CN102547385A - Distributed stream pushing method, device and system - Google Patents

Distributed stream pushing method, device and system Download PDF

Info

Publication number
CN102547385A
CN102547385A CN 201110451761 CN201110451761A CN102547385A CN 102547385 A CN102547385 A CN 102547385A CN 201110451761 CN201110451761 CN 201110451761 CN 201110451761 A CN201110451761 A CN 201110451761A CN 102547385 A CN102547385 A CN 102547385A
Authority
CN
Grant status
Application
Patent type
Prior art keywords
push
execution unit
slice
message
flow
Prior art date
Application number
CN 201110451761
Other languages
Chinese (zh)
Inventor
宋波
Original Assignee
深圳市同洲视讯传媒有限公司
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

Links

Abstract

The invention discloses a distributed stream pushing method which comprises the steps: a stream pushing main control unit receives a request playing message and obtains a slice to be pushed according to the request playing message; the stream pushing main control unit queries position information of the slice to be pushed to a distributed file system; the stream pushing main control unit receives information of a storage node where the slice to be pushed is, and finds a corresponding stream pushing executing unit according to the information of the storage node, wherein the information of the storage node is returned by the distributed file system; the stream pushing main control unit sends a stream pushing message to the stream pushing executing unit; the stream pushing executing unit data of the slice to be pushed from the storage node of the distributed file system through an internal bus according to the stream pushing message; and the stream pushing executing unit pushes the data of the slice to be pushed. The invention also discloses a device and a system.

Description

一种分布式推流的方法、装置及系统 A distributed plug flow method, apparatus and system for

技术领域 FIELD

[0001] 本发明涉及数字电视领域,尤其涉及一种分布式推流的方法、装置及系统。 [0001] The present invention relates to digital television, and more particularly relates to a distributed plug flow method, apparatus and system. 背景技术 Background technique

[0002] VOD (Video On Demand,视频点播)系统是目前广电的核心系统,VOD支持点播、回看、时移等业务。 [0002] VOD (Video On Demand, Video on Demand) system is the core broadcasting systems, VOD support demand, looking back, times and other services. 随着省网整合和双向改造完成,VOD系统的规模也越来越大。 With the provincial network integration and two-way transformation is complete, the size of the VOD system is also growing. 并发用户数从万级别上升到十万、百万级别。 Number of concurrent users increased from million level to a hundred thousand, one million level. 在VOD系统中,SS (Mreaming Server,视频推流服务器) 负责推送音视频流到IPQAM(IP Quadrature Amplitude Modulation,IP正交幅度调制)设备。 In the VOD system, SS (Mreaming Server, Video Streaming Server) is responsible for audio and video streams to push IPQAM (IP Quadrature Amplitude Modulation, IP quadrature amplitude modulation) device. IPQAM将音视频流调制到电视电缆Cable上,传输到机顶盒终端。 IPQAM audio and video stream is modulated onto a TV cable Cable, is transmitted to the terminal STB. 十万、百万级并发对SS系统要求越来越高。 Hundred thousand, one million concurrent on the SS system increasingly demanding. 如何提高SS整个系统的并发性能、稳定性是各个系统方案、设备提供商的一个难题。 How to improve the entire system of concurrent performance SS, stability is a problem individual system solutions, equipment providers.

[0003] 传统的SS集群推流方案如图1所示,每台SS自身包含存储或者外部挂接存储设备。 [0003] SS conventional plug-flow solutions cluster shown in Figure 1, each SS itself comprises a storage device or an external memory attached. 音视频媒体数据采用整个文件的方式保存在存储设备中,SS从存储设备中读取文件, 将数据推送到IPQAM设备,再由IPQAM调制成RF(Radio frequency,射频)信号,推送到用户终端(例如:机顶盒、数字电视一体机等数字电视接收终端)。 Audio and video media data file using the entire way in the storage device, SS read files from the storage device, the data is pushed to IPQAM device, then by IPQAM modulated to RF (Radio frequency, RF) signal, pushed to the user terminal ( For example: top boxes, integrated digital TVs and other digital television receiving terminal). 但由于用户行为的不规律性,可能导致SS服务集群出现负载不均衡的情况。 However, due to irregularities in users' behavior, may cause load imbalance situation SS service cluster. 例如某些SS上,由于其上面分布的节目点播的人数比较多,导致负载很重,其他的SS服务器则比较空闲,负载重的服务器变成了瓶颈。 For example, some SS, due to the relatively large number of programs distributed thereon demand, leading to a heavy load, the other SS server is relatively idle, heavily loaded server becomes a bottleneck.

发明内容 SUMMARY

[0004] 本发明实施例在于提供一种分布式推流的方法、装置及系统,可以通过将视频推流服务器分成推流主控单元和推流执行单元,并通过将文件切成小片,分布到各个存储中, 使负载更均衡。 [0004] The embodiment of the present invention is to provide a method embodiment, apparatus and system for a distributed plug flow, plug flow can be the video server into a plug flow through the main control unit and plug flow execution unit, and cut into small pieces by the file distribution to each store, the load is more balanced.

[0005] 本发明是这样实现的,一种分布式推流的方法,包括: [0005] The method of the present invention is implemented as a distributed plug flow, comprising:

[0006] 推流主控单元接收请求播放消息,并根据所述请求播放消息,获得即将推送的切片; [0006] Streaming playback master unit receives a request message, and play message according to the request, get pushed upcoming sections;

[0007] 所述推流主控单元向分布式文件系统询问即将推送的切片的位置信息; [0007] The main control unit interrogation plug flow coming push slice position information to distributed file system;

[0008] 所述推流主控单元接收所述分布式文件系统返回的所述即将推送的切片所在的存储节点信息,根据所述存储节点信息找到对应的推流执行单元; [0008] The push slice information storage node where the master unit receives the stream distributed file system returns the coming push, push find the corresponding execution unit stream information according to the storage node;

[0009] 所述推流主控单元向所述推流执行单元发送推流消息; [0009] main control unit of the plug-flow flow execution unit sends a push message to the plug-flow;

[0010] 所述推流执行单元根据所述推流消息,通过内部总线从所述分布式文件系统的存储节点读取所述即将推送的切片数据; [0010] The execution unit of the plug-flow stream of the push message, reads the slice data is about to push the storage node from the distributed file system by an internal bus;

[0011] 所述推流执行单元推送所述即将推送的切片数据。 [0011] The push slice data flow execution unit is about to push the push.

[0012] 本发明的另一目的在于提供一种视频推流装置,包括: [0012] Another object of the present invention is to provide a video stream pushing apparatus, comprising:

[0013] 推流主控单元,用于接收请求播放消息,并根据所述请求播放消息,获得即将推送的切片,向分布式文件系统询问所述即将推送的切片的位置信息,并接收所述分布式文件系统返回的所述即将推送的切片所在的存储节点信息,根据所述存储节点信息找到对应的推流执行单元,并向对应的推流执行单元发送推流消息; [0013] The plug-flow control unit, for receiving a request message playback, and play message according to the request, obtain the upcoming push slice, location information of the inquiry is about to push sections distributed file system, and receiving the information of the storage node is about to return to the distributed file system where the push slice, find the corresponding plug flow execution unit according to the storage node information corresponding to the plug-flow flow execution unit sends a push message;

[0014] 推流执行单元,设置在分布式文件系统的存储节点中,与所述推流主控单元通过网络相连,与所述分布式文件系统的存储节点通过系统内部总线相连,接收所述推流消息, 并根据所述推流消息,通过内部总线从所述分布式文件系统读取即将推送的切片的数据, 推送所述即将推送的切片的数据。 [0014] The plug flow execution unit, disposed in the storage node of a distributed file system, said main control unit and the plug-flow is connected via a network, and the storage node of the distributed file system via an internal bus system are connected to receive push message flow, according to the push message stream, the read data is about to push the slices from the distributed file system via the internal bus, the data is about to push the push slice.

