • the client When a client (the client logged in by the user) enters the channel, the client sends a session notification to the service node, which informs at least the channel identifier of the carrying channel and the identity of the client.
  • Service The service node establishes a connection between the channel and the client according to the channel identifier.
  • the service node generates a message notification to the session process, and the message notification carries the identifier of the client, and the session process forwards the message notification to the live broadcast server.
  • the live server writes the identity of the client to the proxy cache.
  • FIG. 7 is a timing diagram of the client entering the channel of an example.
  • the process of the client entering the channel is: the client sends a channel message to the service node, and then the client logs in to the service node.
  • the service node forwards the incoming channel message to the session process, and the session process forwards the incoming channel message to the live broadcast server, and the live broadcast server inserts the client into the cache (ie, inserts the client into the currently online set of clients or queues) In the linked list), and returning a corresponding notification message (watching the live broadcast notification or queuing waiting notification) to the session process, the session process forwards the notification message to the service node, and the service node sends the notification message to the client.
  • the service node When a client exits the channel, the service node generates a message notification to the live broadcast server, the message notifying the identifier of the client and the channel identifier of the channel to be exited.
  • the server in the live broadcast first checks the set of clients that are currently online access according to the identifier of the client. If the identifier of the client is in the client set that is currently online, the client directly deletes the client. Identification, if the identity of the client is not in the client set that is currently online access, find the identity of the client from the queuing list and delete it, then update the ordering of the queuing list, modify the ordering of the queuing client, and obtain an update.
  • the queuing list is followed by the ranking information of the client behind the client, and the obtained ranking information is notified to the client behind the client.
  • the check scheme for losing the client exit channel message is as follows:
  • the anchor starts broadcasting
  • the number of clients connected to the current channel is obtained from the client list maintained by the service node, and the number of the client is uploaded to the live server.
  • the server in the live broadcast compares the number of clients with the number of all clients recorded, and checks the channel once in a full amount to determine whether there is still a client that has exited the channel in the currently connected client.
  • the specific scheme can set a flag for invalidation in a set time (such as 1 minute) in the cache.
  • a flag for invalidation in a set time (such as 1 minute) in the cache.
  • the server in the live broadcast sends a query notification message to the query process.
  • the query process queries the number of clients from the client list maintained by the service node, and sends the number of clients to the live broadcast server.
  • the live server compares the number of clients with the number of clients recorded, finds the abnormal client and cleans up the abnormal client, and then notifies the client of the updated latest ranking information.
  • FIG. 8 is an example of a channel client data repair timing diagram.
  • the client of the anchor periodically sends the message 1 to the service node, the service node forwards the message 2 to the session process, the session process forwards the message 3 to the live broadcast server, and the server test of the live broadcast is established.
  • the number of connected clients sends a query client number message to the client query process, and the client query process sends a query client number message to the service node.
  • the service node After receiving the query client number message, the service node returns a message 4 containing the number of clients to the client query process, and the client query process forwards the number of the client to the live server.
  • the server in the live broadcast compares the number of clients that have established a connection with the number of clients that are queried.
  • the client queries the process to send a query client list message, and the client query process forwards the query list message to the service node.
  • the node returns the client list maintained by itself to the client query process, and the client query process sends a message 5 containing the client list to the live broadcast server.
  • the server in the live broadcast server deletes the client that has exited the channel in the currently accessed client set according to the client of the current channel recorded in the client list, and then updates the ranking information in the queue list. Then, the updated ranking information is forwarded to the client by the session process, the client query process, and the service node through the notification message, and the latest queue order of the client is notified.
  • FIG. 9 is a schematic structural diagram of a channel access system in a live broadcast according to an embodiment, including:
  • connection module configured to acquire an incoming channel request of the first client, and establish a connection between the first client and the channel corresponding to the incoming channel request;
  • the determining module determines whether the number of clients currently accessed by the channel online reaches a predetermined access threshold
  • a live broadcast module configured to send the service information of the channel to the first client if the access threshold is not reached;
  • the queuing module is configured to add the first client to the queuing list for caching if the access threshold has been reached.
  • FIG. 10 is a schematic structural diagram of a channel access system between live broadcasts according to another embodiment.
  • the channel access system of the live room of the present invention may further include:
  • the exiting module is configured to obtain an exit channel request of the second client, disconnect the second client from the channel corresponding to the exit channel request, and query whether the second client is in the currently accessed client; if yes, Then, the second client is deleted from the currently online client, and if not, the second client is deleted from the queuing list, and the ordering of the queuing list is updated.
  • the determining module of the channel access system of the live room of the present invention may further include:
  • the queuing-free module is configured to determine, according to the identity information of the first client, whether the first client is a queuing-free client, and if yes, send the service information of the channel to the first client, and if not, execute the first client The steps to join the queued list for caching.
  • the channel access system of the live room of the present invention may further include:
  • a notification module configured to acquire, by the set period, the number of the first client in the queue list of the channel and the ranking information of each first client; according to the number of the first client and the ranking of each first client The information generates the ranking information of each first client; the ranking information of each first client is sent to each first client.
  • the notification module is configured to establish a sliding window sliding along a time axis, and establish a sliding time axis according to the time when each first client enters the channel; The sliding window slides along the sliding time axis, and sends the ranking information of each first client located in the sliding window to the corresponding first client.
  • the channel access system of the live room of the present invention may further include:
  • the checking module is configured to obtain the current number of clients of the channel, and compare the current number of clients with the sum of the number of currently accessed clients and the number of first clients in the queuing list, if the current number of clients is If the sum is inconsistent, the client that is not currently in the channel is searched, the found client is deleted, and the queuing list is updated.
  • the current number of clients includes the actual number of clients currently connected online and the queuing list. The actual number of first clients.
  • the check module further includes:
  • a triggering module configured to set a flag in the cache to be invalid within a set time; when the first client enters the channel, determine whether the flag exists in the cache; if yes, the current client of the channel is not executed The number of steps, if not present, performs the step of obtaining the current number of clients for the channel and then rewriting the flag in the cache.
  • the channel access system of the live broadcast room of the present invention has a one-to-one correspondence with the channel access method of the live broadcast of the present invention, and the technical features and beneficial effects of the embodiment of the channel access method between the live broadcasts are applicable to channel access between live broadcasts. In the embodiment of the system, it is hereby declared.
  • FIG. 11 is a schematic structural diagram of a channel access system in a live broadcast according to an embodiment.
  • the channel access system in the live broadcast includes: a receiver 1101, a processor 1102, and a transmitter 1103.
  • the processor 1102 is configured to acquire, by the receiver 1101, an incoming channel request of the first client, and establish a connection of the first client with a channel corresponding to the incoming channel request.
  • the processor 1102 is configured to determine whether the number of clients currently accessed by the channel is up to a predetermined schedule. Access threshold
  • the transmitter 1103 is configured to: if the access threshold is not reached, send the service information of the channel to the first client;
  • the processor 1102 is configured to add the first client to the queuing list to perform caching if the access threshold has been reached.
  • the processor 1102 is further configured to: obtain, by the receiver 1101, an exit channel request of the second client, disconnect the second client from the channel corresponding to the exit channel request, and query whether the second client is currently connected online. In the incoming client; if yes, the second client is deleted from the currently online client, and if not, the second client is deleted from the queue list, and the ordering of the queue list is updated.
  • the processor 1102 is further configured to determine, according to the identity information of the first client, whether the first client is a skip-free client;
  • the transmitter 1103 is further configured to: if yes, send the service information of the channel corresponding to the incoming channel request to the first client;
  • the processor 1102 is further configured to: if not, perform the step of determining whether the number of clients that the channel is currently online accessing reaches a predetermined access threshold.
  • the processor 1102 is further configured to acquire, according to a set period, the number of the first client in the queuing list of the channel and the ranking information of each first client; according to the number of the first client and the ranking information of each first client Generating ranking information of each first client;
  • the transmitter 1103 is further configured to separately send the ranking information of each first client to each first client.
  • the processor 1102 is configured to establish a sliding window sliding along a time axis, and establish a sliding time axis according to a time when each first client enters the channel; and sliding the sliding window along the sliding time axis;
  • the transmitter 1103 is configured to send the ranking information of each first client located in the sliding window to the corresponding first client.
  • the processor 1102 is further configured to obtain the current number of clients of the channel, and the current number of clients. Compared with the sum of the number of clients currently connected online and the number of the first client in the queuing list, if the current number of clients is inconsistent with the sum, the client that is not currently in the channel is found, and the deletion is found.
  • the client, and update the queue list, the current number of clients includes the actual number of clients currently online access and the actual number of first clients in the queue list.
  • the processor 1102 is further configured to: in the cache, set a flag bit that fails within a set time; when the first client enters the channel, determine whether a flag bit exists in the cache; if yes, perform the acquisition channel current The step of the number of clients, if not present, performs the step of obtaining the current number of clients of the channel, and then rewriting the flag in the cache.
  • the service information of the channel is sent to the first client; Enter the threshold and add the first client to the queue list for caching. Since the queue list is set, after the number of clients that are watching the live content reaches the access threshold of the channel, the client that enters the channel is cached through the queue list, thereby effectively avoiding the client exceeding the access threshold from being repeated multiple times.
  • Sending a request to enter the channel reduces the number of incoming channel requests that the server needs to process during the live broadcast, alleviating the processing pressure of the server between the live broadcasts, and effectively managing the client entering the channel.
  • the storage medium may be a magnetic disk, an optical disk, a read-only memory (ROM), or a random access memory (RAM).

