KR102024637B1 - Open platform for live broadcast - Google Patents

Abstract

It provides open live platform technology. The system for the open live platform according to the embodiments of the present invention may dynamically allocate resources at the time when the broadcast is actually broadcast through the logical streaming cloud.

Description

OPEN PLATFORM FOR LIVE BROADCAST}

The following description relates to an open live platform technology, and more particularly, to a system for an open live platform and a method of operating the open live platform.

Live broadcasting is to broadcast a live broadcast simultaneously, for example, Korean Patent Laid-Open No. 10-2014-0115661 discloses a method, a server, and a system for providing an interactive live broadcast service to a plurality of user terminals.

In addition, there is a live broadcast platform for relaying various broadcasters who want to broadcast broadcasts and various viewers who want to watch live broadcasts of broadcasters, such as a personal broadcast live service, in which anyone can broadcast a video as a broadcast station.

In such a live broadcast platform of the prior art, each sender transmits its own streaming data to the relay server assigned to the sender, the assigned relay server can encode the received streaming data and send it to the viewers. To this end, in a live relay platform, when a broadcaster registers a broadcast, the broadcast server must allocate a relay server for the broadcaster, and provide the broadcaster with a broadcast address for the assigned relay server. In this case, the relay server must be prepared with resources allocated to process broadcast transmission regardless of transmission. However, since the time at which the broadcasters register the broadcast and occupy the resources of the relay server does not coincide with the broadcast time of the broadcasters, the resources of the relay server to which the broadcast is registered are not displayed even while the broadcast is not being broadcast. The resources allocated for the broadcast are wasted. For example, assuming that 10 relay servers can be allocated for registration of 3 broadcasts, the number of broadcasts that can be registered is fixed to 30. At this time, even if all 30 broadcasts are not transmitting broadcasts, a new relay server is required to register a new broadcast.

In addition, since information on broadcasts allocated to the relay servers must be continuously maintained separately from the center, there is a problem in that the burden for managing the entire system is large.

Such a live broadcast platform of the prior art is not suitable for an open live broadcast service in which arbitrary broadcasts can be generated, not a fixed number of broadcasts, and it is expensive because relay servers are inefficiently used even if a service is provided. There is a problem.

In addition, such a live broadcast platform of the prior art has a problem that it is difficult to cope with the failure of the relay server. For example, when a failure occurs in the relay server 1, all broadcasts allocated to the relay server 1 are not transmitted. If the failure of the physical machine where the relay server 1 is implemented, the time to resolve the failure will be longer. A complicated process is required such as notifying the sender of the failure of relay server 1, registering the broadcast again, and notifying the sender of the address of the new relay server. As the processing time for recovering the fault becomes longer, There is a possibility that a serious problem of not being able to broadcast a live broadcast in time occurs.

The present invention provides a system of an open live platform that can dynamically allocate resources when a broadcast is actually broadcasted through a logical streaming cloud, and a method of operating the open live platform.

Dynamic and autonomous coordination systems (coordinators) separately and dynamically, both publishing point servers for processing broadcast requests and encoding servers for encoding streaming data for broadcasts, when central broadcasts occur, without centralized control. According to the present invention, a system of an open live platform and an operating method of the open live platform, which can efficiently utilize all servers without having to wait in a reserved state for a specific broadcast in advance, are provided.

In the event of a publishing point server or encoding server failure, a system of an open live broadcast platform capable of automatically failing over by reallocating a new publishing point server or encoding server from a pool of servers and a method of operating the open live platform may be provided.

Implement logical streaming cloud for each country or region divided by region, manage servers of logical streaming cloud for each country through coordination system, and if broadcasts are transmitted and viewed between different zones, proxy The present invention provides a system of an open live platform that can minimize a user section having a slow network speed through acceleration, and a method of operating the open live platform.

A system of an open live platform that can stably and efficiently use a physical machine by dynamically adjusting the number of encoding servers to be operated on one physical machine according to the encoding performance of the physical machine through a coordination system, and a method of operating the open live platform. To provide.

Even for a plurality of broadcasts provided by each of a plurality of different services for streaming, each channel is stably managed without interference from other services by independently managing a channel regardless of the service for each broadcast. The present invention provides a system of an open live broadcast platform capable of providing each of a plurality of broadcasts provided to a viewer of a corresponding service, and a method of operating the open live broadcast platform.

The present invention provides a system of an open live broadcast platform capable of assigning priority to each channel and guaranteeing broadcast transmission of an important channel using the assigned wired priority, and a method of operating the open live broadcast platform.

A system for an open live platform, comprising: a first server pool managing at least one first server for receiving streaming data transmitted from a sender's terminal; A second server pool managing at least one second server for receiving and encoding the streaming data from a first server allocated in the first server pool; And a coordinator for managing states of the servers of the first server pool and the second server pool, and dynamically allocating a second server in the second server pool according to a request of the first server allocated in the first server pool. And in response to the streaming data being transmitted from the sender's terminal, dynamically allocating any first server in the first server pool, and receiving the transmitted streaming data through the assigned first server. It provides a system for an open live platform, characterized in that.

A method of operating an open live platform, the method comprising: receiving an allocation request of a second server for encoding the streaming data from any first server allocated in the first server pool in response to the streaming data being transmitted from the sender terminal. ; And dynamically allocating an arbitrary second server in a second server pool according to the received allocation request, wherein the dynamically allocated second server receives and encodes the streaming data from the first server. It provides a method of operation characterized in that.

Provided is a computer-readable recording medium, in which a program for causing the computer to execute the operation method is recorded.

Logical streaming clouds allow you to dynamically allocate resources when broadcasts actually occur.

Dynamic and autonomous coordination systems (coordinators) separately and dynamically, both publishing point servers for processing broadcast requests and encoding servers for encoding streaming data for broadcasts, when central broadcasts occur, without centralized control. By allocating, all servers can be efficiently utilized without having to wait in advance for a specific broadcast.

In the event of a publishing point server or encoding server failure, the failed server can be recovered automatically by reassigning a new publishing point server or encoding server from the pool of servers.

Implement logical streaming cloud for each country or region divided by region, manage servers of logical streaming cloud for each country through coordination system, and if broadcasts are transmitted and viewed between different zones, proxy Acceleration minimizes user segments on slower networks.

The coordination system allows the physical machine to be used stably and efficiently by dynamically adjusting the number of encoding servers to be operated on one physical machine according to the encoding performance of the physical machine.