[0015] 本发明的又一目的在于提供一种分布式推流的系统,视频推流装置和分布式文件系统: [0015] A further object of the present invention to provide a system for distributed plug flow, plug flow apparatus and video distributed file system:

[0016] 所述视频推流的装置,包括: [0016] The pushing means video stream, comprising:

[0017] 推流主控单元,用于接收请求播放消息,并根据所述请求播放消息,获得即将推送的切片,向所述分布式文件系统询问所述即将推送的切片的位置信息,并接收所述分布式文件系统返回的所述即将推送的切片所在的存储节点信息,根据所述存储节点信息找到对应的推流执行单元,并向对应的推流执行单元发送推流消息; [0017] The plug-flow control unit, for receiving a request message playback, and play message according to the request, obtain the upcoming push slice, location information of the inquiry is about to push the slices distributed file system, and receives storing the node information distributed file system is about to push the return sections are located, to find the corresponding plug flow execution unit according to the storage node information corresponding to the plug-flow flow execution unit sends a push message;

[0018] 推流执行单元,与推流主控单元相连,并通过内部总线与所述分布式文件系统相连,接收所述推流消息,并根据所述推流消息,通过内部总线从所述分布式文件系统读取即将推送的切片数据,推送所述即将推送的切片数据; [0018] The plug flow execution unit, the main control unit is connected with plug flow, and through an internal bus connected to said distributed file system, receiving the push message flow, according to the push message flow through said internal bus from distributed file system reads the data will be pushed slice, slice data is about to push the pushed;

[0019] 分布式文件系统,与所述视频推流装置相连,用于接收询问所述即将推送的切片的位置信息的消息,并根据所述即将推送的切片的信息获得该即将推送的切片所在的存储节点信息,并将所述存储节点信息返回给所述视频推流装置。 [0019] The distributed file system, connected to the video stream pushing means for receiving location information query message is about to push the slice and the slice obtained where the upcoming The information pushed to the pushed upcoming sections the information storage node and the storage node information returned to the video stream pushing means.

[0020] 实施上述实施例,通过将视频推流服务器分成推流主控单元和推流执行单元,并通过将文件切成小片,分布到各个存储节点中,能够同时利用多台服务器的能力,有效的解决不均衡性,大大提高系统的并发性能;而且通过将推流执行单元与存储节点部署在一个系统上,使得推流执行单元从存储节点中读取数据通过的是存储节点的系统内部总线,从而能有效的加快数据读取速度,同时没有额外网络开销。 [0020] Example embodiments described above, by pushing the video streaming server plug flow into the main control unit and plug flow execution unit, and cut into small pieces by the document, distributed to the respective storage node, the ability to simultaneously use multiple servers, effective solution to the imbalance, the system greatly improve concurrency; push through and flow execution unit deployed on the storage node with a system, such that the plug flow execution unit reads data from the storage node is an internal node of the storage system bus, so that it can effectively speed up the data read speed, and there is no additional network overhead.

附图说明 BRIEF DESCRIPTION

[0021] 为了更清楚地说明本发明实施例或现有技术中的技术方案,下面将对实施例或现有技术描述中所需要使用的附图作简单地介绍,显而易见地,下面描述中的附图仅仅是本发明的一些实施例,对于本领域普通技术人员来讲,在不付出创造性劳动性的前提下,还可以根据这些附图获得其他的附图。 [0021] In order to more clearly illustrate the technical solutions in the embodiments or the prior art embodiment of the present invention, briefly introduced hereinafter, embodiments are described below in the accompanying drawings or described in the prior art needed to be used in describing the embodiments the drawings are only some embodiments of the present invention, those of ordinary skill in the art is concerned, without any creative effort, and can obtain other drawings based on these drawings.

[0022] 图1是现有技术SS网络架构示意图; [0022] FIG. 1 is a prior art schematic diagram of a network architecture SS;

[0023] 图2是本发明实施例的一种分布式推流的方法流程示意图; [0023] FIG. 2 is a process flow of a distributed embodiment of the present invention plug flow schematic;

[0024] 图3是本发明实施例的分布式推流网络架构示意图; [0024] FIG. 3 is a schematic view of a distributed network architecture of the present invention, an embodiment of the plug flow;

[0025] 图4是本发明一种分布式推流的方法的另一实施例的流程示意图; [0025] FIG. 4 is a schematic flow diagram of another method of the present invention is a distributed plug flow embodiment;

[0026] 图5是本发明实施例的一种视频推流装置的结构示意图; [0026] FIG. 5 is a schematic structure of a video stream pushing apparatus of the present embodiment of the invention;

[0027] 图6是本发明实施例的一种分布式推流的系统结构示意图。 [0027] FIG. 6 is a diagram showing the system structure of a distributed embodiment of the present invention plug flow.

具体实施方式 Detailed ways

[0028] 为了使本发明所要解决的技术问题、技术方案及有益效果更加清楚明白,以下结合附图及实施例,对本发明进行进一步详细说明。 [0028] In order to make the technical problem to be solved by the present invention, technical solutions and beneficial effects clearer, the accompanying drawings and the following embodiments, the present invention will be further described in detail.

[0029] 请参阅图2示出的本发明实施例的一种分布式推流的方法流程示意图,包括: [0029] See flowchart of a method of plug flow in a distributed embodiment of the present invention shown in FIG. 2 embodiment, comprising:

[0030] 步骤S201 :推流主控单元接收请求播放消息,并根据请求播放消息,获得即将推送的切片。 [0030] Step S201: The main control unit receiving a plug flow request message playback, and playback according to the request message, about to push slices obtained.

[0031] 具体的,该请求播放消息包括影片播放的位置。 [0031] Specifically, the request message including the playback position of video playback.

[0032] 步骤S202 :推流主控单元向分布式文件系统询问该即将推送的切片的位置信息。 [0032] Step S202: The main control unit plug flow position information of the inquiry to the upcoming sections pushed distributed file system.

[0033] 步骤S203 :推流主控单元接收该分布式文件系统返回的该即将推送的切片所在的存储节点信息,根据该存储节点信息找到对应的推流执行单元。 [0033] Step S203: The main control unit receives the plug flow information of the slice where the storage node is about to push the distributed file system of the returned, find the corresponding execution units based on the plug flow information storage node.

[0034] 步骤S204 :推流主控单元向该推流执行单元发送推流消息。 [0034] Step S204: The main control unit to push the plug flow stream flow execution unit to send push messages.

[0035] 步骤S205 :推流执行单元根据该推流消息,通过内部总线从分布式文件系统的存储节点读取即将推送的切片数据。 [0035] Step S205: plug flow execution unit, slice data storage node is about to push the distributed file system is read from the internal bus via stream based on the push message.

[0036] 步骤S206 :该推流执行单元推送即将推送的切片数据。 [0036] Step S206: The Push slice data stream coming push push execution unit.