Landscapes

  • Engineering & Computer Science (AREA)
  • Databases & Information Systems (AREA)
  • Human Computer Interaction (AREA)
  • Multimedia (AREA)
  • Signal Processing (AREA)
  • Two-Way Televisions, Distribution Of Moving Picture Or The Like (AREA)

Abstract

本发明提供一种直播间的频道访问方法和系统，其中方法包括：获取第一客户端的进入频道请求，建立第一客户端与该进入频道请求对应的频道的连接；判断该频道当前已在线接入的客户端数量是否达到预定的接入阈值；若未达到该接入阈值，将该频道的业务信息发送至第一客户端；若已达到该接入阈值，将第一客户端加入到排队链表中进行缓存。本发明的技术方案，从技术上限制了直播间的频道同时观看直播的客户端数，将超过接入阈值的客户端加入到排队链表中，避免了这些客户端重复多次发送进入频道请求，减少了进入频道请求的数量，缓解了直播间服务器的处理压力，提高了直播系统的稳定性，同时也可以实现对直播间频道的客户端访问进行有效管理。

Description

直播间的频道访问方法和系统
本申请要求于2014年10月20日提交中国专利局、申请号为201410559842.0、发明名称为“直播间的频道访问方法和系统”的中国专利申请的优先权，其全部内容通过引用结合在本申请中。
技术领域
本发明涉及互联网技术领域，特别是涉及一种直播间的频道访问方法和系统。
背景技术
随着网络技术的发展，客户端网络带宽的提高，在线直播取得了飞速发展，出现了大量的在线直播系统，如语音直播系统、视频点播直播系统和教育培训直播系统等。在线直播系统主要是由直播间构成，每个直播间内有多个不同类型的频道，如娱乐频道或音乐频道等。客户端登录直播间后，可以进入频道中收看主播的直播内容。一般登录这些在线直播系统的客户端数量很大，对网络带宽以及服务器处理能力提出了更高的要求。
以视频直播间举例，在线视频播放系统通过广播服务器将直播内容广播给客户端。随着客户端数量的增加，广播服务器的处理压力不断增加。为了缓解广播服务器的处理压力，广播服务器设置了在线观看人数的上限值，在同一时间广播服务器允许在线观看直播内容的客户端的数量不超过该上限值，以此来缓解广播服务器的处理压力。
但是，限制同一时间在线观看人数的上限值，当在线观看直播内容的客户端数量达到该上限值后，未接入的客户端可能会不断尝试进入频道，而重复多次发送进入频道请求给广播服务器，致使广播服务器需要处理的进入频道请求的数量倍增，使广播服务器的处理压力过大，甚至导致广播服务器崩溃，影响 直播系统的稳定性，而且直播服务器也无法对客户端进行有效管理。
发明内容
基于此，有必要针对上述技术问题，提供一种可以提高直播系统的稳定性、实现直播间对客户端进行有效管理的直播间的频道访问方法和系统。
一种直播间的频道访问方法，包括如下步骤：
获取第一客户端的进入频道请求，建立所述第一客户端与所述进入频道请求对应的频道的连接；
判断所述频道当前已在线接入的客户端数量是否达到预定的接入阈值；
若未达到所述接入阈值，将所述频道的业务信息发送至所述第一客户端；
若已达到所述接入阈值，将所述第一客户端加入到排队链表中进行缓存。
一种直播间的频道访问系统，包括：
连接模块，用于获取第一客户端的进入频道请求，建立所述第一客户端与所述进入频道请求对应的频道的连接；
判断模块，用判断所述频道当前已在线接入的客户端数量是否达到预定的接入阈值；
直播模块，用于若未达到所述接入阈值，将所述频道的业务信息发送至所述第一客户端；
排队模块，用于若已达到所述接入阈值，将所述第一客户端加入到排队链表中进行缓存。
一种直播间的频道访问系统，包括：接收器、处理器和发送器；
所述处理器，用于通过所述接收器获取第一客户端的进入频道请求，建立所述第一客户端与所述进入频道请求对应的频道的连接；
所述处理器，用于判断所述频道当前已在线接入的客户端数量是否达到预定的接入阈值；
所述发送器，用于若未达到所述接入阈值，将所述频道的业务信息发送至 所述第一客户端；
所述处理器，用于若已达到所述接入阈值，将所述第一客户端加入到排队链表中进行缓存。
上述实施例的直播间的频道访问方法，通过接收和响应客户端进入频道请求，并建立与该进入频道请求对应的频道的连接；利用预定的接入阈值判断该频道当前已在线接入的客户端数量，在未达到接入阈值情况下，控制客户端观看直播内容；如果达到预定的接入阈值，将该客户端加入到排队链表中进行缓存。从技术上限制了直播间的频道同时观看直播的客户端数，将超过接入阈值的客户端加入到排队链表中进行缓存，有效避免了超过接入阈值的客户端重复多次发送进入频道请求，减少了直播间服务器需要处理的进入频道请求的数量，缓解了直播间服务器的处理压力，提高了直播系统的稳定性，同时也可以实现对直播间频道的客户端访问进行有效管理。
附图说明
图1为一个实施例的应用系统结构示意图；
图2为一个实施例的直播间的频道访问方法的流程图；
图3为一个实施例的客户端进入频道的处理流程图；
图4为一个应用UI示意图；
图5a为利用滑动窗口通知客户端的第一示意图；
图5b为利用滑动窗口通知客户端的第二示意图；
图6为一个应用实例的服务端的系统结构图；
图7为一个实例的客户端进入频道时序图；
图8为一个实例的频道客户端数据修复时序图；
图9为一个实施例的直播间的频道访问系统的结构示意图；
图10为另一个实施例的直播间的频道访问系统的结构示意图；
图11为一个实施例的直播间的频道访问系统的结构示意图。
具体实施方式
为了使本发明的目的、技术方案及优点更加清楚明白，以下结合附图及实施例，对本发明进行进一步详细说明。应当理解，此处所描述的具体实施例仅仅用以解释本发明，并不用于限定本发明。
本发明实施例所提供的直播间的频道访问方法和系统，适用于语音直播系统、视频点播直播系统、教育培训直播系统等，其可应用于如图1所示的系统中，图1为一个实施例的应用系统结构示意图，请参考图1所示，该系统包括终端设备02和服务端04。
终端设备02可以是诸如PC(Personal Computer，个人电脑)、Smart phone(智能手机)、Tablet PC(Tablet Personal Computer，平板电脑)或PDA(Personal Digital Assistant，个人数字助理)等设备，但并不局限于此。
终端设备02可以通过有线或无线等方式接入公共网络，并与服务端04进行通信，公共网络可以为GPRS(General Packet Radio Service，通用无线分组服务)或Internet(互联网)等。终端设备02上可运行客户端，该客户端可以访问直播间频道。该客户端通过公共网络访问服务端04，并进入到对应的频道请求所需的直播内容。客户端将服务端04返回的直播内容呈现给用户。
所述终端设备02可以是指一个设备也可以是指多个设备；服务端04包括实现各种后台功能的服务器设备，可以是一个服务器设备也可以是多个服务器设备构成的服务器群。
参考图2所示，图2为一个实施例的直播间的频道访问方法的流程图，包括如下步骤：
步骤S10，获取第一客户端的进入频道请求，建立第一客户端与该进入频道请求对应的频道之间的连接。
其中，该进入频道请求中至少携带第一客户端的标识和所要进入的频道的频道标识。该进入频道请求对应的频道即为该进入频道请求携带的频道标识对应的频道。
在此步骤中，对于进入直播间频道的第一客户端，直播间服务器在接收其进入频道请求后，直播间服务器响应该进入频道请求，根据该进入频道请求中携带的第一客户端的标识和频道标识建立第一客户端与该频道标识对应的频道之间的通信连接。
步骤S20，判断该频道当前已在线接入的客户端数量是否达到预定的接入阈值，如果是，则执行步骤S40，如果否，则执行步骤S30。
在此步骤中，预设了在直播间的频道中在线接入的客户端的接入阈值，该接入阈值用于控制观看直播内容的客户端数量。在新的客户端进入频道后，判断当前已在线接入的客户端数量是否达到预定的接入阈值，从而确定该新的客户端是否需要进行排队。