Even for a plurality of broadcasts provided by each of a plurality of different services for streaming, each channel is stably managed without interference from other services by independently managing a channel regardless of the service for each broadcast. Each of the plurality of broadcasts provided by may be provided to viewers of the corresponding service.

Priority can be given for each channel, and the broadcast transmission of important channels can be guaranteed using the assigned wired priority.

1 is a view showing an example of the overall configuration of an open live broadcast platform according to an embodiment of the present invention.
2 is a diagram illustrating an example of a configuration of a live streaming farm according to an embodiment of the present invention.
3 is a diagram illustrating an example of load balancing of an L4 switch according to an embodiment of the present invention.
4 is a diagram illustrating an example of a configuration of an encoder server pool according to an embodiment of the present invention.
5 is a diagram illustrating an example of a configuration of a VOD farm according to an embodiment of the present invention.
6 to 8 are diagrams showing an example of a method of operating an open live broadcast platform according to an embodiment of the present invention.
9 is a diagram illustrating an example of a broadcast address and a viewing address for live broadcasting in an embodiment of the present invention.
10 illustrates an example of a process of processing a broadcast transmission request according to an embodiment of the present invention.
11 is a diagram illustrating an example of an encoding process according to an embodiment of the present invention.
12 is a diagram for one example of a process of dynamically allocating each server when a transmission occurs in one embodiment of the present invention.
13 and 14 illustrate examples of an autonomous failover process for a publishing point server according to an exemplary embodiment of the present invention.
15 and 16 illustrate an example of a process of autonomous failover for an encoder server according to an embodiment of the present invention.
FIG. 17 is a diagram for one embodiment of an open live platform according to a zone according to one embodiment of the present invention; FIG.
18 is a diagram for explaining an example of proxy acceleration according to an embodiment of the present invention.
FIG. 19 is a diagram for explaining the concept of a multi-tenant of an open live platform according to an embodiment of the present invention.
20 is a flowchart illustrating an example of a method of operating an open live broadcast platform according to an embodiment of the present invention.
21 is a flowchart illustrating an example of a method for failback of an open live relay platform according to an embodiment of the present invention.

Hereinafter, exemplary embodiments will be described in detail with reference to the accompanying drawings.

1 is a view showing an example of the overall configuration of an open live broadcast platform according to an embodiment of the present invention, Figure 2 is a view showing an example of the configuration of a live streaming farm according to an embodiment of the present invention, Figure 3 is a diagram illustrating an example of load balancing of an L4 switch according to an embodiment of the present invention, and FIG. 4 is a diagram illustrating an example of a configuration of an encoder server pool according to an embodiment of the present invention. 5 is a diagram illustrating an example of a configuration of a VOD farm according to an embodiment of the present invention.

As shown in FIG. 1, the open live platform 100 according to the present embodiment includes a live streaming farm 110, a coordinator 120, and a video on demand farm 130. can do.

The live streaming farm 110 may serve to relay the streaming data transmitted from the sender 140 to the viewer 150. To this end, as shown in FIG. 2, the live streaming farm 110 includes a first L4 switch 210, a publishing point server pool 220, an encoder server pool 230, The distributed file system 240 may include a distributed file system 240, a proxy server pool 250, and a second L4 switch 260. In FIG. 1, one sender 140 and one viewer 150 are shown. However, the sender 140 and the viewer 150 may mean a plurality of senders and a plurality of viewers. May mean a terminal of a sender for transmitting a broadcast and a terminal of a viewer receiving a broadcast.

As illustrated in FIG. 3, the first L4 switch 210 may be used for load balancing the plurality of publishing point servers 310 included in the publishing point server pool 220. For example, the first L4 switch 210 may determine which publishing point server of the plurality of publishing point servers 310 delivers the streaming data transmitted from the sender 140.

As described with reference to FIG. 3, the publishing point server pool 220 is a pool of the plurality of publishing point servers 310, and each publishing point server is transmitted through the first L4 switch 210. The transmission request of the 140 may be processed, and streaming data may be delivered to an encoder server included in the encoder server pool 230. At this time, one publishing point server may process a plurality of transmission requests. In addition, one publishing point server may be implemented as one physical machine, but a plurality of publishing point servers may be implemented on one physical machine. For example, if the three physical machines are implemented to include three publishing point servers, a total of nine publishing point servers may be included in the publishing point server pool 220. In FIG. 3, the publishing point server pool 220 includes three publishing point servers. However, the publishing point server may be freely added to the publishing point server pool 220 or the publishing point server pool 220 may be used. May be excluded.

Encoder server pool 230 may be a pool including a plurality of encoder servers 410 as shown in FIG. Although the plurality of encoder servers 410 shows an example including three encoder servers as shown in FIG. 4, the encoder server may be freely added to the encoder server pool 230 or the encoder server pool 230 as necessary. ) May be excluded. In addition, each of the plurality of encoder servers 410 may be implemented as one physical machine as described above, but may be implemented to include a plurality of encoder servers on one physical machine. For example, if the three physical machines are implemented to include four encoder servers each, a total of 12 encoder servers may be included in the encoder server pool 220.

In addition, one physical machine may be implemented to include at least one publishing point server and at least one encoder server. In other words, the publishing point server, the encoder server, and the proxy server to be described later may be implemented in the form of individual physical machines, or may be implemented in the form of a plurality of servers included on one physical machine. .

Each of the plurality of encoder servers 410 may be implemented as a pair of one encoder server and one uploader. For example, FIG. 4 illustrates an example in which one encoder server is implemented as a pair of encoder server 1 411 and uploader 1 412.

The encoder server may encode streaming data received from the publishing point server into a preset profile. For example, when providing a live broadcast to the viewer 150 using an HTTP Live Streaming (HLS) algorithm, the encoder server may generate a TS (Transport Segment) file by encoding streaming data received from the publishing point server. have.

The uploader uploads the TS files resulting from the encoding to the distributed file system 240 included in the live streaming farm 110 and the distributed file system 520 included in the VOD farm 130 shown in FIG. can do. In addition, the uploader generates playback information files (e.g., m3u8 files in the HTTP Live Streaming (HLS) algorithm) of the playback information files (for example, information on what media and how to play them) according to each service type. It may be generated and uploaded to the distributed file system 240 included in the live streaming farm 110. In this case, the playback information file according to the service type may include a playback information file (for example, an m3u8 file for a time machine) for providing a time machine function, a playback information file (for example, a m3u8 file for a live) and / or a VOD broadcast for live broadcasting. May include a playback information file (for example, an m3u8 file for VOD).