[0037] 实施上述实施例,通过将视频推流服务器分成推流主控单元和推流执行单元,并通过将文件切成小片,分布到各个存储中,能够同时利用多台服务器的能力,有效的解决不均衡性,大大提高系统的并发性能,而且由于推流执行单元与分布式文件系统读取数据通过的是内部总线,即读取本地切片文件的数据,能有效的加快数据读取速度,同时没有额外网络开销。 [0037] Example embodiments described above, by pushing the video streaming server plug flow into the main control unit and plug flow execution unit, and cut into small pieces by the document, distributed to each memory, the ability to simultaneously use multiple servers, effective solution imbalance, greatly improving performance of the system complicated, and since the execution unit and the plug-flow distributed file system is read internal bus, local data slice i.e., read through the file data, which can effectively speed up the data read speed while no additional network overhead.

[0038] 下面结合图4示出的本发明一种分布式推流的方法的另一实施例的流程示意图,在该实施例中,为了更清楚的描述,可以结合图3示出的分布式推流网络架构图,该分布式推流网络架构包括:SS Master (Streaming Server Master,推流主控单元),SS Workerl (Streaming Server Worker,推流执行单元),SS Worker2 和SS Worker3,其中SS Master管理SS Worker 1, SS Worker2和SS Worker30还包括分布式文件系统,该分布式文件系统包括存储节点1,存储节点2和存储节点3,以及文件控制单元。 Schematic representation of another method of the invention [0038] below with reference to FIG. 4 shows a distributed embodiment of the plug flow, in this embodiment, for greater clarity of description, may be combined with the distributed illustrated in FIG. 3 plug flow network architecture, the network architecture includes a distributed plug flow: SS Master (streaming Server Master, the extrusion flow master unit), SS Workerl (streaming Server Worker, extrusion flow execution unit), and worker2 SS SS Worker3, wherein SS Master management SS Worker 1, SS worker2 SS Worker30 and further comprising a distributed file system, the distributed file system includes a storage node, the storage node and the storage node 2 3, and a file control unit. 其中SS Workerl 和存储节点1部署在一起,SS Workerl从存储节点1读取数据是通过系统内部总线;SS Worker2和存储节点2部署在一起,SS Worker2从存储节点2读取数据是通过系统内部总线;SS Worker3和存储节点3部署在一起,SS Worker3从存储节点3读取数据是通过系统内部总线。 Wherein SS Workerl storage node 1 and deployed together, SS Workerl reading data from storage node 1 through the internal system bus; SS Worker2 deployed together and the storage node 2, SS Worker2 2 reads data from the storage node by the internal system bus ; SS Worker3 deployed together and the storage node 3, SS Worker3 3 reads data from the storage node by an internal system bus. 文件控制单元负责管理整个分布式文件系统的切片文件的存储。 File slice file storage control unit is responsible for managing the entire distributed file system. 在分布式文件系统中,数据以切片的形式存储在各个存储节点中。 In a distributed file system, the data stored in the respective storage nodes in the form of slices. 比如,一份为100MB的数据,可以被切为10份,每个切片大小为10MB,切片1、切片2、切片3、切片4存储在存储节点1,切片5、切片6、切片7存储在存储节点2,切片8、切片9、切片10存储在存储节点3。 For example, a 100MB of data, can be cut into 10 parts, each slice size is 10MB, slice 1, slice 2, slice 3, slice 4 is stored in the storage node 1, slice 5, slice 6, slice 7 stores in 3 storage nodes in the storage node 2, sections 8, 9 slice, slice 10 is stored. 以下实施例,都以用户终端点播该100MB的数据为例。 The following examples, all the data of the user terminal on demand Example 100MB.

[0039] 步骤S401 :用户终端向SS Master发出请求播放消息。 [0039] Step S401: The user terminal displays the request message to the SS Master.

[0040] 具体的,该请求播放消息包括影片播放的位置。 [0040] Specifically, the request message including the playback position of video playback.

[0041] 步骤S402 =SS Master根据请求播放消息,获得即将推送的切片。 [0041] Step S402 = SS Master PLAY message according to the request, get pushed upcoming sections.

[0042] 具体的,SS Master根据影片播放的位置,获得即将推送的切片。 [0042] Specifically, SS Master video playback depending on the position, about to push to get a slice. 比如,用户终端发送从影片第30MB的地方开始播放,则SS Master获得即将推送的切片即为切片4。 For example, the user terminal sends the movie to start playing from the first 30MB place, the SS Master is about to push to get a slice is the slice 4.

[0043] 步骤S403 =SS Master向分布式文件系统询问该即将推送的切片的位置信息。 [0043] Step S403 = SS Master interrogates the coming push slice location information distributed file system.

[0044] 具体的,SS Master向文件控制单元发送询问切片4的位置信息的请求,即询问切片4存储在哪个存储节点上。 [0044] Specifically, SS Master control unit sends an inquiry to the file requesting location information of the slice 4, slice i.e. asking 4 on which is stored in the storage node. [0045] 步骤S404 :分布式文件系统返回该即将推送的切片所在的存储节点信息给SS Master0 [0045] Step S404: distributed file system returns the storage node sections where the information is about to push to the SS Master0

[0046] 具体的,存储节点信息包括存储节点编号或IP地址。 [0046] Specifically, the storage node information includes a storage node number or IP address. 文件控制单元得知切片4存储在存储节点1上,将存储节点1的编号或者IP地址返回给SS Master。 The control unit that the file on the storage node 1, node 1 returns the stored number or IP address to 4 stores SS Master slice.

[0047] 步骤S405 =SS Master根据该存储节点信息找到对应的SS Worker。 [0047] Step S405 = SS Master SS Worker finds the corresponding node based on the stored information.

[0048] 具体的,因为SS Workerl和存储节点1部署在一起,所以SS Master根据存储节点1的信息可以找到SS Workerl0 [0048] Specifically, since the memory node 1 and the SS Workerl deployed together, the SS Master SS Workerl0 The information can be found in the storage node 1

[0049] 步骤S406 =SS Master向该SS Worker发送推流消息,通知该SS Worker开始推流。 [0049] Step S406 = SS Master transmits to the SS Worker plug flow message, notifying the SS Worker plug flow starts.

[0050] 具体的,SS Master通知SS Workerl开始推流,推送的内容是切片4。 [0050] Specifically, SS Master inform SS Workerl plug flow starts, the slice 4 is pushed content. 该推流消息包括即将推送的切片信息、推送码流大小、推流起始位置和推流终止位置、IPQAM地址等。 The push message comprises a stream of information coming slice push, push the size of the code stream, the start position of plug flow and plug flow end position, IPQAM address.

[0051] 步骤S407 :该SS Worker根据推流消息,从分布式文件系统读取数据到缓存。 [0051] Step S407: The plug flow of the SS Worker message, reads data from the cache to the distributed file system.

[0052] 具体的,SS Workerl从存储节点1读取切片4的数据内容到缓存。 [0052] Specifically, SS Workerl content data read from the storage sections 4 to node 1 buffer. 因为SS Workerl 和存储节点1部署在一起,SS Workerl从存储节点1读取切片4的数据通过是内部总线, 即读取本地切片文件的数据,能有效的加快数据读取速度,同时没有额外网络开销。 Since the storage node 1 and SS Workerl deployed together, SS Workerl slice data read from the storage node 14 via the internal bus, the local data slice i.e., read file, which can effectively speed up the data read speed, and there is no additional network overhead.