步骤S30，若未达到该接入阈值，则将该频道的业务信息发送至第一客户端。
在此步骤中，在该频道当前已在线接入的客户端数量未达到预定的接入阈值的情况下，获取该频道的业务信息，将该频道的业务信息发送至第一客户端，使第一客户端对应的用户可以观看该频道的直播内容。
步骤S40，若已达到该接入阈值，则将第一客户端加入到排队链表中进行缓存。
在此步骤中，在该频道当前已在线接入的客户端数量达到预定的接入阈值的情况下，将进入该频道的第一客户端加入到该频道对应的排队链表中进行缓存，从而实现对进入该频道的客户端的有效控制和管理。
对于排队链表，可以以客户端的ID(Identity，身份标识)为排队链表的数据元素，第一客户端的ID可以为第一客户端对应的用户的登录账号。在该排队链表中以第一客户端进入该频道的时间来排列数据元素的逻辑顺序。该排队链表中的客户端是暂时不允许观看该频道的直播内容的。
上述步骤提供的直播间的频道访问方法，通过接收和响应第一客户端的进入频道请求，并建立第一客户端与该进入频道请求对应的频道之间的连接；利用预定的接入阈值判断该频道当前已在线接入的客户端数量是否达到该接入 阈值，在未达到接入阈值的情况下，控制第一客户端观看直播内容；如果达到预定的接入阈值，则将第一客户端加入到该频道对应的排队链表中进行缓存。从技术上限制了直播间的频道中同时观看直播内容的客户端数量，避免了同一时间内由于观看直播内容的客户端数量过多而对频道的直播间服务器造成过大处理压力，杜绝了直播系统崩溃的情况发生，提高了直播系统的稳定性，同时也可以实现对访问直播间的频道的客户端进行有效管理。
对于上述频道对应的排队链表，可以选择使用Redis(重申)来缓存直播间的频道对应的排队链表，实现对数据的高速缓存。Redis是一个key-value(分布式存储系统)存储系统，它支持存储多种value(值)类型，包括string(字符串)、list(链表)、set(集合)、zset(sorted set，有序集合)和hash(哈希类型)等。这些数据类型都支持push/pop(推进/弹出)、add/remove(添加/移除)、取交集、取并集和取差集等丰富的操作。而且Redis也支持各种不同方式的排序。
对于上述频道对应的排队链表，优选的，可以采用跳跃表(skiplist)这种链表，来实现高效排队算法。跳跃表是链表的一种，跳跃表主要由以下几个部分构成：
(1)head(表头)：负责维护跳跃表的节点指针；
(2)node(节点)：实际保存元素值，每个节点有一层或多层；
(3)level(层)：保存着指向该层下一个节点的节点指针；
(4)tail(表尾)：全部由null(空)组成。
跳跃表是在链表的基础上增加了跳转功能，可以轻松地实现Log N次计算。跳跃表是一种随机化的数据结构，基于并联的链表，可以在对数期望时间内简单直观地完成查找、删除或插入等操作。采用跳跃表，遍历过程总是从高层开始，然后随着元素值范围的缩小，直至降低到低层。
在一个实施例中，对于本发明的直播间的频道访问方法，还可以包括如下步骤：
获取第二客户端的退出频道请求，断开第二客户端与该退出频道请求对应 的频道的连接；
查询第二客户端是否在当前在线已接入的客户端中；
若是，则将其从当前在线已接入的客户端中删除，若否，则从排队链表中删除第二客户端，并更新该排队链表的排序。
其中，该退出频道请求中至少携带第二客户端的标识和所要退出的频道的频道标识。该退出频道请求对应的频道即为该退出频道请求携带的频道标识对应的频道。在上述步骤中，对于退出频道的第二客户端，可以首先根据该退出频道请求携带的第二客户端的标识查找第二客户端是否在当前已在线接入的客户端中，如果在，则将第二客户端从当前已在线接入的客户端中删除，如果不在，则说明第二客户端在排队链表中，从排队链表中找到第二客户端的标识并将其删除，然后更新该排队链表的排序；如果在当前已在线接入的客户端中删除第二客户端，则可以移入排队链表中的一个客户端作为在线接入的客户端，这样排队链表中的排序发生了变化，更新该排队链表中的数据元素的逻辑顺序；如果是在排队链表中删除第二客户端，则将其对应数据元素删除后，排队链表中的排序发生了变化，更新排队链表中数据元素的逻辑顺序。
在一个实施例中，本发明的直播间的频道访问方法，在判断该频道当前已在线接入的客户端数量是否达到预定的接入阈值的步骤前，还可以包括如下步骤：
根据第一客户端的身份信息判断第一客户端是否为免排队客户端，若是，将该频道的业务信息发送至第一客户端，若否，则判断该频道当前已在线接入的客户端数量是否达到预定的接入阈值。
上述步骤的方案，实现对客户端进入频道后更加有效的管理，通过基于管理目的的技术手段来给客户端设置对应的服务等级，既可以实现服务的增值，同时也可以作为直播间业绩的考核指标。
参考图3，图3为一个实施例的客户端进入频道的处理流程图；主要包括如下步骤：
s301，读取进入频道的第一客户端；
s302，读取该频道的在线客户端的上限值；
s303，判断第一客户端是否为免排队客户端，若是，执行s304，否则执行s305；
s304，将第一客户端放入当前已在线接入的客户端集合中，并通知其观看直播；
s305，读取该频道当前已在线接入的客户端数量；
s306，判断当前已在线接入的客户端数量是否超过上限值，若是，执行s308，否则执行s307；
s307，将第一客户端放入当前已在线接入的客户端集合中，并通知其观看直播；
s308，读取该频道当前的排队链表中的客户端数量；
s309，将第一客户端放入排队链表中进行缓存；
s310，根据当前的排队链表中的客户端数量生成第一客户端排队的排位信息。
上述客户端进入频道的流程，通过给客户端设置对应的服务等级，设置了免排队客户端，免排队客户端可以直接在线接入频道来观看直播内容，促使客户端选择更高的服务等级，这样可以实现服务的增值，而进入频道的客户端的服务等级也可以作为该频道对应的直播间业绩的考核指标。
参考图4所示，图4为一个应用UI(User Interface，用户界面)示意图，从图中可以看出，贵宾客户端(免排队客户端)可以通过相应的通道快速进入频道，而普通客户端则需要进行排队等候。
在一个实施例中，本发明的直播间的频道访问方法，还可以包括如下步骤：
以设定的周期获取该频道的排队链表中的第一客户端的数量和各个第一客户端的排序信息；根据第一客户端的数量和各个第一客户端的排序信息生成各个第一客户端的排位信息；将各个第一客户端的排位信息分别发送至各个第一客户端。
上述步骤的方案，在第一客户端进入到频道后，如果第一客户端是需要排 队的客户端，则直播间服务器可以实时地通知第一客户端当前自己排队的信息，如排位信息，即排在第一客户端前的客户端数量，从而可以提高第一客户端的应用体验。
在一个实施例中，考虑到直播间的频道的排队客户端数量巨大，实时地通知每一个排队客户端，需要耗费较大的计算资源，为此，本发明提供一种改进的通知方案，定期对所有排队客户端通知其排队状态。
作为一种实施方式，对于将各个第一客户端的排位信息分别发送至各个第一客户端的步骤，可以包括如下：
建立沿时间轴滑动的滑动窗口，并根据各个第一客户端进入频道的时间建立滑动时间轴；将该滑动窗口沿该滑动时间轴进行滑动，并将位于该滑动窗口内的每个第一客户端的排位信息发送至对应的第一客户端。
例如，如图5a所示，第一客户端ID1、ID2和ID3进入频道的时间分别为t0、t1和t2，根据第一客户端1、2和3进入频道的时间分别为t0、t1和t2建立如图5a所示的滑动时间轴，将该滑动窗口沿滑动时间轴进行滑动，第一次滑动时第一客户端ID1和ID2位于该滑动窗口内，将第一客户端ID1和ID2的排位信息分别发送给第一客户端ID1和ID2，第二次滑动时，第一客户端ID3位于该滑动窗口内，将第一客户端ID3的排位信息发送给第一客户端ID3。
参考图5b所示，图5b为利用滑动窗口通知客户端的示意图，滑动窗口沿着滑动时间轴在排队客户端上移动，每次滑动时只会通知位于滑动窗口内的客户端，一般情况下，可以设置滑动窗口在预设时间滑动完整个滑动时间轴，以每隔该预设时间通知一次所有的排队客户端，预设时间可以为10秒或20秒等。