The distributed file system 240 may store and provide encoded files (eg, TS files) and playback information files (eg, m3u8 files) for streaming to the viewer 150.

The proxy server pool 250 may include a plurality of proxy servers, and the proxy server selected through the second L4 switch 260 receives an external streaming request and delivers it to the distributed file system 240. can do. In this case, the distributed file system 240 advances the stored files (encoded files (e.g., TS files) and playback information files (e.g., m3u8 files)) according to the external streaming request delivered. By providing through the selected proxy server, a live broadcast service may be provided to the viewer 150.

The coordinator 120 may manage the state of the servers included in the open live platform 100, allocate an encoder server, and manage the overall state and settings of the open live platform 100, such as information on a channel currently being transmitted. . For example, the servers included in the open live platform 100 may undergo a process of registering themselves with the coordinator 120 when executed, and the coordinator 120 may manage the registered servers. In addition, the coordinator 120 may allocate an encoder server according to a request of the publishing point server. The coordinator 120 may be implemented using, for example, a ZOOkeeper system, which is a framework that provides auxiliary functions such as information sharing, locks, and events among nodes in a distributed environment.

The VOD farm 130 may be responsible for generating and storing a VOD file of a live broadcast image and providing a VOD service. As shown in FIG. 5, the VOD farm 130, a distributed file system 520, It may include a proxy server pool 530 and a third L4 switch 540.

When the VOD generation pool 510 receives a VOD file generation request, the VOD generation pool 510 may download encoded files (eg, TS files) from the distributed file system 520 and generate a VOD file (eg, an MP4 file). have. In addition, the VOD generation pool 510 may upload the generated VOD file to the distributed file system 520 again. The VOD generation pool 510 may include a plurality of VOD generation servers.

The distributed file system 520 may store and provide encoded files for the VOD file and a VOD file generated by the VOD generation pool 510.

The proxy server pool 530 may include a plurality of proxy servers. When an external VOD file request is received, the third L4 switch 540 may select one proxy server among a plurality of proxy servers included in the proxy server pool 530 and transmit the external VOD file request. The proxy server may forward an external VOD file request to the distributed file system 520. In this case, the distributed file system 520 may provide the requested VOD file to the outside through a proxy server.

6 to 8 are diagrams showing an example of a method of operating an open live broadcast platform according to an embodiment of the present invention.

6 illustrates steps 1-1 to 1-6 of the entire processes (steps 1-1 to 1-10) for live streaming.

Process 1-1 may be an example of a process in which the sender 140 transmits streaming data to the first L4 switch 210.

Process (1-2) in which the first L4 switch 210 selects any publishing point server 610 included in the publishing point server pool 220, and forwards the transmission request to the selected publishing point server 610 It may be an example of a process.

Process 1-3 may be an example of a process in which the publishing point server 610 requests the coordinator 120 to assign any one of the encoder servers included in the encoder server pool 230 for streaming.

Processes 1-4 include that the coordinator 120 selects an arbitrary encoder server 620 to be used for streaming in the encoder server pool 230 and selects the encoder information selected for the publishing point server 610 together with the streaming start event. It may be an example of a process of transferring to the server 620.

Process (1-5) may be an example of a process in which the encoder server 620 selected by the coordinator 120 accesses the publishing point server 610 to receive streaming data and encode the received streaming data into a preset profile. have.

Steps 1-6 are steps in which the encoder server 620 stores the encoded data (eg, TS files generated according to the encoding) in a designated path (eg, temporary storage 630 of FIG. 6). It may be an example.

7 illustrates steps 1-7 to 1-10 of the entire processes (steps 1-1 to 1-10) for live streaming. Processes 1-6 and 1-7 may be connected via (A) shown in FIGS. 6 and 7.

In steps 1-7, the uploader 710 obtains encoded data (TS files) from the temporary storage 630 shown in FIG. 6, and reproduces a reproduction information file (for example, m3u8) corresponding to the encoded data. Files). As described above, the uploader 710 may be paired with the encoder server 620.

Processes 1-8 upload the uploader 710 encoded data to the distributed file system 240 of the live streaming farm 110 and the distributed file system 520 of the VOD farm 130, respectively. The reproduction information file may be an example of a process of uploading to the distributed file system 240 of the live streaming farm 110. In this case, the process of uploading the encoded data to the distributed file system 520 of the VOD farm 130 may be connected through (C) illustrated in FIGS. 7 and 8.

Process (1-9) is a data and playback information file encoded by the proxy server 720, one of the proxy servers included in the proxy server pool 250 from the distributed file system 240 of the live streaming farm 110 This may be an example of the process of requesting and receiving.

Process 1-10 may be an example of a process of performing live streaming by delivering the encoded data and the reproduction information file received from the proxy server 720 to the viewer 150.

8 illustrates the entire processes (processes 2-1 to 2-5) for generating a VOD file and providing a VOD service. Such FIG. 8 may be connected to FIGS. 6 and 7 through (B) and (C).

Process 2-1 may be an example of a process in which the VOD generation server 810, which is one of the VOD generation servers included in the VOD generation pool 510, receives a VOD generation request from the coordinator 120.

Process 2-2 may be an example of a process in which the VOD generation server 810 downloads encoded data to generate a VOD file from the distributed file system 520 included in the VOD farm 130. The encoded data may be data stored in the distributed file system 520 by the uploader 710 through steps 1-8 of FIG. 7.

Process 2-3 may be an example of a process of generating a VOD file using the downloaded data by the VOD generation server 810 and uploading it to the distributed file system 520.

Step 2-4 is a process in which the proxy server 820, which is one of the proxy servers included in the proxy server pool 530, downloads the VOD file from the distributed file system 520 at the request of the VOD requester 830. It may be an example of.

Process 2-5 may be an example of a process in which the proxy server 820 provides the downloaded VOD file to the VOD requester 830.

Through these processes 2-1 to 2-5, a VOD file for live streamed broadcast may be generated and provided.

The open live broadcast system according to the embodiments of the present invention has the features of 1 to 6 as follows.

1. Logical Streaming cloud (Logical Streaming Cloud)

In order to broadcast a broadcast using the open live broadcast system 100 according to embodiments of the present invention, the sender 140 must register a broadcast. In this case, when a broadcast is registered, one channel may be generated, and the generated channel may be provided with a stream key necessary for broadcasting the broadcast. The sender 140 may always broadcast the same address using the stream key, and the open live system 100 may always provide the viewer 150 with the broadcast of the corresponding channel at the same address. In other words, the sender 140 can always broadcast the same address, the viewer 150 can always watch a particular broadcast of the sender 140 at the same address.