[0053]步骤 S408 :该SS Worker 推送数据到IPQAM。 [0053] Step S408: the SS Worker push data to IPQAM.

[0054] 具体的,SS Workerl通过IP网络推送切片4的数据到IPQAM。 [0054] Specifically, SS Workerl push data slice through the IP network 4 to IPQAM. IPQAM调制数据, 输出RF,用户终端接收RF经过处理后观看节目。 IPQAM data modulated RF output, the user terminal receives RF viewing the program after treatment.

[0055] 在执行步骤S408的同时,该SS Worker还执行: [0055] In step S408 performed at the same time, the SS Worker further performs:

[0056] 步骤S409 :该SS Worker向SS Master发送获取下一个SS Worker的消息。 [0056] Step S409: transmitting to the SS Worker SS Master SS Worker get the next message.

[0057] 步骤S410 =SS Master根据下一个切片信息,向分布式文件系统询问下一个切片的位置信息。 [0057] Step S410 = SS Master next slice information according to a distributed file system asks slice position information.

[0058] 具体的,下一个切片是切片5,询问切片5存储在哪个存储节点上。 [0058] Specifically, the next slice is a slice 5, slice interrogation 5 on which the storage node is stored.

[0059] 步骤S411 :分布式文件系统返回该下一个切片所在的存储节点信息给SS Master0 [0059] Step S411: a distributed file system returns the next slice location information to the storage node SS Master0

[0060] 具体的,文件控制单元得知切片5存储在存储节点2上,将存储节点2的编号或者IP地址返回给SS Master。 [0060] Specifically, the control unit that the file on the storage node 2, node 2 returns the number or the stored IP address to the memory 5 SS Master slice.

[0061] 其中,存储节点2和存储节点3可以互为备份,所以切片5也可能存储在存储节点3上,这时,文件控制单元也可以将存储节点3的编号或者IP地址返回给SS Master。 [0061] wherein the storage node and the storage node 2 can back up each other 3, 5 may also be sliced ​​so on the storage node 3, this time, the control file storage unit may also store the node ID 3 or IP address may be returned to the SS Master .

[0062] 步骤S412 =SS Master根据该存储节点信息找到对应的SS Worker。 [0062] Step S412 = SS Master SS Worker finds the corresponding node based on the stored information.

[0063] 具体的,因为SS Worker2和存储节点2部署在一起,SS Worker2从存储节点2读取数据通过是内部总线,所以SS Master根据存储节点2的信息可以找到SS Worker2。 [0063] Specifically, since the memory node 2 and SS Worker2 deployed together, SS Worker2 2 to read data from the memory node via an internal bus, SS Master SS Worker2 be found based on information stored in the node 2.

[0064] 当文件控制单元将存储节点3也返回给SS Master时,SS Master得知切片5可以从存储节点2或者存储节点上获得,即SS Master可以选择SS Workerl或SS Worker2。 [0064] When the file control unit 3 returns to the storage node when the SS Master, SS Master 5 slices that can be obtained from the memory node 2 or the storage node, i.e., may be selected SS Workerl SS Master or SS Worker2. 这时,SS Master可以根据负载均衡的策略,来选择其中一个SSWorker。 At this time, SS Master can load balance strategy to choose a SSWorker. 在本实施例中,SS Master 选择SS Worker2。 In the present embodiment, SS Master select SS Worker2.

[0065]步骤 S413 :SS Master 将下一个SS Worker 的信息发给当前SS Worker。 [0065] Step S413: SS Master will next information is destined to the SS Worker SS Worker.

[0066]具体的,SS Master 将SS Worker2 的信息发给SS Workerl。 [0066] Specifically, SS Master information SS Worker2 issued to SS Workerl.

[0067] 步骤S414 :当当前SS Worker即将完成推流任务时,当前SS Worker向下一个SSWorker发送准备推流消息。 [0067] Step S414: when the current SS Worker plug flow is about to complete the task, a current SS Worker SSWorker down stream message sent ready to push.

[0068] 具体的,该准备推流消息包括即下一个切片信息、推送码流大小、推流起始和终止位置、IPQAM地址等。 [0068] Specifically, the message comprising a stream ready to push the next slice information i.e., the size of the stream push, plug flow start and end positions, address IPQAM. 当SS Workerl离推流结束还剩IOms时,SS Workerl向SS Worker2 发送准备推流消息,该推流消息包括即将推送的切片5信息等。 When the SS Workerl left end from plug flow IOms, SS Workerl sends a message flow SS Worker2 ready to push the push message includes a stream information or the like is about 5 slices pushed.

[0069] 步骤S415 :下一个SS Worker接收到准备推流消息后,从分布式文件系统读取数据到缓存。 [0069] Step S415: receiving a next SS Worker ready to push the message stream, read data from the cache to the distributed file system.

[0070] 具体的,SS Worker2从存储节点2读取切片5的数据内容到缓存。 [0070] Specifically, SS Worker2 from the data storage node 2 reads the contents of the slice 5 to the cache.

[0071] 步骤S416 :当当前SS Worker完成推流任务时,当前SS Worker立刻通知下一个SS Worker启动推流,同时将推流结束信息汇报给SS Master。 [0071] Step S416: when the current SS Worker Task complete plug flow, the current SS Worker SS Worker immediately notice a start plug flow, while the report information to the push end of stream SS Master.

[0072] 具体的,SS Workerl完成推流任务时,SS Workerl立刻通知SS Worker2启动推流,同时将推流结束信息汇报给SS Master。 [0072] Specifically, when the SS Workerl complete plug-flow task, SS Workerl immediately notify SS Worker2 start plug-flow, while the flow will push the end of the reporting information to the SS Master.

[0073] 步骤S417 :下一个SS Worker接收到通知后,延时一段时间。 [0073] Step S417: SS Worker next after receiving the notification, delaying for a period of time.

[0074] 具体的,SS Worker2接收到通知后,判断延时时间是否到达预设的延时时间,例如,预设的延时时间为20ms,则延时20ms。 After [0074] Specifically, SS Worker2 receiving the notification, it is determined whether the delay time reaches a predetermined delay time, for example, the predetermined delay time is 20ms, then the delay of 20ms. 如果SS Worker 2马上推流,IPQAM可能会先收到SS Worker2的数据,后收到SS Workerl的数据,出现乱序的问题。 If SS Worker 2 immediately plug flow, IPQAM may receive the first SS Worker2 data, after receiving SS Workerl data, problems arise out of order. 其中,具体的延时时间与SS Workerl推流结束时间有关。 Wherein, the specific delay time and the end time SS Workerl relevant plug flow.

[0075] 步骤S418 :下一个SS Worker推送数据到IPQAM。 [0075] Step S418: The next push data to SS Worker IPQAM.

[0076] 具体的,SS Worker2通过IP网络推送切片5的数据到IPQAM。 [0076] Specifically, SS Worker2 push data slice through the IP network 5 to IPQAM. IPQAM调制数据, 输出RF,用户终端接收RF经过处理后观看节目。 IPQAM data modulated RF output, the user terminal receives RF viewing the program after treatment.

[0077] 重复步骤S409直至所有用户需要的切片数据推送完毕。 [0077] Step S409 is repeated until all slices of the data required by the user will push.