优选地，滑动窗口沿滑动时间轴在排队客户端上进行滑动时，每次进入该滑动窗口内的第一客户端的数量相同，且连续两次滑动时进入该滑动窗口内的第一客户端中不存在重复的第一客户端。例如，假设每次进入该滑动窗口内的第一客户端的数量均为5个，则连续两次滑动该滑动窗口时，这两次进入该滑动窗口内的第一客户端一共有10个，这10个第一客户端中不存在相同的第一客户端。
上述实施例的方案，可以提高服务器资源的利用效率，减少排队客户端数量巨大时通知客户端处理对直播间服务器运行的影响，减轻直播间服务器的负担。
前述实施例中，提供了进入频道后通知客户端当前自己排队的信息的方案，但在实际应用中，通知消息是需要在多个节点中进行转发，如果发生单点故障，则可能会重发了一些通知消息。例如，一客户端已经进入频道，但是由于该客户端发生故障重复向服务器发送了进入频道请求。据此，为了避免由于重发消息而导致将已经进入频道的客户端作为新请求的客户端来进行处理，可以进一步设置重复消息检查机制。
作为一种实施方式，该重复消息检查机制具体可以包括如下步骤：
在建立第一客户端与直播间的频道之间的连接后，判断当前在线已接入的客户端中或排队链表中是否存在该进入的第一客户端；若存在，则判定该进入的第一客户端为已存在客户端，并将上一次发送的排位信息发送至第一客户端；若不存在，则判定该进入的客户端为新进入的客户端，将最新生成的排位信息发送至第一客户端。
上述实施例的方案，通过进一步设置重复消息检查机制，可以有效减少错误消息影响，提高发送通知消息的效率。
如上述实施例所述，在实际应用中，由于整个直播间的消息往往需要在多个节点中进行转发，在这种情况下，往往难以保证转发消息的完整性，据此，作为本发明的直播间的频道访问方法的实施例，还可以提供数据完整性的检查机制，通过这些检查机制来相互验证数据是否完整。
在一个实施例中，本发明的直播间的频道访问方法，还可以包括针对于丢失客户端退出频道消息的检查方案，该检查方案主要是针对于一些已经退出频道的客户端，由于退出频道消息在转发过程中丢失，从而导致退出频道的客户端仍然缓存在当前已在线接入的客户端中或排队链表中，而仍占用服务资源的情况。由此而提出的检查机制，包括如下步骤：
获取该频道当前客户端数量，将当前客户端数量与当前已在线接入的客户 端数量和排队链表中第一客户端的数量的和值进行对比，若当前客户端数量与该和值不一致，则查找出当前已经不在该频道的客户端，删除查找出的客户端，并更新该排队链表。
一般情况下，接入频道的客户端对应的服务节点都会维护一个客户端列表，通过这个客户端列表可以获取当前频道的数据准确性是较高的客户端数量，由于该客户端列表是由服务节点来维护的，所以通过该客户端列表获取的该客户端数量的准确性很高，可以认为该客户端数量与当前进入该频道的所有客户端的实际数量相等。据此，当频道的主播开播时，主播定期(如每分钟)记录当前频道的客户端数量，并上传一个心跳消息到直播间服务器，该心跳消息携带该客户端数量，记录当前频道的客户端数量，直播间服务器收到此客户端数量后，将此客户端数量与所有已建立连接的客户端数量(当前已在线接入的客户端数量和排队链表中第一客户端的数量的和值)进行比对，如果不一致，直播间服务器触发全量核查一次，检查已建立连接的客户端中是否存在已经不在该频道的客户端。
另外，对于上述检查机制，还可以在主播开播后，定时检查频道当前的客户端数量并触发上述检查过程。
作为一种优选实施方式，对于上述实施例的检查机制，为了更好的进行流量控制，可以以定时方式进行触发该检查过程，具体如下：
在缓存中设置一个在设定时间内失效的标志位；
在第一客户端进入该频道时，判断该缓存中是否存在该标志位；
若存在，则不执行获取该频道当前客户端数量的步骤，
若不存在，则执行获取该频道当前客户端数量的步骤，然后在缓存中重新写入该标志位。
上述定时的触发策略，可以通过在缓存中设置一个在设定时间(如1分钟)内失效的标志位来实现的，当有客户端进入该频道时，如果缓存中没有这个标志位，则触发一次检查过程，然后在缓存中写入这个标志位。当有其他客户端进入该频道时，发现缓存中存在这个标志位，就不触发检查过程。通过上述技 术手段，通过最短在设定时间内触发一次的周期，这样在没有客户端进出频道时，就不会触发检查过程，避免重复频繁触发检查过程，并且实现了更好的流量控制功能。
进一步地，在清理不存在的客户端后，对当前客户端队列情况进行修复，该修复流程如下：
根据删除客户端后当前已在线接入的客户端数量与接入阈值之间的差距，将排队链表中前部的客户端移动到当前已在线接入的客户端集合中，并向移动的这些客户端发送进入频道的通知消息；在移动客户端操作后，以排队链表中剩余的客户端数量来更新排队链表的排序并生成剩余的各个客户端的排位信息，将各个客户端的排位信息分别发送至各个客户端。
上述实施例的处理方案，在进行一次删除操作后，及时对当前已在线接入的客户端和排队链表进行更新，并能够及时将客户端的状态通知客户端，保证了当前缓存数据的实时性。
上述各个实施例，阐述了本发明的直播间的频道访问方法的多种实施方式，这些实施例的方案可以在如图1中所示的服务端04上实现，针对于服务端04的应用以及当前直播间的系统架构，下面结合附图阐述若干具体应用实例。
参考图6所示，图6为一个应用实例的服务端的系统结构图，主要包括服务节点(Service)、会话进程(Session)、直播间服务器、缓存代理、查询进程(Query)。
服务节点维护频道内的所有客户端上的客户端列表，并与客户端进行通信交互。每个频道对应一个会话进程，直播间服务器对整个直播间的所有频道的进入频道请求或退出频道请求进行处理。缓存代理通过链表形式缓存客户端数据。
客户端进入频道的过程如下：
当一个客户端(用户登录的客户端)进入频道时，该客户端发送一个会话通知至服务节点，该会话通知至少携带频道的频道标识和该客户端的标识。服 务节点根据该频道标识建立该频道与该客户端之间的连接。服务节点产生一个消息通知至会话进程，该消息通知中携带该客户端的标识，会话进程将该消息通知转发给直播间服务器。直播间服务器将该客户端的标识写入到代理缓存中。
针对于进入频道流程，下面结合附图阐述一个应用实例，参考图7所示，图7为一个实例的客户端进入频道时序图。
如图7中流程所示，客户端进入频道的流程为：由该客户端发送进频道消息给服务节点，然后该客户端登录到服务节点。服务节点转发该进频道消息至会话进程，会话进程转发该进频道消息至直播间服务器，直播间服务器在缓存中插入该客户端(即将该客户端插入当前已在线接入的客户端集合或排队链表中)，并返回相应的通知消息(观看直播通知或排队等候通知)至会话进程，会话进程转发该通知消息至服务节点，服务节点将该通知消息发送至该客户端。
客户端退出频道过程如下：
当一个客户端退出频道时，服务节点产生一个消息通知至直播间服务器，该消息通知携带该客户端的标识和所要退出的频道的频道标识。直播间服务器收到该消息通知后，首先根据该客户端的标识查看当前已在线接入的客户端集合，如果该客户端的标识在当前已在线接入的客户端集合中，则直接删除该客户端的标识，如果该客户端的标识不在当前已在线接入的客户端集合中，则从排队链表中找到该客户端的标识并将其删除，然后更新排队链表的排序，修改排队客户端的排序，并获取更新排队链表后排在该客户端后面的客户端的排位信息，并将获取的排位信息通知给排在该客户端后面的客户端。
丢失客户端退出频道消息的检查方案如下：
在主播开播时，从服务节点维护的客户端列表中获取到当前频道已建立的所有连接的客户端数量，并将该客户端数量上传至直播间服务器。直播间服务器将该客户端数量与记录的所有客户端数量进行对比，对频道进行全量核查一次，判断当前已建立连接的客户端中是否还存在已经退出频道的客户端。
具体方案可以在缓存中设置一个设定时间(如1分钟)内失效的标志位，当有客户端进入频道时，如果发现缓存中没有这个标志位，则触发检查流程，然后在缓存中写入这个标志位。当有其他客户端进入频道时，发现缓存中存在这个标志位，就不触发检查流程。
异常数据的修复过程如下：