9 is a diagram illustrating an example of a broadcast address and a viewing address for live broadcasting in an embodiment of the present invention. 9 shows a sender address' rtmp: // generated using the stream key 'xyzc123' assigned to the sender 140 when the sender 140 transmits the broadcast using the Real Time Messaging Protocol (RTMP). An example of broadcasting a broadcast to tv.bbb.com/xyzc123 'is shown. In addition, FIG. 9 shows a viewer address 'http: // streamfarm.' Generated using the name 'aaa' of the sender 140 when the viewer 150 watches a broadcast using the HLS (HTTP Live Streaming) protocol. An example of watching the broadcast at bbb.com/aaa.m3u8 'is shown. Accordingly, the viewer 150 may access a specific broadcast in a concept similar to the concept of viewing a broadcast of a specific broadcaster of a TV on a specific channel.

The channel generated at the time of broadcasting registration generates only the information on the broadcast, and is not limited to the number of generations because it is not connected to a physical server, and the increase in the number of generations does not affect the system at all. In other words, even if a channel is created, resources of the open live platform 100 are not allocated for the channel. In fact, when the resources of the open live broadcast platform 100 are used, the broadcast transmission occurs. Such a structure is suitable for an open live service structure in which arbitrary broadcasters register a plurality of broadcasts and have a short actual broadcast time. Do.

2. Dynamic and Autonomous Coordination System

The servers of the open live platform 100 are dynamically allocated at the time when the transmission occurs, and remain in the pool in the pool until the transmission occurs. When a request for transmission occurs, an arbitrary publishing point server is selected from the publishing point server pool 220 by the L4 switch (first L4 switch 210), and the encoding server pool 230 is selected by the coordination 120. The encoder server may be selected according to the encoder server selection algorithm in.

The encoder server selection algorithm is as follows (a), (b), (c).

(a) Select an encoder server on standby among encoder servers on the same local machine as the publishing point server.

(b) If there is no encoder server that reads (a) above, select an encoder server in standby state among the encoder servers in the same region as the publishing point server.

(c) If there is no encoder server that satisfies the above-mentioned (a) or (b), select an encoder server that is broadcasting a channel having a lower priority than a channel that has requested broadcasting. (In this case, the selected encoder server types the job of the channel currently being processed and processes the request for transmission of the newly requested broadcast.)

The open live platform 100 may process a broadcast transmission request autonomously at each step without centralized control.

10 illustrates an example of a process of processing a broadcast transmission request according to an embodiment of the present invention. In the embodiment of FIG. 10, the first L4 switch 210 selects the publishing point server 1 1010 according to the broadcast transmission request from the sender 140, and transmits the broadcast transmission request to the selected publishing point server 1 1010. An example of passing is shown. In addition, the publishing point server 1 1010 selected in the embodiment of FIG. 10 notifies the coordinator 120 of the requested broadcast and confirms whether the requested broadcast is a registered normal broadcast through the coordinator 120, and then sends a caller ( 140 may play a role of receiving the streaming data of the broadcast from the broadcast. As described with reference to FIG. 6, the coordinator 120 may allocate an encoder server for the publishing point server 1 1010 with respect to a broadcast transmission request for a normal broadcast.

11 is a diagram illustrating an example of an encoding process according to an embodiment of the present invention. In the embodiment of FIG. 11, the coordinator 120 selects an encoder server according to a broadcast transmission request for a normal broadcast (in the embodiment of FIG. 11, encoder server 1 1110) and a task list of the selected encoder server 1 1110. An example of registering a broadcast transmission job is shown in FIG. At this time, each of the standby encoder server may monitor its own work list. In other words, the encoder server 1 1110 monitors its own task list 'Encoder server 1 task list' 1120 and uses the registered information as a broadcast transmission task is registered by the coordinator 120. The streaming data can be received and an encoding operation can be performed on the received streaming data. For example, the encoder server 1 1110 may identify the publishing point server 1 1010 through the registered information, and may request streaming data from the identified publishing point server 1 1010. The publishing point server 1 1010 may transmit the streaming data received from the sender 140 through the first L4 switch 210 to the encoder server 1 1110 according to a request of the encoder server 1 1110. In this case, the encoder server 1 1110 may encode streaming data received from the publishing point server 1 1010.

12 is a diagram for one example of a process of dynamically allocating each server when a transmission occurs in one embodiment of the present invention. Processes ①, process ②, process ③, process ④ and process ⑤ illustrated in FIG. 12 may correspond to the processes (1-1) to (1-5) described above with reference to FIG. 6.

Process ① may be an example of a process in which the sender 140 attempts to broadcast.

Process ② may be an example of a process in which the first L4 switch 210 selects a publishing point server 1210 that is one of the publishing point servers included in the publishing point server pool 220, and transmits a transmission request.

Process ③ may be an example of a process in which the publishing point server 1210 receiving the transmission request requests the coordinator 120 to allocate the encoding server while registering the broadcast transmission request.

Process ④ may be an example of a process in which the coordinator 120 selects an encoder server 1220 which is one of encoder servers included in the encoder server pool 230 and registers streaming request information. The registration of the streaming request information may correspond to registering a broadcast transmission job in the job list of the selected encoder server 1220 as described with reference to FIG. 11.

In the process ⑤, the encoder server 1220 accesses the publishing point server 1210 to request streaming data, receives the requested streaming data from the publishing point server 1210, and encodes the data according to a preset profile such as a set image quality. It may be an example of. At this time, the encoder server 1220 may generate a TS file for each picture quality.

When transmission of the broadcast ends, the assigned publishing point server 1210 and the encoder server 1220 wait in the publishing point server pool 220 and the encoder server pool 230 in a standby state which is available again.

As described above, in the embodiments of the present invention, since the publishing point server and the encoder server are dynamically allocated only when the broadcast is actually transmitted, the server for the specific broadcast needs to wait in a pre-reserved state (pre-allocated state). All servers can be used efficiently. In addition, since the capacity of the open live platform 100 can be increased simply by adding a server to each server pool, the capacity of the open live platform 100 can be easily increased.

3. Automatic failover mechanism failover  mechanism)

In the open live platform 100, the publishing point server and the encoder server are separated from each other, and since a server is not associated with a specific channel, it is possible to quickly recover from abnormal operation.