[0078] 在以上步骤中,SS Worker可以定期汇报推流状态信息给SS Master,推流状态信息包括当前的推流位置、推流片序号等,以便SS Master实时监控各个SS Worker。 [0078] In the above step, SS Worker plug flow may periodically report status information to the SS Master, plug-flow state information including the current position of the plug-flow, plug flow sheet number, etc., so that real-time monitoring of each SS Master SS Worker.

[0079] 在以上实施例中,由一个SS Master管理多个SS Worker。 [0079] In the above embodiment, a plurality of a managed SS Master SS Worker. 然而,在本发明另一实施例中,可通过由多个SS Master组成的SS Master服务集群,来构成该分布式推流网络架构,通过SS Master服务集群管理多个SS Worker。 However, in another embodiment of the present invention, by SS Master service cluster composed of a plurality of SS Master, constituted the distributed network architecture plug-flow through the plurality of cluster management service SS Master SS Worker.

[0080] 实施上述实施例,通过将视频推流服务器分成SS Master和SS Worker,并通过将文件切成小片,分布到各个存储中,能够同时利用多台服务器的能力,有效的解决不均衡性,大大提高系统的并发性能,而且由于SS Worker与存储节点部署在一块,SS Worker从存储系统读取数据通过的是内部总线,能有效的加快数据读取速度,同时没有额外网络开销。 [0080] Example embodiments described above, by pushing the video stream into a server SS Master and SS Worker, and cut into small pieces by the document, distributed to each memory, the ability to simultaneously use multiple servers, unevenness effective solution , greatly improving the performance of the system complicated, and because the SS Worker deployed in a storage node, SS Worker data read from the storage system is an internal bus, which can effectively speed up the data read speed, and there is no additional network overhead.

[0081] 下面结合图5示出的本发明实施例的一种视频推流装置的结构示意图,进一步详细说明本发明实施例一种视频推流装置的结构。 Streaming video schematic structure of a device according to the present embodiment of the invention [0081] below with reference to FIG. 5 shows a further embodiment of a structure of a video stream pushing device according to the present invention will be described in detail.

[0082] 推流主控单元501,用于接收用户终端发送的请求播放消息,并根据请求播放消息,获得即将推送的切片,向分布式文件系统询问该即将推送的切片的位置信息,并接收该分布式文件系统返回的该即将推送的切片所在的存储节点信息,根据该存储节点信息找到对应的推流执行单元,并向对应的推流执行单元发送推流消息。 [0082] Streaming main control unit 501 configured to receive a request sent by the user terminal displays the message, and displays the message according to the request, obtain the upcoming push slice, the slice position information inquiry soon pushed to the distributed file system, and receives information of the storage node is about to push the distributed file system returns slice resides, finds the corresponding plug flow execution unit based on the storage node information, and sends a push message stream corresponding to a plug flow execution unit.

[0083] 推流执行单元502,设置在分布式文件系统的存储节点中,与推流主控单元501相连,与分布式文件系统的存储节点通过系统内部总线相连,接收该推流消息,并根据该推流消息,通过内部总线从分布式文件系统读取数据,推送该数据。 [0083] Streaming executing unit 502, the storage node disposed in a distributed file system, the master control unit is connected to the plug flow 501, and the storage node is connected to a distributed file system via an internal bus system, receives the push message stream, and the flow of the push message, reads data from the internal bus via the distributed file system, the data is pushed.

[0084] 实施上述实施例,通过将视频推流服务器分成推流主控单元和推流执行单元,并通过将文件切成小片,分布到各个存储中,能够同时利用多台服务器的能力,有效的解决不均衡性,大大提高系统的并发性能,而且由于推流执行单元与分布式文件系统读取数据通过的是内部总线,能有效的加快数据读取速度,同时没有额外网络开销。 [0084] Example embodiments described above, by pushing the video streaming server plug flow into the main control unit and plug flow execution unit, and cut into small pieces by the document, distributed to each memory, the ability to simultaneously use multiple servers, effective solution imbalance, greatly improving performance of the system complicated, and since the internal bus is a plug flow execution units and read data distributed file system, which can effectively speed up the data read speed, and there is no additional network overhead.

[0085] 具体的,推流执行单元502,还用于向SS Master发送获取下一个SS Worker的消肩、ο [0085] Specifically, the plug flow execution unit 502 is further configured to send to SS Master shoulder obtain cancellation of the next SS Worker, o

[0086] 推流主控单元501,还用于接收该获取下一个SS Worker的消息,并根据下一个切片信息,向分布式文件系统询问下一个切片的位置信息,接收该分布式文件系统返回的下一个切片所在的存储节点信息,根据该存储节点信息找到下一个推流执行单元,并将下一个推流执行单元信息返回。 [0086] Streaming main control unit 501 is further configured to receive the get the next message SS Worker, and a slice according to the following information, to the location information of the distributed file system asks a slice, to receive the distributed file system returns next slice information storage node is located, according to this information to find a storage node plug flow execution unit, the execution unit and a plug-flow information is returned.

[0087] 具体的,在根据该存储节点信息找到下一个推流执行单元步骤中,SS Master可以根据负载均衡的策略,来选择其中一个SS Worker。 [0087] Specifically, in the step of performing a push unit according to the flow information is found in the storage node, SS Master can load balance strategy to choose a SS Worker.

[0088] 推流执行单元502,与其他推流执行单元502相连,还用于当完成推流任务时,向下一个SS Worker发送推流消息。 [0088] The plug flow execution unit 502, connected to another plug flow execution unit 502 is further configured to push upon completion of the task flow down a plug flow SS Worker transmitted message.

[0089] 具体的,该推流消息包括即将推送的切片信息、推送码流大小、推流起始和终止位置、IPQAM地址等。 [0089] Specifically, the slice information includes a push message stream coming push, push stream size, plug flow start and end positions, address IPQAM.

[0090] 具体的,在当完成推流任务时,向下一个SS Worker发送推流消息步骤之前,推流执行单元502还用于当即将完成推流任务时,向下一个SS Worker发送准备推流消息;具体的,该准备推流消息包括即将推送的切片信息、推送码流大小、推流起始和终止位置、IPQAM 地址等。 [0090] Specifically, upon completion of the task flow pushed down a SS Worker message before sending step plug flow, plug flow execution unit 502 is further configured to plug flow when the task being completed, the next transmission ready to push SS Worker message stream; specifically, the stream sections ready to push message includes information about to push, push stream size, plug flow start and end positions, address IPQAM. 下一个SS Worker接收到准备推流消息后,从分布式文件系统读取数据到缓存;当下一个SS Worker接收到推流消息后,会延时一段时间,再将数据推送到IPQAM。 After a next SS Worker ready to push the received message stream, the distributed file system data read from the cache; SS Worker immediately after receiving a push message stream, will be delayed for some time, and then push data to IPQAM.

[0091] 实施上述实施例,通过将视频推流服务器分成推流主控单元和推流执行单元,并通过将文件切成小片,分布到各个存储中,能够同时利用多台服务器的能力,有效的解决不均衡性,大大提高系统的并发性能,而且由于推流执行单元与分布式文件系统读取数据通过的是内部总线,能有效的加快数据读取速度,同时没有额外网络开销。 [0091] Example embodiments described above, by pushing the video streaming server plug flow into the main control unit and plug flow execution unit, and cut into small pieces by the document, distributed to each memory, the ability to simultaneously use multiple servers, effective solution imbalance, greatly improving performance of the system complicated, and since the internal bus is a plug flow execution units and read data distributed file system, which can effectively speed up the data read speed, and there is no additional network overhead.