当触发检查流程时，直播间服务器发一个查询通知消息至查询进程，查询进程从服务节点维护的客户端列表中查询客户端数量，并将该客户端数量发送至直播间服务器。直播间服务器将该客户端数量与记录的客户端数量进行比较，找出异常客户端并对异常客户端进行清理，然后将更新后的最新的排位信息通知到客户端。
针对于频道客户端数据修复流程，下面结合附图阐述一个应用实例，参考图8所示，图8为一个实例的频道客户端数据修复时序图。
如图8中流程所示，主播开播后，由主播的客户端定期发送消息1至服务节点，服务节点转发消息2至会话进程，会话进程转发消息3至直播间服务器，直播间服务器检测已建立连接的客户端数量，发送查询客户端数量消息至客户端查询进程，客户端查询进程向服务节点发送查询客户端数量消息。服务节点接收查询客户端数量消息后，返回包含客户端数量的消息4至客户端查询进程，客户端查询进程转发该客户端数量至直播间服务器。直播间服务器将已建立连接的客户端数量与查询的客户端数量进行对比，如果存在不一致，则向客户端查询进程发送查询客户端列表消息，客户端查询进程转发查询列表消息至服务节点，服务节点将自己维护的客户端列表返回至客户端查询进程，客户端查询进程将包含客户端列表的消息5发送至直播间服务器。直播间服务器收到客户端列表后，根据客户端列表记录的当前频道的客户端，将当前已在线接入的客户端集合中已经退出频道的客户端删除，然后更新排队链表中排位信息，然后将更新后的排位信息通过通知消息依次由会话进程、客户端查询进程、服务节点转发至客户端，通知客户端最新的排队顺序。
参考图9所示，图9为一个实施例的直播间的频道访问系统的结构示意图，包括：
连接模块，用于获取第一客户端的进入频道请求，建立第一客户端与该进入频道请求对应的频道的连接；
判断模块，用判断该频道当前已在线接入的客户端数量是否达到预定的接入阈值；
直播模块，用于若未达到接入阈值，将该频道的业务信息发送至第一客户端；
排队模块，用于若已达到接入阈值，将第一客户端加入到排队链表中进行缓存。
为了更加清晰本发明的直播间的频道访问系统，下面结合图10阐述若干实施例。
参考图10所示，图10为另一个实施例的直播间的频道访问系统的结构示意图。
在一个实施例中，本发明的直播间的频道访问系统还可以包括：
退出模块，用于获取第二客户端的退出频道请求，断开第二客户端与该退出频道请求对应的频道的连接；查询第二客户端是否在当前已在线接入的客户端中；若是，则将第二客户端从当前已在线接入的客户端中删除，若否，则从排队链表中删除第二客户端，并更新该排队链表的排序。
在一个实施例中，本发明的直播间的频道访问系统的判断模块前还可以包括：
免排队模块，用于根据第一客户端的身份信息判断第一客户端是否为免排队客户端，若是，将该频道的业务信息发送至第一客户端，若否，则执行将第一客户端加入到排队链表中进行缓存的步骤。
在一个实施例中，本发明的直播间的频道访问系统还可以包括：
通知模块，用于以设定的周期获取该频道的排队链表中的第一客户端数量和各个第一客户端的排序信息；根据第一客户端数量和各个第一客户端的排序 信息生成各个第一客户端的排位信息；将各个第一客户端的排位信息分别发送至各个第一客户端。
在一个实施例中，本发明的直播间的频道访问系统，所述通知模块，用于建立沿时间轴滑动的滑动窗口，并根据各个第一客户端进入该频道的时间建立滑动时间轴；将该滑动窗口沿该滑动时间轴进行滑动，并将位于该滑动窗口内的每个第一客户端的排位信息发送至对应的第一客户端。
在一个实施例中，本发明的直播间的频道访问系统还可以包括：
检查模块，用于获取该频道当前客户端数量，将当前客户端数量与当前已在线接入的客户端数量和排队链表中第一客户端额数量的和值进行对比，若当前客户端数量与该和值不一致，则查找出当前已经不在该频道的客户端，删除查找出的客户端，并更新该排队链表，该当前客户端数量包括当前已在线接入的客户端的实际数量和排队链表中第一客户端的实际数量。
在一个实施例中，本发明的直播间的频道访问系统，所述检查模块前还包括：
触发模块，用于在缓存中设置一个在设定时间内失效的标志位；在第一客户端进入该频道时，判断缓存中是否存在该标志位；若存在，则不执行获取该频道当前客户端数量的步骤，若不存在，则执行获取该频道当前客户端数量的步骤，然后在缓存中重新写入该标志位。
本发明的直播间的频道访问系统与本发明的直播间的频道访问方法一一对应，在上述直播间的频道访问方法的实施例阐述的技术特征及其有益效果均适用于直播间的频道访问系统的实施例中，特此声明。
参考图11所示，图11为一个实施例的直播间的频道访问系统的结构示意图，该直播间的频道访问系统包括：接收器1101、处理器1102和发送器1103；
处理器1102，用于通过接收器1101获取第一客户端的进入频道请求，建立第一客户端与该进入频道请求对应的频道的连接；
处理器1102，用于判断该频道当前已在线接入的客户端数量是否达到预定 的接入阈值；
发送器1103，用于若未达到接入阈值，将该频道的业务信息发送至第一客户端；
处理器1102，用于若已达到接入阈值，将第一客户端加入到排队链表中进行缓存。
其中，处理器1102，还用于通过接收器1101获取第二客户端的退出频道请求，断开第二客户端与该退出频道请求对应的频道的连接；查询第二客户端是否在当前已在线接入的客户端中；若是，则将第二客户端从当前已在线接入的客户端中删除，若否，则从排队链表中删除第二客户端，且更新排队链表的排序。
其中，处理器1102，还用于根据第一客户端的身份信息判断第一客户端是否为免排队客户端；
发送器1103，还用于若是，则将该进入频道请求对应的频道的业务信息发送至第一客户端；
处理器1102，还用于若否，则执行判断该频道当前已在线接入的客户端数量是否达到预定的接入阈值的步骤。
其中，处理器1102，还用于以设定的周期获取该频道的排队链表中的第一客户端数量和各个第一客户端的排序信息；根据第一客户端数量和各个第一客户端的排序信息生成各个第一客户端的排位信息；
发送器1103，还用于将各个第一客户端的排位信息分别发送至各个第一客户端。
其中，处理器1102，用于建立沿时间轴滑动的滑动窗口，并根据各个第一客户端进入频道的时间建立滑动时间轴；将该滑动窗口沿该滑动时间轴进行滑动；
发送器1103，用于将位于滑动窗口内的每个第一客户端的排位信息发送至对应的第一客户端。
其中，处理器1102，还用于获取频道当前客户端数量，将当前客户端数量 与当前已在线接入的客户端数量和排队链表中第一客户端的数量的和值进行对比，若当前客户端数量与该和值不一致，则查找出当前已经不在频道的客户端，删除查找出的客户端，并更新排队链表，该当前客户端数量包括当前已在线接入的客户端的实际数量和排队链表中第一客户端的实际数量。
其中，处理器1102，还用于在缓存中设置一个在设定时间内失效的标志位；在第一客户端进入频道时，判断缓存中是否存在标志位；若存在，则不执行获取频道当前客户端数量的步骤，若不存在，则执行获取频道当前客户端数量的步骤，然后在缓存中重新写入标志位。
在本发明实施例中，判断频道当前已在线接入的客户端数量是否达到预定的接入阈值；若未达到接入阈值，将该频道的业务信息发送至第一客户端；若已达到接入阈值，将第一客户端加入到排队链表中进行缓存。由于设置了排队链表，当正在观看直播内容的客户端的数量达到该频道的接入阈值之后，通过该排队链表来缓存进入该频道的客户端，有效避免了超过接入阈值的客户端重复多次发送进入频道请求，减少了直播间服务器需要处理的进入频道请求的数量，缓解了直播间服务器的处理压力，实现了对进入频道的客户端的有效管理。
本领域普通技术人员可以理解实现上述实施例方法中的全部或部分流程，是可以通过计算机程序来指令相关的硬件来完成，所述的程序可存储于一计算机可读取存储介质中，该程序在执行时，可包括如上述各方法的实施例的流程。其中，所述的存储介质可为磁碟、光盘、只读存储记忆体(Read-Only Memory，ROM)或随机存储记忆体(Random Access Memory，RAM)等。
以上所述实施例仅表达了本发明的几种实施方式，其描述较为具体和详细，但并不能因此而理解为对本发明专利范围的限制。应当指出的是，对于本领域的普通技术人员来说，在不脱离本发明构思的前提下，还可以做出若干变形和改进，这些都属于本发明的保护范围。因此，本发明专利的保护范围应以所附权利要求为准。