13 and 14 illustrate examples of an autonomous failover process for a publishing point server according to an exemplary embodiment of the present invention.

FIG. 13 illustrates processing of a situation in which the publishing point server 1210 described above with reference to FIG. 12 is abnormally terminated. If the publishing point server 1210 is abnormally terminated, the coordinator 120 may detect such an abnormal termination of the publishing point server 1210 and send it to the encoder server 1220 assigned to the publishing point server 1210. You can dispatch the shutdown event. In this case, the encoder server 1220 may be switched back to the standby state for the transmission (streaming) of another broadcast.

FIG. 14 illustrates that when a transmission is retried by the sender 140, another publishing point server 1410 is selected by the first L4 switch 210, and a coordinator (eg, at the request of another publishing point server 1410) is selected. As 120 allocates a new encoder server 1420, the streaming point is resumed by the publishing point server 1410 and the encoder server 1420. In other words, when a failure occurs in the existing publishing point server 1210, a failure recovery may be performed by performing the same process as the initial transmission request according to the transmission retry.

15 and 16 illustrate an example of a process of autonomous failover for an encoder server according to an embodiment of the present invention.

FIG. 15 illustrates processing of a situation in which the encoder server 1220 described above with reference to FIG. 12 is abnormally terminated. When the encoder server 1220 abnormally terminates, the coordinator 120 may detect abnormal termination of the encoder server 1220, and attempts to allocate another encoder server currently assignable.

16 shows that as the coordinator 120 assigns a new encoder server 1420 to the publishing point server 1210, transmission is resumed through the publishing point server 1210 and the new encoder server 1420, so that there is no obstacle to broadcast transmission. It can be recovered.

In such embodiments, since the streaming method between the publishing point server and the encoder server is a method in which the assigned encoder server accesses the corresponding publishing point server to receive streaming data (PULL method), a failure situation of the encoder server in the publishing point server There is no need to respond to this, and no operation such as retry is required.

However, when the server fails over, there is a need to add an operation for processing the player. For example, when failover of the publishing point server or the encoder server occurs, streaming data provided to the viewer 150 is not continuous, so that the player can play without stopping through the processing. For this purpose, the uploader (for example, the uploader paired with the encoder server 1420 in FIG. 16) first checks the existence of the m3u8 file for the channel before generating the m3u8 file, and then has already m3u8 file. If there exists, add '# EXT ※ DISCONTINUITY' according to HLS protocol and attach TS list to already existing m3u8 file. If '# EXT ※ DISCONTINUITY' exists in the m3u8 file, the player detects that it is not continuous with the previous streaming data, and processes the data afterwards so that playback does not occur even if a server failover occurs.

4. Basically global platform

The open live platform 100 may divide a country or a region unit into regions and configure live streaming farms according to the regions.

FIG. 17 is a diagram for one embodiment of an open live platform according to a zone according to one embodiment of the present invention; FIG. 17 illustrates a first live streaming farm 1710 and a first VOD farm 1720 implemented in a zone of a first country, and a second live streaming farm 1730 and a second VOD farm implemented in a zone of a second country. 1720 illustrates an example in which the coordinator 120 is connected to each other. In this case, broadcasting and viewing can be performed in different zones. In this case, the concept of proxy acceleration may be used for fast broadcast streaming. Proxy acceleration is a method of minimizing a user section having a slow network speed and providing a streaming service using a fast server-to-server line when different transmission and viewing areas are different.

18 is a diagram for explaining an example of proxy acceleration according to an embodiment of the present invention. 18 illustrates broadcasting of a sender 1810 of a first country through a publishing point server pool 1820, an encoder server pool 1830, a distributed file system 1840, and a proxy server 1850 implemented in a first country. An example of providing the first viewer 1860 located in the first country is shown. In this case, as illustrated in FIG. 18, a white arrow represents a user network section, a gray arrow represents a network section between servers, and a black arrow represents a server dedicated network section between regions.

Since the first viewer 1860 is located in the first country, fast broadcast streaming may also be provided through the proxy server 1850. On the other hand, when the second viewer 1870 located in the second country is provided with broadcast streaming through the proxy server 1850, it is difficult to receive fast broadcast streaming because a user section having a slow network speed becomes very long. Therefore, in the present embodiment, the proxy server 1850 implemented in the first country quickly transmits the broadcast streaming to the proxy server 1880 implemented in the second country through a dedicated line section for proxy acceleration indicated by a black arrow. The second viewer 1870 located in the second country may be provided with bounce streaming through the proxy server 1880 implemented in the second country. In this case, since a user section having a slow network speed is minimized, broadcast streaming of the sender 1810 of the first country can be quickly provided to the second viewer 1870 located in the second country.

5. Effective machine  Effective machine utilization

In the open live platform 100, the number of encoder servers capable of operating in one physical machine may be set according to the encoding performance of the physical machine in charge of encoding. For example, suppose that one physical machine can simultaneously encode five streaming data with 720p resolution, and only one streaming data with 1080p resolution can be encoded at the same time. In this case, in the open live platform 100, five encoder servers are installed in a corresponding physical machine, and when encoding streaming data having 720p resolution, the encoder servers may be allocated to operate all five installed encoder servers. . If one encoder server encodes streaming data having a 1080p resolution, the other four encoder servers may control not to be allocated by the coordinator 120.

6. Multi- Tenant (Multi-tenant)

The open live platform 100 may be used simultaneously in several different services, and each channel may belong to a specific service category at the time of creation. However, since the open live platform 100 independently manages streaming data regardless of a service belonging to a channel, the open live platform 100 can use the open live platform 100 stably without interference from each other.

FIG. 19 is a diagram for explaining the concept of a multi-tenant of an open live platform according to an embodiment of the present invention. 19 is not a different sender using the same service, but a broadcaster broadcasts broadcasters whose services are different from each other, such as the first service 1910, the second service 1920, and the third service 1930. The case is shown. In this case, the live streaming farm 1940 according to the embodiments of the present invention manages each channel independently of a service (as described above, a transmitter is generated using a stream key assigned to one broadcast). Broadcasts are transmitted through the same address, and viewers watch the broadcast through the same address assigned to view the broadcast.), So that the broadcast can be relayed without interference between services. Accordingly, the first service viewer 1950, the second service viewer 1960, and the third service viewer 1970 may be provided with broadcast streams of respective different services.