[0092] 下面结合图6示出的本发明实施例一种分布式推流的系统示意图,进一步详细说明本发明实施例一种分布式推流的系统,包括视频推流装置601,分布式文件系统602 : The system schematic diagram of a distributed embodiment of the present invention plug flow [0092] below with reference to FIG. 6 illustrates a further detailed embodiment of a distributed system embodiment of the present invention, plug flow, plug flow device 601 includes a video, distributed file system 602:

[0093] 视频推流装置601,包括: [0093] Streaming video device 601, comprising:

[0094] 推流主控单元6011,用于接收请求播放消息,并根据请求播放消息,获得即将推送的切片,向分布式文件系统602询问该即将推送的切片的位置信息,并接收该分布式文件系统602返回的该即将推送的切片所在的存储节点信息,根据该存储节点信息找到对应的推流执行单元,并向对应的推流执行单元发送推流消息。 [0094] Streaming master unit 6011, configured to receive a request to play message, a message according to the request and play, obtained soon push slice, the interrogation position is about to push information to a slice distributed file system 602, and receives the distributed information of the storage node is about to push the file system 602 returns slice resides, finds the corresponding plug flow execution unit based on the storage node information, and sends a push message stream corresponding to a plug flow execution unit.

[0095] 推流执行单元6012,设置在分布式文件系统的存储节点中,与推流主控单元501 通过网络相连,与所述分布式文件系统602的存储节点通过系统内部总线相连,接收该推流消息,并根据该推流消息,通过内部总线从分布式文件系统602读取数据,推送该数据。 [0095] The plug flow execution unit 6012, the storage node disposed in a distributed file system, the master control unit is connected to the plug flow through the network 501, and the storage node of the distributed file system 602 through an internal bus system are connected to receive the push message flow, and flow based on the push message, reads data from the distributed file system 602 via an internal bus, the data is pushed.

[0096] 分布式文件系统602,与视频推流装置601相连,用于接收询问该即将推送的切片的位置信息的消息,并根据该即将推送的切片的信息获得该即将推送的切片所在的存储节点信息,并将该存储节点信息返回给视频推流装置601。 [0096] distributed file system 602, and the video device 601 is connected to plug flow, for receiving a query message is about to push the slice position information, and stores the obtained push upcoming sections based on the information about to where the push slice node information, and returns the information to the video storage node plug flow device 601.

[0097] 在本发明实施例中,所述连接推流执行单元6012与推流主控单元501的网络,即可以是有线通信网络,也可以是无线通信网络。 [0097] In an embodiment of the present invention, the network connection plug flow execution unit 6012 and the main control unit 501 of the plug flow, i.e., may be a wired communication network, or a wireless communication network. [0098] 实施上述实施例,通过将视频推流服务器分成推流主控单元和推流执行单元,并通过将文件切成小片,分布到各个存储中,能够同时利用多台服务器的能力,有效的解决不均衡性,大大提高系统的并发性能,而且由于推流执行单元与分布式文件系统读取数据通过的是内部总线,能有效的加快数据读取速度,同时没有额外网络开销。 [0098] Example embodiments described above, by pushing the video streaming server plug flow into the main control unit and plug flow execution unit, and cut into small pieces by the document, distributed to each memory, the ability to simultaneously use multiple servers, effective solution imbalance, greatly improving performance of the system complicated, and since the internal bus is a plug flow execution units and read data distributed file system, which can effectively speed up the data read speed, and there is no additional network overhead.

[0099] 需要说明的是,通过以上的实施方式的描述,本领域的技术人员可以清楚地了解到本发明可借助软件加必需的硬件平台的方式来实现,当然也可以全部通过硬件来实施。 [0099] Incidentally, the above description of embodiments, those skilled in the art may clearly understand that the present invention may be implemented by software plus a necessary hardware platform, and definitely may also be implemented by hardware. 基于这样的理解,本发明的技术方案对背景技术做出贡献的全部或者部分可以以软件产品的形式体现出来,该计算机软件产品可以存储在存储介质中,如ROM/RAM、磁碟、光盘等,包括若干指令用以使得一台计算机设备(可以是个人计算机,服务器,或者网络设备等)执行本发明各个实施例或者实施例的某些部分所述的方法。 Based on such understanding, the technical solutions of the present invention contributes to the prior art all or part may be embodied in a software product, which computer software product may be stored in a storage medium such as ROM / RAM, magnetic disk, optical disk, etc. , including several instructions for instructing a computer device (may be a personal computer, a server, or network device) to perform various method embodiments or some parts of the embodiment of the present invention.

[0100] 以上所揭露的仅为本发明实施例中的较佳实施例而已,当然不能以此来限定本发明之权利范围,因此依本发明权利要求所作的等同变化,仍属本发明所涵盖的范围。 [0100] equivalent variations only preferred embodiments of the present invention, in one embodiment only, but not intended to limit the scope of the present invention, according to the claims made by the claimed invention disclosed above, the present invention still encompassed range.

Claims (12)