Claims (21)

  1. 一种直播间的频道访问方法，其特征在于，包括如下步骤：
    获取第一客户端的进入频道请求，建立所述第一客户端与所述进入频道请求对应的频道的连接；
    判断所述频道当前已在线接入的客户端数量是否达到预定的接入阈值；
    若未达到所述接入阈值，将所述频道的业务信息发送至所述第一客户端；
    若已达到所述接入阈值，将所述第一客户端加入到排队链表中进行缓存。
  2. 根据权利要求1所述的直播间的频道访问方法，其特征在于，还包括：
    获取第二客户端的退出频道请求，断开所述第二客户端与所述退出频道请求对应的频道的连接；
    查询所述第二客户端是否在当前已在线接入的客户端中；
    若是，则将所述第二客户端从所述当前已在线接入的客户端中删除，若否，则从所述排队链表中删除所述第二客户端，且更新所述排队链表的排序。
  3. 根据权利要求1或2所述的直播间的频道访问方法，其特征在于，在判断所述频道当前已在线接入的客户端数量是否达到预定的接入阈值的步骤前，还包括：
    根据所述第一客户端的身份信息判断所述第一客户端是否为免排队客户端，若是，则将所述进入频道请求对应的频道的业务信息发送至所述第一客户端，若否，则执行判断所述频道当前已在线接入的客户端数量是否达到预定的接入阈值的步骤。
  4. 根据权利要求1-3任一项权利要求所述的直播间的频道访问方法，其特征在于，还包括：
    以设定的周期获取所述频道的排队链表中的第一客户端数量和各个第一客户端的排序信息；根据所述第一客户端数量和所述各个第一客户端的排序信息生成所述各个第一客户端的排位信息；将所述各个第一客户端的排位信息分别发送至所述各个第一客户端。
  5. 根据权利要求4所述的直播间的频道访问方法，其特征在于，所述将所述各个第一客户端的排位信息分别发送至所述各个第一客户端的步骤包括：
    建立沿时间轴滑动的滑动窗口，并根据所述各个第一客户端进入所述频道的时间建立滑动时间轴；将所述滑动窗口沿所述滑动时间轴进行滑动，并将位于所述滑动窗口内的每个第一客户端的排位信息发送至对应的第一客户端。
  6. 根据权利要求1-5任一项权利要求所述的直播间的频道访问方法，其特征在于，还包括：
    获取所述频道当前客户端数量，将所述当前客户端数量与所述当前已在线接入的客户端数量和所述排队链表中第一客户端的数量的和值进行对比，若所述当前客户端数量与所述和值不一致，则查找出当前已经不在所述频道的客户端，删除所述查找出的客户端，并更新所述排队链表，所述当前客户端数量包括所述当前已在线接入的客户端的实际数量和所述排队链表中第一客户端的实际数量。
  7. 根据权利要求6所述的直播间的频道访问方法，其特征在于，在所述获取所述频道当前客户端数量的步骤前，还包括：
    在缓存中设置一个在设定时间内失效的标志位；
    在所述第一客户端进入所述频道时，判断所述缓存中是否存在所述标志位；
    若存在，则不执行获取所述频道当前客户端数量的步骤，
    若不存在，则执行获取所述频道当前客户端数量的步骤，然后在缓存中重新写入所述标志位。
  8. 一种直播间的频道访问系统，其特征在于，包括：
    连接模块，用于获取第一客户端的进入频道请求，建立所述第一客户端与所述进入频道请求对应的频道的连接；
    判断模块，用判断所述频道当前已在线接入的客户端数量是否达到预定的接入阈值；
    直播模块，用于若未达到所述接入阈值，将所述频道的业务信息发送至所述第一客户端；
    排队模块，用于若已达到所述接入阈值，将所述第一客户端加入到排队链表中进行缓存。
  9. 根据权利要求8所述的直播间的频道访问系统，其特征在于，还包括：
    退出模块，用于获取第二客户端的退出频道请求，断开所述第二客户端与所述退出频道请求对应的频道的连接；查询所述第二客户端是否在当前已在线接入的客户端中；若是，则将所述第二客户端从所述当前已在线接入的客户端中删除，若否，则从所述排队链表中删除所述第二客户端，且更新所述排队链表的排序。
  10. 根据权利要求8或9所述的直播间的频道访问系统，其特征在于，所述判断模块前还包括：
    免排队模块，用于根据所述第一客户端的身份信息判断所述第一客户端是否为免排队客户端，若是，则将所述频道的业务信息发送至所述第一客户端，若否，则执行判断所述频道当前已在线接入的客户端数量是否达到预定的接入阈值的步骤。
  11. 根据权利要求8-10任一项权利要求所述的直播间的频道访问系统，其特征在于，还包括：
    通知模块，用于以设定的周期获取所述频道的排队链表中的第一客户端数量和各个第一客户端的排序信息；根据所述第一客户端数量和所述各个第一客户端的排序信息生成所述各个第一客户端的排位信息；将所述各个第一客户端的排位信息分别发送至所述各个第一客户端。
  12. 根据权利要求11所述的直播间的频道访问系统，其特征在于，所述通知模块，用于建立沿时间轴滑动的滑动窗口，并根据所述各个第一客户端进入所述频道的时间建立滑动时间轴；将所述滑动窗口沿所述滑动时间轴进行滑动，并将位于所述滑动窗口内的每个第一客户端的排位信息发送至对应的第一客户端。
  13. 根据权利要求8-12任一项权利要求所述的直播间的频道访问系统，其特征在于，还包括：
    检查模块，用于获取所述频道当前客户端数量，将所述当前客户端数量与所述当前已在线接入的客户端数量和所述排队链表中第一客户端的数量的和值进行对比，若所述当前客户端数量与所述和值不一致，则查找出当前已经不在所述频道的客户端，删除所述查找出的客户端，并更新所述排队链表，所述当前客户端数量包括所述当前已在线接入的客户端的实际数量和所述排队链表中第一客户端的实际数量。
  14. 根据权利要求13所述的直播间的频道访问系统，其特征在于，所述检查模块前还包括：
    触发模块，用于在缓存中设置一个在设定时间内失效的标志位；在所述第一客户端进入所述频道时，判断所述缓存中是否存在所述标志位；若存在，则不执行获取所述频道当前客户端数量的步骤，若不存在，则执行获取所述频道当前客户端数量的步骤，然后在所述缓存中重新写入所述标志位。
  15. 一种直播间的频道访问系统，其特征在于，包括：接收器、处理器和发送器；
    所述处理器，用于通过所述接收器获取第一客户端的进入频道请求，建立所述第一客户端与所述进入频道请求对应的频道的连接；
    所述处理器，用于判断所述频道当前已在线接入的客户端数量是否达到预定的接入阈值；
    所述发送器，用于若未达到所述接入阈值，将所述频道的业务信息发送至所述第一客户端；
    所述处理器，用于若已达到所述接入阈值，将所述第一客户端加入到排队链表中进行缓存。
  16. 根据权利要求15所述的直播间的频道访问系统，其特征在于，所述处理器，还用于通过所述接收器获取第二客户端的退出频道请求，断开所述第二客户端与所述退出频道请求对应的频道的连接；查询所述第二客户端是否在当前已在线接入的客户端中；若是，则将所述第二客户端从所述当前已在线接入的客户端中删除，若否，则从所述排队链表中删除所述第二客户端，且更新所 述排队链表的排序。
  17. 根据权利要求15或16所述的直播间的频道访问系统，其特征在于，
    所述处理器，还用于根据所述第一客户端的身份信息判断所述第一客户端是否为免排队客户端；
    所述发送器，还用于若是，则将所述进入频道请求对应的频道的业务信息发送至所述第一客户端；
    所述处理器，还用于若否，则执行判断所述频道当前已在线接入的客户端数量是否达到预定的接入阈值的步骤。
  18. 根据权利要求15-17任一项权利要求所述的直播间的频道访问系统，其特征在于，
    所述处理器，还用于以设定的周期获取所述频道的排队链表中的第一客户端数量和各个第一客户端的排序信息；根据所述第一客户端数量和所述各个第一客户端的排序信息生成所述各个第一客户端的排位信息；
    所述发送器，还用于将所述各个第一客户端的排位信息分别发送至所述各个第一客户端。
  19. 根据权利要求18所述的直播间的频道访问系统，其特征在于，
    所述处理器，用于建立沿时间轴滑动的滑动窗口，并根据所述各个第一客户端进入所述频道的时间建立滑动时间轴；将所述滑动窗口沿所述滑动时间轴进行滑动；
    所述发送器，用于将位于所述滑动窗口内的每个第一客户端的排位信息发送至对应的第一客户端。
  20. 根据权利要求15-19任一项权利要求所述的直播间的频道访问系统，其特征在于，所述处理器，还用于获取所述频道当前客户端数量，将所述当前客户端数量与所述当前已在线接入的客户端数量和所述排队链表中第一客户端的数量的和值进行对比，若所述当前客户端数量与所述和值不一致，则查找出当前已经不在所述频道的客户端，删除所述查找出的客户端，并更新所述排队链表，所述当前客户端数量包括所述当前已在线接入的客户端的实际数量和所述 排队链表中第一客户端的实际数量。
  21. 根据权利要求20所述的直播间的频道访问系统，其特征在于，所述处理器，还用于在缓存中设置一个在设定时间内失效的标志位；在所述第一客户端进入所述频道时，判断所述缓存中是否存在所述标志位；若存在，则不执行获取所述频道当前客户端数量的步骤，若不存在，则执行获取所述频道当前客户端数量的步骤，然后在缓存中重新写入所述标志位。
PCT/CN2014/096019 2014-10-20 2014-12-31 直播间的频道访问方法和系统 WO2016061898A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201410559842.0 2014-10-20
CN201410559842.0A CN104468395B (zh) 2014-10-20 2014-10-20 直播间的频道访问方法和系统