7. Priority based Streaming Scheduling

The open live platform 100 may give the channel a priority in order to enable the use of servers efficiently and to transmit important broadcasts at any time. The priority of a channel may be set at the time of channel generation, and under normal circumstances does not affect broadcast transmission. However, when there are no more resources (servers) that can be used for broadcast transmission, it is possible to guarantee the transmission of important broadcasts using the 'priority' assigned to the channel.

For example, if a new broadcast transmission is required while all resources (servers) of the open live platform 100 are currently in use, the open live platform 100 is a channel having a lower priority than the channel that requested the new broadcast transmission. Any channel can be selected to terminate the transmission, and the broadcast transmission of the newly requested channel can be serviced. This operation can ensure the broadcast transmission of important broadcasts even when all resources (servers) of the system are currently in use.

The priority may be divided into a plurality of levels such as 'Super', 'High', 'Low', and 'Zero' in order of high priority, and may be allocated to each channel.

The open live platform 100 according to an embodiment may include a first server pool managing at least one first server for receiving streaming data transmitted from a sender's terminal, and a first server pool allocated from the first server pool. A second server pool that manages at least one second server for receiving and encoding the streaming data, and manages states of the first server pool and servers of the second server pool and is allocated in the first server pool. According to the request of the first server may include a coordinator for dynamically allocating the second server in the second server pool. Here, the sender's terminal may correspond to the sender 140, the first server pool to the publishing point server pool 220, the second server pool to the encoder server pool 230, and the coordinator to the coordinator 120, respectively. have. In this case, the open live platform 100 dynamically allocates any first server from the first server pool in response to the streaming data being transmitted from the sender's terminal, and transmits the first server from the first server pool. Streaming data can be received. The open live platform 100 can allocate servers dynamically at the time of the broadcast transmission of the sender, thereby efficiently utilizing resources.

In another embodiment, when the broadcaster of the sender is registered, the coordinator may generate a channel for the broadcast, and generate a broadcast address for transmitting the streaming data of the broadcast by using a unique key assigned to the channel. The open live platform 100 may dynamically allocate an arbitrary first server in the first server pool in response to the streaming data of the broadcast being transmitted through the transmission address. In this case, streaming data can always be transmitted to the same transmission address for a specific broadcast. Similarly, a service may be provided so that viewers can always watch a specific broadcast through the same viewing address by using a viewing address generated by using information about the sender.

In another embodiment, the coordinator may register a broadcast transmission job in a job list of the allocated second server. In this case, each of the at least one second server managed in the second server pool may monitor its own work list, and the second server whose broadcast transmission work is registered in its work list may include information registered in the work list. By using to identify the first server assigned in the first server pool it is possible to receive and encode the streaming data from the assigned first server.

In another embodiment, the open live platform 100 dynamically selects any first server from the first server pool and is sent from the sender's terminal in response to the streaming data being sent from the sender's terminal. The apparatus may further include a switch configured to transfer streaming data to the selected first server. Such a switch may correspond to the first L4 switch 210 described above.

In another embodiment, each of at least one first server managed in the first server pool and at least one second server managed in the second server pool may be registered with the coordinator when executed. In this case, the coordinator may monitor the state of the registered at least one first server and at least one second server. Such monitoring may be used to detect a failure of an assigned first server or an assigned second server.

In another embodiment, when the coordinator detects a failure of the allocated first server, the coordinator sends a streaming end event to the allocated second server to put the allocated second server into a standby state, and As the streaming data is retransmitted from the sender's terminal, a new second server may be allocated in the second server pool according to a request of another first server allocated in the first server pool. In this case, since the streaming data is received and encoded through another first server and a new second server, it is possible to cope with a failure of the existing first server.

In another embodiment, when the coordinator detects a failure of the allocated second server, the coordinator may allocate a new second server from the second server pool. In this case, the allocated new second server may access the allocated first server, receive the streaming data from the allocated first server, and encode the received streaming data. In this case, since the streaming data is received and encoded through the existing first server and the new assigned second server, it is possible to cope with a failure of the existing first server. As already explained, the publishing point server as the first server does not need to consider the failure or failover of the encoder server as the second server because the new server allocated requests streaming data to the first server in a PULL manner. do.

In another embodiment, the open live platform 100 receives a broadcast viewing request of a viewer terminal and a distributed file system storing streaming data encoded by the dynamically allocated second server and encodes the broadcast file in the distributed file system. The apparatus may further include a third server pool managing at least one third server for providing the streamed data to the viewer terminal. In this case, the distributed file system includes the distributed file system 240 described with reference to FIG. 2, the third server pool with the proxy server pool 250 described with reference to FIG. 2, and the viewer terminal with the viewer 150 described with reference to FIG. 1. It can respond. In this case, in response to the broadcast viewing request of the viewer terminal, an arbitrary third server is dynamically allocated from the third server pool, and the encoded streaming data is transmitted from the distributed file system through the allocated third server. It can be provided to the viewer terminal.

In another embodiment, the open live platform 100 dynamically selects any third server from the third server pool in response to the broadcast viewing request of the viewer terminal to direct the broadcast viewing request to the selected third server. The switch may further include. In this case, the switch may correspond to the second L4 switch 260 described with reference to FIG. 2.

In another embodiment, the first server pool, the second server pool, the distributed file system, and the third server pool may be configured for each region divided according to at least one of a country and a region. This configuration makes it possible to reduce the slow network user intervals for each zone.

In another embodiment, a dedicated line may be connected between third servers in different zones. In this case, the streaming data transmitted from the first zone is delivered to the first server of the first zone and encoded by the second server of the first zone, and the viewer is located in the first zone via the third server of the first zone. It may be provided to. In addition, the streaming data transmitted from the first zone may be further provided to the viewer located in the second zone through the third server of the second zone connected to the third server of the first zone by a dedicated line. Therefore, even when the viewer located in the second zone watches the broadcast transmitted from the first zone, the user section having a slow network speed can be minimized.

In another embodiment, the open live platform 100 may include at least one fourth server for generating a VOD file using streaming data encoded through the allocated second server according to a VOD file generation request from the coordinator. A fifth server pool that manages at least one fifth server for receiving the request for the VOD file from the requester terminal and providing the generated VOD file to the requester terminal according to the request; It may further include. The fourth server pool may correspond to the VOD generation server pool 510 described above, and the fifth server pool may correspond to the proxy server pool 540, respectively.

20 is a flowchart illustrating an example of a method of operating an open live broadcast platform according to an embodiment of the present invention. The operation method according to the present embodiment may be performed by the coordinator 120 included in the open live broadcast platform 100. The coordinator 120 may be implemented by a physical machine including at least one processor and at least one memory.