  1. 1. 一种分布式推流的方法,其特征在于,包括:推流主控单元接收请求播放消息,并根据所述请求播放消息,获得即将推送的切片; 所述推流主控单元向分布式文件系统询问即将推送的切片的位置信息; 所述推流主控单元接收所述分布式文件系统返回的所述即将推送的切片所在的存储节点信息,根据所述存储节点信息找到对应的推流执行单元; 所述推流主控单元向所述推流执行单元发送推流消息;所述推流执行单元根据所述推流消息,通过内部总线从所述分布式文件系统的存储节点读取所述即将推送的切片数据;所述推流执行单元推送所述即将推送的切片数据。 1. A method of distributed plug flow, characterized by comprising: a master unit receives a request to push stream playback message, and play message according to the request, get pushed upcoming sections; push the flow distribution to the main control unit files coming asked pushed slice position information; storing the information of the node main control unit receives the plug flow distributed file system is about to push the return sections are located, to find the corresponding push information according to the storage node flow execution unit; the main control unit to send a plug flow stream pushed to perform the push message flow; flow execution unit in accordance with the push of the push message stream, the storage node is read from the distributed file system via internal bus the slice data taken soon pushed; slice data stream the execution unit to push the push soon pushed.
  2. 2.如权利要求1所述的方法,其特征在于,在所述推流执行单元从所述缓存中推送数据的同时,还包括:所述推流执行单元向所述推流主控单元发送获取下一个推流执行单元的消息; 所述推流主控单元接收到获取下一个推流执行单元的消息,根据下一个切片信息,向所述分布式文件系统询问下一个切片的位置信息;所述推流主控单元接收所述分布式文件系统返回的所述下一个切片所在的下一个存储节点信息,根据所述下一个存储节点信息找到对应的下一个推流执行单元,并将下一个推流执行单元信息返回所述推流执行单元;当所述推流执行单元完成推送数据后,发送推流消息给所述下一个推流执行单元,所述下一个推流执行单元接收到所述推流消息后,推送下一个切片的数据。 2. The method according to claim 1, wherein, while the pushing unit to push data flow execution from the cache, further comprising: a plug flow execution unit transmits the master unit to the plug-flow obtaining the next message a plug flow execution unit; push the master control unit receives a message stream next execution unit acquires a plug flow, according to the next slice information, inquiry next slice location information to the distributed file system; the plug flow next node storing the master unit receives the information distributed file system returns the next slice location to find a corresponding lower plug flow execution unit next to the storage node according to the information, and the a plug-flow execution unit returns information of the plug-flow execution unit; when the execution unit completes the push push data stream, sends a push message to the next stream plug-flow execution unit, the execution unit next receives the plug flow after the push message stream, the next push of a data slice.
  3. 3.如权利要求2所述的方法,其特征在于,在所述当所述推流执行单元完成推送数据后,发送推流消息给所述下一个推流执行单元,所述下一个推流执行单元接收到所述推流消息开始推送下一个切片的数据之前,还包括:所述推流执行单元在完成推送数据前,向所述下一个推流执行单元发送准备推流消息,所述准备推流消息包括下一个切片信息、推送码流大小、推流起始和终止位置、IP正交幅度调制器地址;所述下一个推流执行单元接收到所述准备推流消息后,从所述分布式文件系统读取下一个切片的数据到缓存。 3. The method according to claim 2, wherein, after said plug flow when the execution unit completes pushing data, sends a push message to the next stream plug-flow execution unit, the next plug flow before the execution unit receives a data message at the beginning of one slice to push the plug-flow, further comprising: a plug flow before completion of the execution unit to push data to the execution unit next transmits a plug flow stream message ready to push the ready to push message flow slice information includes a next push stream size, plug flow start and end positions, the IP address of the quadrature amplitude modulator; push the next flow execution unit after receiving the message stream ready to push from the distributed file system reads data of the next slice to the cache.
  4. 4.如权利要求2所述的方法,其特征在于,在所述下一个推流执行单元接收到所述推流消息的步骤之后,推送下一个切片的数据的步骤之前,还包括步骤:判断延时时间是否到达预设的延时时间。 4. The method according to claim 2, characterized in that, after the plug flow execution unit next to the step of receiving the push message stream, a step of pushing the data slice, the method further comprising the step of: determining delay time is preset delay time is reached.
  5. 5.如权利要求1所述的方法,其特征在于,所述根据所述存储节点信息找到对应的推流执行单元还包括:所述推流主控单元根据负载均衡策略,选择其中一个推流执行单元。 5. The method according to claim 1, wherein said plug flow to find the corresponding execution unit further comprises a storage node according to the information: the master control unit push stream according to the load balancing policy, select a plug flow execution unit.
  6. 6.如权利要求1所述的方法,其特征在于,还包括:所述推流消息包括即将推送的切片信息、推送码流大小、推流起始和终止位置、IP正交幅度调制器地址。 6. The method according to claim 1, characterized in that, further comprising: the push message includes information stream coming slice pushed, the push stream size, plug flow start and end positions, the IP address of the quadrature amplitude modulator .
  7. 7. 一种视频推流装置,其特征在于,包括:推流主控单元,用于接收请求播放消息,并根据所述请求播放消息,获得即将推送的切片,向分布式文件系统询问所述即将推送的切片的位置信息,并接收所述分布式文件系统返回的所述即将推送的切片所在的存储节点信息,根据所述存储节点信息找到对应的推流执行单元,并向对应的推流执行单元发送推流消息;推流执行单元,设置在分布式文件系统的存储节点中,与所述推流主控单元通过网络相连,与所述分布式文件系统的存储节点通过系统内部总线相连,接收所述推流消息,并根据所述推流消息,通过内部总线从所述分布式文件系统读取即将推送的切片的数据,推送所述即将推送的切片的数据。 A push video stream apparatus, comprising: pushing the flow control unit, for receiving a request message playback, and play message according to the request, get pushed upcoming sections, to interrogate the distributed file system information storage node is about to push the slice position information, and receiving a return of the distributed file system is about to push the sections where to find the corresponding plug flow execution unit according to the storage node information corresponding to the plug flow plug flow execution unit transmits messages; plug flow execution unit provided in the storage node of a distributed file system, connected to the main control unit via a plug flow network connected to the storage node of the distributed file system via the internal bus system , receiving the data stream push message, and the message according to the plug-flow, read from the slice soon pushed through the internal bus distributed file system, the data is about to push the push slice.
  8. 8.如权利要求7所述的装置,其特征在于,还包括:所述推流执行单元,还用于向所述推流主控单元发送获取下一个推流执行单元的消息;所述推流主控单元,还用于接收到获取下一个推流执行单元的消息,根据下一个切片信息,向所述分布式文件系统询问下一个切片的位置信息,接收所述分布式文件系统返回的所述下一个切片所在的下一个存储节点信息,根据所述下一个存储节点信息找到对应的下一个推流执行单元,并将下一个推流执行单元信息返回所述推流执行单元;所述推流执行单元,还用于当完成推送数据后,发送推流消息给所述下一个推流执行单元,所述下一个推流执行单元接收到所述推流消息开始推送下一个切片的数据。 8. The apparatus according to claim 7, characterized in that, further comprising: a plug flow execution unit is further for transmitting the acquired message to the master unit a plug-flow plug flow execution unit; said push the flow control unit, is further configured to acquire the next message is received a plug flow execution unit, a slice according to the following information, to the location information of the distributed file system under examination a slice, to receive the distributed file system returns the next next slice where a storage node information to find the corresponding lower flow execution unit in accordance with a push of the lower storage node information, and the next push of the return flow execution unit to push information flow execution unit; the plug flow execution unit is further configured to push the data when complete, sends a push message to the streaming data flow execution unit to push the next, the next plug flow execution unit receives the message begins to push a plug flow of said slice .
  9. 9.如权利要求8所述的装置,其特征在于,还包括:所述推流执行单元,还用于在完成推送数据前,向所述下一个推流执行单元发送准备推流消息,所述准备推流消息包括下一个切片信息、推送码流大小、推流起始和终止位置、 IP正交幅度调制器地址;所述下一个推流执行单元,还用于接收到所述准备推流消息后,从所述分布式文件系统读取下一个切片的数据到缓存。 9. The apparatus according to claim 8, characterized in that, further comprising: a flow execution unit to push the further configured to push data before completion, to the next unit transmits a stream ready to push push message stream execution, the said setup message includes a lower plug flow slice information, push stream size, plug flow start and end positions, the IP address of the quadrature amplitude modulator; push the next flow execution unit is further configured to receive the ready to push after the message stream, the read data from the next slice to the distributed file system cache.
  10. 10.如权利要求8所述的装置,其特征在于,还包括:所述下一个推流执行单元,还用于接收到所述推流消息后,判断延时时间是否到达预设的延时时间,若到达预设的延时时间,则推送下一个切片的数据。 10. The apparatus according to claim 8, characterized in that, further comprising: pushing the next flow execution unit is further configured to receive the message after the plug flow, it is determined whether the delay time reaches a preset time delay time, when the time reaches a preset time delay, the next slice data is pushed.
  11. 11.如权利要求7所述的装置,其特征在于,所述根据所述存储节点信息找到对应的推流执行单元还包括:所述推流主控单元根据负载均衡策略,选择其中一个推流执行单元。 11. The apparatus according to claim 7, wherein said plug flow to find the corresponding execution unit further comprises a storage node according to the information: the master control unit push stream according to the load balancing policy, select a plug flow execution unit.
  12. 12. —种分布式推流的系统,其特征在于,包括视频推流装置和分布式文件系统:所述视频推流的装置,包括:推流主控单元,用于接收请求播放消息,并根据所述请求播放消息,获得即将推送的切片,向所述分布式文件系统询问所述即将推送的切片的位置信息,并接收所述分布式文件系统返回的所述即将推送的切片所在的存储节点信息,根据所述存储节点信息找到对应的推流执行单元,并向对应的推流执行单元发送推流消息;推流执行单元,与推流主控单元相连,并通过内部总线与所述分布式文件系统相连,接收所述推流消息,并根据所述推流消息,通过内部总线从所述分布式文件系统读取即将推送的切片数据,推送所述即将推送的切片数据;分布式文件系统,与所述视频推流装置相连,用于接收询问所述即将推送的切片的位置信息的消息,并根 12. - A distributed plug flow system, wherein the apparatus includes a video stream and pushing a distributed file system: the pushing means video stream, comprising: a plug flow control unit, for receiving a request message playback, and storage sections where the player according to the request message, about to push the slice obtained, location information of the inquiry is about to push the slices distributed file system, and receiving the distributed file system is about to push the return node information, to find the storage node information corresponding to plug flow execution unit, and sends the push message stream corresponding to plug flow execution unit; plug flow execution unit, the main control unit is connected to the plug flow, and through the internal bus and the distributed file system is connected to receive the push message flow and message flow according to the push, to push the read slice data coming from the internal bus via the distributed file system, said slice data about to push pushed; distributed file system, connected to the video stream pushing means for receiving a location information query message is about to push the slice, and the root 据所述即将推送的切片的信息获得该即将推送的切片所在的存储节点信息,并将所述存储节点信息返回给所述视频推流装置。 According to the information obtained about to push the slice information storage node is about to push the slice is located, and the storage node information returned to the video stream pushing means.
CN 201110451761 2011-12-29 2011-12-29 Distributed stream pushing method, device and system CN102547385A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN 201110451761 CN102547385A (en) 2011-12-29 2011-12-29 Distributed stream pushing method, device and system