Publications (1)

Publication Number Publication Date
WO2016061898A1 true WO2016061898A1 (zh) 2016-04-28

Family

ID=52913791

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2014/096019 WO2016061898A1 (zh) 2014-10-20 2014-12-31 直播间的频道访问方法和系统

Country Status (2)

Country Link
CN (1) CN104468395B (zh)
WO (1) WO2016061898A1 (zh)

Cited By (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111355970A (zh) * 2019-05-15 2020-06-30 深圳市鸿合创新信息技术有限责任公司 一种直播管理方法及装置、电子设备
CN112312148A (zh) * 2020-09-29 2021-02-02 北京达佳互联信息技术有限公司 业务功能开启方法、装置、电子设备及存储介质
CN113115065A (zh) * 2021-04-13 2021-07-13 北京字跳网络技术有限公司 一种基于直播的数据处理方法及装置
CN113422969A (zh) * 2021-05-25 2021-09-21 北京达佳互联信息技术有限公司 信息拉取方法、装置及服务器
CN113992925A (zh) * 2021-10-12 2022-01-28 江西创成微电子有限公司 直播场景下的通知运维人员的方法和系统、终端设备
CN114124863A (zh) * 2021-09-14 2022-03-01 微梦创科网络科技（中国）有限公司 一种消息连续性识别方法、装置和系统
WO2023151426A1 (zh) * 2022-02-09 2023-08-17 上海哔哩哔哩科技有限公司 直播间人数统计方法及装置