In step 2010, when the broadcaster's broadcast is registered, the coordinator 120 may generate a channel for broadcasting and generate a broadcast address for broadcasting streaming data by using a unique key assigned to the channel. have.

In step 2020, the coordinator 120 requests the allocation of the second server for encoding the streaming data from any first server allocated in the first server pool in response to the streaming data being sent from the sender terminal through the sending address. Can be received.

In step 2030, the coordinator 120 may dynamically allocate any second server in the second server pool according to the received allocation request. In this case, the dynamically allocated second server may receive and encode streaming data from the first server. To this end, the coordinator 120 may register a broadcast transmission job for streaming data in the job list of any second server in step 2030. Each of the at least one second server managed in the second server pool may monitor its own work list, wherein the second server registered in the work list has information registered in its work list. The first server may be allocated from the first server pool to receive and encode streaming data from the first server.

The encoded streaming data may be stored in the distributed file system, and the encoded streaming data stored in the distributed file system may be downloaded and provided to the viewer terminal from a third server dynamically allocated from the third server pool according to the request of the viewer terminal. Can be.

In operation 2040, the coordinator 120 may allocate an arbitrary fourth server for generating a VOD file in the fourth server pool. In this case, the allocated fourth server may generate a VOD file using the encoded streaming data.

21 is a flowchart illustrating an example of a method for failback of an open live relay platform according to an embodiment of the present invention. The failure recovery method of FIG. 21 may be included in the operation method of FIG. 20, and may be executed in parallel with the operation method of FIG. 20.

In operation 2110, the coordinator 120 may monitor a state of at least one first server managed in the first server pool and at least one second server managed in the second server pool. As described above, the coordinator 120 may manage the states of all servers of the open live platform 100.

In operation 2120, the coordinator 120 may detect a failure of the first server. If the failure of the first server is detected, the coordinator 120 may perform step 2130 if the failure of the first server is not detected. Can be.

In operation 2130, the coordinator 120 may transfer the streaming end event to the allocated second server to switch the allocated second server to the standby state. The second server in the standby state may wait for encoding of another broadcast.

In operation 2140, the coordinator 120 may allocate a new second server from the second server pool according to a request of another first server allocated from the first server pool as streaming data is retransmitted from the sender's terminal. . Optionally, the new second server may be an existing second server previously switched to the standby state, or may be a second server different from the existing second server. After step 2140, the coordinator 120 may perform step 2110 again to monitor the status of the servers.

In operation 2150, the coordinator 120 may detect a failure of the second server. In this case, when the failure of the second server is detected, the coordinator 120 may perform step 2160, and if the failure of the second server is not detected, the coordinator 120 may perform step 2110 again to monitor the status of the servers. .

In operation 2160, the coordinator 120 may allocate a new second server from the second server pool. In this case, the allocated new second server may access the allocated first server to receive the streaming data from the allocated first server and to encode the received streaming data. After step 2160, the coordinator 120 may perform step 2110 again to monitor the status of the servers.

As described above, according to embodiments of the present invention, a resource may be dynamically allocated at the time when a broadcast is actually transmitted through a logical streaming cloud. In addition, through a dynamic and autonomous coordination system (coordinator), a publishing point server for processing the broadcast request and a encoding server for encoding the streaming data of the broadcast separately when a broadcast transmission occurs without centralized control, and Dynamic allocation allows for efficient use of all servers without the need for servers to be reserved for specific broadcasts in advance. In addition, when a publishing point server or encoding server fails, the failed server can be automatically recovered by reassigning a new publishing point server or encoding server from the pool of servers. In addition, a logical streaming cloud can be implemented for each region (region) divided by country or region, and the servers of the logical streaming cloud for each country can be managed through a coordination system. In addition, proxy acceleration can be used to minimize the duration of slow network users. In addition, through the coordination system, the physical machine can be used stably and efficiently by dynamically adjusting the number of encoding servers to be operated on one physical machine according to the encoding performance of the physical machine. In addition, since a plurality of broadcasts provided in each of a plurality of different services for streaming are independently managed regardless of the service for each broadcast, the services are stably different without interference from the different services. Each of the plurality of broadcasts provided in each of the services may be provided to viewers of the corresponding service. In addition, priority can be given to each channel, and the broadcast transmission of important channels can be guaranteed using the assigned wired priority.

The system or apparatus described above may be implemented as a hardware component or a combination of hardware components and software components. For example, the devices and components described in the embodiments are, for example, processors, controllers, arithmetic logic units (ALUs), digital signal processors, microcomputers, field programmable gate arrays (FPGAs). Can be implemented using one or more general purpose or special purpose computers, such as a programmable logic unit (PLU), a microprocessor, or any other device capable of executing and responding to instructions. The processing device may execute an operating system (OS) and one or more software applications running on the operating system. The processing device may also access, store, manipulate, process, and generate data in response to the execution of the software. For convenience of explanation, one processing device may be described as being used, but one of ordinary skill in the art will appreciate that the processing device includes a plurality of processing elements and / or a plurality of types of processing elements. It can be seen that it may include. For example, the processing device may include a plurality of processors or one processor and one controller. In addition, other processing configurations are possible, such as parallel processors.

The software may include a computer program, code, instructions, or a combination of one or more of the above, and configure the processing device to operate as desired, or process it independently or collectively. You can command the device. Software and / or data may be any type of machine, component, physical device, virtual equipment, computer storage medium or device in order to be interpreted by or to provide instructions or data to the processing device. It can be embodied in. The software may be distributed over networked computer systems so that they may be stored or executed in a distributed manner. Software and data may be stored on one or more computer readable media.

The method according to the embodiment may be embodied in the form of program instructions that can be executed by various computer means and recorded in a computer readable medium. The computer readable medium may include program instructions, data files, data structures, etc. alone or in combination. The medium may be to continue to store a computer executable program, or to temporarily store for execution or download. In addition, the medium may be a variety of recording means or storage means in the form of a single or several hardware combined, not limited to a medium directly connected to any computer system, it may be distributed on the network. Examples of the medium include magnetic media such as hard disks, floppy disks and magnetic tape, optical recording media such as CD-ROMs and DVDs, magneto-optical media such as floptical disks, And ROM, RAM, flash memory, and the like, configured to store program instructions. In addition, examples of another medium may include a recording medium or a storage medium managed by an app store that distributes an application, a site that supplies or distributes various software, a server, or the like. Examples of program instructions include not only machine code generated by a compiler, but also high-level language code that can be executed by a computer using an interpreter or the like.