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN 201110451761 CN102547385A (en) 2011-12-29 2011-12-29 Distributed stream pushing method, device and system

Publications (1)

Publication Number Publication Date
CN102547385A true true CN102547385A (en) 2012-07-04

Family

ID=46353141

Family Applications (1)

Application Number Title Priority Date Filing Date
CN 201110451761 CN102547385A (en) 2011-12-29 2011-12-29 Distributed stream pushing method, device and system

Country Status (1)

Country Link
CN (1) CN102547385A (en)

Citations (9)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101141627A (en) * 2007-10-23 2008-03-12 深圳市迅雷网络技术有限公司 Storage system and method of stream media file
CN101242520A (en) * 2007-01-12 2008-08-13 博路技术有限公司 Data distribution and buffering
CN101552800A (en) * 2008-04-01 2009-10-07 华为技术有限公司 Method, device and system for providing media content
KR20100055297A (en) * 2008-11-17 2010-05-26 에스케이텔레콤 주식회사 System and method for simultaneous multimedia streaming using redirected url of distributed contents
US20100242079A1 (en) * 2009-03-18 2010-09-23 Steven Riedl Apparatus and methods for network video recording
US20110083149A1 (en) * 2009-10-06 2011-04-07 At&T Intellectual Property I, L.P. Apparatus and method for providing media content
CN102203758A (en) * 2008-09-19 2011-09-28 杰出网络公司 Content delivery network stream server vignette distribution
US20110246616A1 (en) * 2010-04-02 2011-10-06 Ronca David R Dynamic Virtual Chunking of Streaming Media Content
CN102232298A (en) * 2011-04-07 2011-11-02 华为技术有限公司 Method, device and system for transmitting and processing media content

Patent Citations (9)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101242520A (en) * 2007-01-12 2008-08-13 博路技术有限公司 Data distribution and buffering
CN101141627A (en) * 2007-10-23 2008-03-12 深圳市迅雷网络技术有限公司 Storage system and method of stream media file
CN101552800A (en) * 2008-04-01 2009-10-07 华为技术有限公司 Method, device and system for providing media content
CN102203758A (en) * 2008-09-19 2011-09-28 杰出网络公司 Content delivery network stream server vignette distribution
KR20100055297A (en) * 2008-11-17 2010-05-26 에스케이텔레콤 주식회사 System and method for simultaneous multimedia streaming using redirected url of distributed contents
US20100242079A1 (en) * 2009-03-18 2010-09-23 Steven Riedl Apparatus and methods for network video recording
US20110083149A1 (en) * 2009-10-06 2011-04-07 At&T Intellectual Property I, L.P. Apparatus and method for providing media content
US20110246616A1 (en) * 2010-04-02 2011-10-06 Ronca David R Dynamic Virtual Chunking of Streaming Media Content
CN102232298A (en) * 2011-04-07 2011-11-02 华为技术有限公司 Method, device and system for transmitting and processing media content

Similar Documents

Publication Publication Date Title
US20110307534A1 (en) Distributed file system supporting data block dispatching and file processing method thereof
US20130111028A1 (en) Method and apparatus for selecting an access method for delivery of media
US20060242664A1 (en) Content providing server, information processing device and method, and computer program
US20050034164A1 (en) Network AV system
US20110246661A1 (en) Streaming playback and dynamic Ad insertion
US20060212531A1 (en) Content providing server, information processing device and method, and computer program
US20130145016A1 (en) Methods and apparatuses for domain management
US20060195545A1 (en) Information processing apparatus and content information processing method
US20120284765A1 (en) Sharing of Subscriber-Recorded Digital Video Recorder Content
US20100011088A1 (en) System and Method for Combining Pull and Push Modes
US20120159542A1 (en) Current Device Location Advertisement Distribution
US20050055718A1 (en) Peer-to-peer architecture for sharing video on demand content
US20120141092A1 (en) Hierarchical Storage Management for Data
JP2008021293A (en) Content management method and apparatus
CN101964787A (en) Method, device and system for implementation of multiple-terminal breakpoint broadcast of programs
US20140165119A1 (en) Offline download method, multimedia file download method and system thereof
CN102572555A (en) Method and system for realizing live video playback at HTTP live streaming (HLS) client
CN102724196A (en) Method and device for playing media resource data
CN103747112A (en) CDN back-to-source method, first server and system thereof
CN102638443A (en) Method and system for sharing streaming media resources, and device with DLNA (Digital Living Network Alliance) function
US20090172742A1 (en) Systems, methods, and computer products for recording of repeated programs
JP2005020180A (en) Information processor, information processing method and computer program
US20120084821A1 (en) Demand-Based Edge Caching Video Conten System and Method
CN101645928A (en) Content resource caching method, device and system
KR20080003961A (en) System and method for moving picture contents service

Legal Events

Date Code Title Description
C06 Publication
C41 Transfer of patent application or patent right or utility model
ASS Succession or assignment of patent right

Owner name: SHENZHEN LONGSHI MEDIA CO., LTD.

Free format text: FORMER OWNER: SHENZHEN COSHIP VIDEO COMMUNICATION CO., LTD.

Effective date: 20130423

ASS Succession or assignment of patent right

Owner name: SHENZHEN TONGZHOU ELECTRONIC CO., LTD.

Free format text: FORMER OWNER: SHENZHEN LONGSHI MEDIA CO., LTD.

Effective date: 20140508

C41 Transfer of patent application or patent right or utility model
C10 Entry into substantive examination
C41 Transfer of patent application or patent right or utility model
WD01