Families Citing this family (22)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN105141978B (zh) * 2015-08-07 2018-05-18 小米科技有限责任公司 视频访问控制方法、装置和云服务器
CN105430529A (zh) * 2015-11-26 2016-03-23 航天恒星科技有限公司 一种视频数据发送方法及系统
CN105516733B (zh) * 2015-12-04 2018-11-06 珠海多玩信息技术有限公司 交互系统及其交互方法
CN106295174A (zh) * 2016-08-09 2017-01-04 北京千安哲信息技术有限公司 一种提供身体状况评测服务的方法和装置
CN106937136B (zh) * 2017-03-29 2020-05-12 武汉斗鱼网络科技有限公司 基于网络直播间统计信息的数据延迟方法及系统
CN108966150B (zh) * 2017-05-17 2021-12-28 北京博瑞彤芸科技股份有限公司 一种健康信息的获取方法
CN108965198A (zh) * 2017-05-17 2018-12-07 北京博瑞彤芸文化传播股份有限公司 一种歌曲信息的处理方法
CN108966151A (zh) * 2017-05-17 2018-12-07 北京博瑞彤芸文化传播股份有限公司 一种选择信息的处理方法
CN108966029A (zh) * 2017-05-17 2018-12-07 北京博瑞彤芸文化传播股份有限公司 一种直播信息的处理方法
CN108964938A (zh) * 2017-05-17 2018-12-07 北京博瑞彤芸文化传播股份有限公司 一种互动信息的接入方法
CN108966312A (zh) * 2017-05-17 2018-12-07 北京博瑞彤芸文化传播股份有限公司 一种互动信息的处理方法
CN107635155B (zh) * 2017-09-20 2020-08-07 香港乐蜜有限公司 一种信息发送方法、装置、设备及存储介质
CN109660489B (zh) * 2017-10-10 2021-07-23 武汉斗鱼网络科技有限公司 直播间登录限制方法、存储介质、电子设备及系统
CN108184140B (zh) * 2018-01-10 2020-03-17 武汉斗鱼网络科技有限公司 一种直播连麦方法、服务器、存储介质及计算机设备
CN109218761A (zh) * 2018-08-07 2019-01-15 邓德雄 一种在视频直播和录像之间切换的方法及系统
CN109257614B (zh) * 2018-09-11 2021-05-07 腾讯数码（天津）有限公司 一种直播会话接入控制方法、装置及存储介质
CN109547481A (zh) * 2018-12-28 2019-03-29 深圳竹云科技有限公司 一种基于Redis有序集合和令牌方式的网站用户会话管理方法
CN109697126B (zh) * 2018-12-29 2021-03-16 网易（杭州）网络有限公司 一种针对服务器的数据处理方法和装置
CN110730161B (zh) * 2019-09-09 2020-08-04 光通天下网络科技股份有限公司 网络靶场实现方法、装置、设备、介质及系统
CN112354175A (zh) * 2020-12-02 2021-02-12 网易（杭州）网络有限公司 游戏广播通知方法、装置、设备和存储介质
CN113727128B (zh) * 2021-08-31 2023-07-07 上海哔哩哔哩科技有限公司 热点流量处理方法及系统
CN115037954B (zh) * 2022-05-18 2024-06-07 阿里云计算有限公司 访问直播的控制方法、装置及系统

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101990262A (zh) * 2009-08-05 2011-03-23 北京大学 无线通信网络的接入控制方法、装置及小区基站
CN102204268A (zh) * 2011-05-31 2011-09-28 华为技术有限公司 直播多码率切换的方法和设备
CN103702226A (zh) * 2013-12-31 2014-04-02 广州华多网络科技有限公司 直播类客户端频道信息的展示方法与装置
US20140136660A1 (en) * 2012-11-12 2014-05-15 Benu Networks, Inc. Extending multicast/broadcast services to wide area networks

Family Cites Families (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US9723042B2 (en) * 2012-02-16 2017-08-01 Telefonaktiebolaget L M Ericsson P2P streaming support
CN104102693B (zh) * 2014-06-19 2017-10-24 广州华多网络科技有限公司 对象处理方法和装置

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101990262A (zh) * 2009-08-05 2011-03-23 北京大学 无线通信网络的接入控制方法、装置及小区基站
CN102204268A (zh) * 2011-05-31 2011-09-28 华为技术有限公司 直播多码率切换的方法和设备
US20140136660A1 (en) * 2012-11-12 2014-05-15 Benu Networks, Inc. Extending multicast/broadcast services to wide area networks
CN103702226A (zh) * 2013-12-31 2014-04-02 广州华多网络科技有限公司 直播类客户端频道信息的展示方法与装置

Cited By (12)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111355970A (zh) * 2019-05-15 2020-06-30 深圳市鸿合创新信息技术有限责任公司 一种直播管理方法及装置、电子设备
CN112312148A (zh) * 2020-09-29 2021-02-02 北京达佳互联信息技术有限公司 业务功能开启方法、装置、电子设备及存储介质
CN112312148B (zh) * 2020-09-29 2023-02-28 北京达佳互联信息技术有限公司 业务功能开启方法、装置、电子设备及存储介质
CN113115065A (zh) * 2021-04-13 2021-07-13 北京字跳网络技术有限公司 一种基于直播的数据处理方法及装置
CN113115065B (zh) * 2021-04-13 2023-05-12 北京字跳网络技术有限公司 一种基于直播的数据处理方法及装置
CN113422969A (zh) * 2021-05-25 2021-09-21 北京达佳互联信息技术有限公司 信息拉取方法、装置及服务器
CN113422969B (zh) * 2021-05-25 2023-08-08 北京达佳互联信息技术有限公司 信息拉取方法、装置及服务器
CN114124863A (zh) * 2021-09-14 2022-03-01 微梦创科网络科技（中国）有限公司 一种消息连续性识别方法、装置和系统
CN114124863B (zh) * 2021-09-14 2023-11-14 微梦创科网络科技（中国）有限公司 一种消息连续性识别方法、装置和系统
CN113992925A (zh) * 2021-10-12 2022-01-28 江西创成微电子有限公司 直播场景下的通知运维人员的方法和系统、终端设备
CN113992925B (zh) * 2021-10-12 2023-11-14 江西创成微电子有限公司 直播场景下的通知运维人员的方法和系统、终端设备
WO2023151426A1 (zh) * 2022-02-09 2023-08-17 上海哔哩哔哩科技有限公司 直播间人数统计方法及装置

Also Published As

Publication number Publication date
CN104468395B (zh) 2018-11-06
CN104468395A (zh) 2015-03-25

Similar Documents

Publication Publication Date Title
WO2016061898A1 (zh) 直播间的频道访问方法和系统
CN104967861B (zh) Cdn网络中的视频缓存系统及方法
US20140298395A1 (en) Methods and systems for playing video on multiple terminals
CN111200657B (zh) 一种管理资源状态信息的方法和资源下载系统
US20150229590A1 (en) Method, server, client, and system for releasing instant messaging key-value data
KR101321664B1 (ko) 무선 통신 디바이스들 간의 그룹 통신들을 지원하는 개선된 데이터베이스 아키텍처
EP2288085A1 (en) P2p based method, device and system for playing media
CN109462631B (zh) 数据处理方法、装置、存储介质及电子装置
CN112016030B (zh) 消息推送的方法、装置、服务器和计算机存储介质
US20200044994A1 (en) Data storage systems and methods using a real-time messaging system
US12056089B2 (en) Method and system for deleting obsolete files from a file system
US20120296946A1 (en) File management apparatus and file management apparatus controlling method
CN111200622B (zh) 一种资源传输方法及装置、存储介质
US20180212816A1 (en) Data storage method, non-volatile computer storage medium, electronic equipment, service capability exposure function, and base station
CN112121413A (zh) 功能服务的响应方法、系统、装置、终端及介质
US10901915B2 (en) Caching system for eventually consistent services
WO2014153963A1 (en) Methods and systems for playing video on multiple terminals
WO2022187008A1 (en) Asynchronous replication of linked parent and child records across data storage regions
CN107992489A (zh) 一种数据处理方法及服务器
US10853892B2 (en) Social networking relationships processing method, system, and storage medium
US7979419B2 (en) Distributed search methods for time-shifted and live peer-to-peer video streaming
EP3274844B1 (en) Hierarchical cost based caching for online media
US20150326686A1 (en) Method, Apparatus and System for Processing User Generated Content
WO2018090315A1 (zh) 数据请求的处理方法及缓存系统
CN104580276A (zh) 信息推送方法、装置、系统及信息接入装置

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 14904619

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

32PN Ep: public notification in the ep bulletin as address of the adressee cannot be established

Free format text: NOTING OF LOSS OF RIGHTS PURSUANT TO RULE 112(1) EPC (EPO FORM 1205 DATED 04/09/2017)

122 Ep: pct application non-entry in european phase

Ref document number: 14904619

Country of ref document: EP

Kind code of ref document: A1