Although the embodiments have been described by the limited embodiments and the drawings as described above, various modifications and variations are possible to those skilled in the art from the above description. For example, the described techniques may be performed in a different order than the described method, and / or components of the described systems, structures, devices, circuits, etc. may be combined or combined in a different form than the described method, or other components Or even if replaced or substituted by equivalents, an appropriate result can be achieved.

Therefore, other implementations, other embodiments, and equivalents to the claims are within the scope of the following claims.

Claims (21)

In the system for open live platform,
A first server pool managing at least one first server for receiving streaming data transmitted from a sender's terminal;
A second server pool managing at least one second server for receiving and encoding the streaming data from a first server allocated in the first server pool; And
Coordinator for managing the state of the servers of the first server pool and the second server pool, and dynamically allocates a second server in the second server pool in response to a request of the first server assigned in the first server pool
Including,
In response to the streaming data being sent from the sender's terminal, dynamically assigning any first server in the first server pool, receiving the transmitted streaming data through the assigned first server,
The coordinator,
When a broadcaster of the sender is registered, a channel for the broadcast is generated, and a broadcast address for transmitting streaming data of the broadcast is generated using a unique key assigned to the channel,
And dynamically assigning any first server in the first server pool in response to the streaming data of the broadcast being transmitted through the transmission address. delete The method of claim 1,
The coordinator,
Register a broadcast transmission job in the assigned job list of the second server,
Each of the at least one second server managed in the second server pool monitors its own work list,
A second server having a broadcast transmission task registered in its task list identifies the first server allocated in the first server pool by using the information registered in the task list, and transmits the streaming data from the allocated first server. System for open live platform, characterized in that receiving and encoding. The method of claim 1,
In response to the streaming data being sent from the sender's terminal, a switch for dynamically selecting an arbitrary first server from the first server pool to deliver streaming data transmitted from the sender's terminal to the selected first server.
System for an open live platform, characterized in that it further comprises. The method of claim 1,
Each of at least one first server managed in the first server pool and at least one second server managed in the second server pool is registered with the coordinator when executed,
The coordinator,
And monitoring the status of the registered at least one first server and at least one second server. The method of claim 1,
The coordinator,
When detecting a failure of the allocated first server, the streaming end event is transmitted to the allocated second server to switch the allocated second server to a standby state, and the streaming data is resent from the sender's terminal. And allocating a new second server from the second server pool according to a request of another first server allocated from the first server pool. The method of claim 1,
The coordinator,
When detecting a failure of the allocated second server, allocate a new second server in the second server pool,
And said new assigned second server connects to said assigned first server to receive said streaming data from said assigned first server and encodes said received streaming data. The method of claim 1,
A distributed file system for storing streaming data encoded by the dynamically allocated second server; And
A third server pool for receiving a broadcast viewing request of a viewer terminal and managing at least one third server for providing the encoded streaming data to the viewer terminal in the distributed file system;
More,
Dynamically assigns an arbitrary third server in the third server pool in response to a broadcast viewing request of the viewer terminal, and transmits the encoded streaming data from the distributed file system to the viewer terminal through the assigned third server System for open live platform, characterized in that for providing. The method of claim 8,
A switch that dynamically selects an arbitrary third server from the third server pool in response to the broadcast viewing request of the viewer terminal to transmit the broadcast viewing request to the selected third server
System for an open live platform, characterized in that it further comprises. The method of claim 8,
A system for an open live platform, characterized in that the first server pool, the second server pool, the distributed file system, and the third server pool are configured for each region divided according to at least one of a country and a region. . The method of claim 10,
Dedicated lines are connected between third servers in different zones,
Streaming data transmitted from the first zone is transmitted to the first server of the first zone, encoded by the second server of the first zone, and provided to the viewer terminal located in the first zone via the third server of the first zone. And a viewer terminal located in the second zone through the third server of the second zone connected to the third server of the first zone by a dedicated line. The method of claim 1,
A fourth server pool managing at least one fourth server for generating a VOD file using streaming data encoded through the allocated second server according to a VOD file generation request from the coordinator; And
A fifth server pool that receives a request for the VOD file from a requester terminal and manages at least one fifth server for providing the generated VOD file to the requester terminal according to the request
System for an open live platform, characterized in that it further comprises. In the operation method of the open live platform,
Receiving an allocation request of a second server for encoding the streaming data from any first server allocated in the first server pool in response to the streaming data being sent from the sender terminal; And
Dynamically allocating any second server in a second server pool according to the received allocation request
Including,
The dynamically allocated second server receives and encodes the streaming data from the first server,
When the broadcaster of the sender is registered, generating a channel for the broadcast, and generating a broadcast address for transmitting the streaming data of the broadcast by using a unique key assigned to the channel;
More,
And an arbitrary first server is dynamically allocated from the first server pool in response to the streaming data of the broadcast being transmitted through the transmission address. The method of claim 13,
Dynamically allocating the second random server includes:
Register a broadcast transmission job for the streaming data in the job list of the second random server;
Each of the at least one second server managed in the second server pool monitors its own work list,
The second server whose broadcast transmission task is registered in its task list identifies the first server allocated in the first server pool by using the information registered in its task list and the streaming data from the allocated first server. And receiving and encoding. The method of claim 13,
The encoded streaming data is stored in a distributed file system,
And encoding the encoded streaming data stored in the distributed file system from the third server dynamically allocated in the third server pool according to the request of the viewer terminal and providing the encoded streaming data to the viewer terminal. delete The method of claim 13,
Monitoring status of at least one first server managed in the first server pool and at least one second server managed in the second server pool
Operation method characterized in that it further comprises. The method of claim 17,
In case of detecting a failure of the allocated first server, transferring the streaming end event to the allocated second server to switch the allocated second server to a standby state; And
Allocating a new second server in the second server pool according to a request of another first server allocated in the first server pool as the streaming data is resent from the sender's terminal;
Operation method characterized in that it further comprises. The method of claim 17,
If a failure of the allocated second server is detected, allocating a new second server from the second server pool
More,
And the allocated new second server connects to the assigned first server to receive the streaming data from the assigned first server and to encode the received streaming data. The method of claim 13,
Allocating any fourth server for generation of a VOD file in a fourth server pool
More,
And generating a VOD file using the encoded streaming data in the allocated fourth server. A computer-readable recording medium having recorded thereon a program for causing a computer to execute the method of any one of claims 13 to 15 or 17 to 20.

Images