KR101667998B1 - System and method for job scheduling for multiple media encoders - Google Patents

Abstract

A task scheduling system and method for a plurality of media encoders is provided. A job scheduling system for a plurality of media encoders includes a job ranking determining unit for determining a job order of content data, a monitoring unit for monitoring the number of encoders whose number of content data in operation or waiting jobs is equal to or less than a preset threshold value, A waiting time predicting unit for predicting the waiting time of the available encoders based on the number of the monitored encoders and the information related to the received encoding task, And a task distributing unit for distributing the content data according to the determined task order based on the estimated waiting time of the encoders.

Encoder, task, priority, wait time, weight, monitoring, prediction

Description

Technical Field [0001] The present invention relates to a scheduling system and method for a plurality of media encoders,

The present invention relates to a task scheduling system and method for a plurality of media encoders, and more particularly, to a task scheduling system and method for a plurality of media encoders that can effectively determine the task sequence, efficiently monitor the task status of the encoder, To a task scheduling system and method.

With the spread of the Internet and the development of digital broadcasting technology, the use of the two-way broadcasting service provided using the high-speed Internet network is increasing.

The bidirectional broadcasting service is a service for providing information services, video contents, broadcasting, and the like to a bidirectional broadcasting receiver (for example, IPTV, DMB receiver, set-top box, etc.) Various contents and additional services provided by the Internet such as home banking, online games, and MP3 can be provided.

In particular, to provide a media on-demand service that provides VOD, AOD, etc. over a network, a digital master or a digital capture must be encoded or transcoded according to the packaging format, codec, resolution and bit rate set in the service . In addition, these tasks are performed in general-purpose or dedicated encoders, and are generally provided in a first-come-first-service (FCFS) mode or a multi-stage queue type job scheduler.

However, when scheduling such as first-come-first-service (FCFS) or multi-stage queue type is used for job scheduling of an encoder / transcoder, the encoder usage rate can not be maximized while considering priority for urgent tasks. Particularly, there has been a problem that low-priority content data may not be continuously processed. In particular, when the encoder servers can not access the shared storage in real time, they can not efficiently handle the work, and there is a problem that the actual work is not performed in a predetermined work order.

On the other hand, as a prior art, Korean Patent Laid-Open No. 10-2006-00034863 discloses an invention entitled " Packet transmission method using weight, " and in providing broadcast and multicast services to a plurality of user terminals, And a technique for packet scheduling using a weight considering a type and a type of a service.

However, since the above-described prior art only determines the weight according to the type or type of the service and determines the packet transmission order using the determined weight, if the packet with the higher transmission order continues to increase, There is a problem in that it can not be efficiently transmitted.

Some embodiments of the present invention provide a system and method for scheduling jobs for a plurality of media encoders that can effectively determine the encoding work order of content data to be distributed.

In addition, one embodiment of the present invention provides a system and method for scheduling jobs for a plurality of media encoders that can efficiently monitor the workload of the encoder and accurately distribute the encoding data of the content data to a plurality of encoders in a predetermined order .

As a technical means for achieving the above technical object, a first aspect of the present invention is a data processing apparatus including a work order determining unit for determining a work order of content data, a number of encoder A job status information receiving unit for receiving information related to the encoding job from the available encoders, a job waiting information receiving unit for collecting the job queues of the available encoders based on the number of the monitored encoders and the information related to the received encoding job, A job scheduling system for a plurality of media encoders including a waiting time predicting unit for predicting time and a job distributing unit for distributing content data according to the determined job order based on the waiting wait time of the predicted encoders have.

The second aspect of the present invention also includes a method comprising the steps of: determining a working order of content data; monitoring whether the number of content data in operation or waiting for work for each available encoder is below a predetermined threshold; Receiving the job status information related to the encoding job, calculating a job wait time of the available encoders based on the number of the content data under the job or waiting job, And distributing the content data to the encoder in accordance with the determined work order based on the predicted job wait time. The present invention also provides a method of scheduling jobs for a plurality of media encoders.

According to the present invention, the encoding task sequence is efficiently determined by flexibly reflecting the weight of the wait time of the task allocation of the waiting content data for the encoding task.

In addition, according to the above-mentioned problem solving means of the present invention, the workload of the encoder is efficiently monitored and the encoding operation is accurately distributed in a predetermined order in a small amount of calculation.

Hereinafter, embodiments of the present invention will be described in detail with reference to the accompanying drawings, which will be readily apparent to those skilled in the art. The present invention may, however, be embodied in many different forms and should not be construed as limited to the embodiments set forth herein. In order to clearly explain the present invention, parts not related to the description are omitted, and like parts are denoted by like reference numerals throughout the specification.

Throughout the specification, when a part is referred to as being "connected" to another part, it includes not only "directly connected" but also "electrically connected" with another part in between . Also, when an element is referred to as "comprising ", it means that it can include other elements as well, without departing from the other elements unless specifically stated otherwise.

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS Hereinafter, the present invention will be described in detail with reference to the accompanying drawings.

1 is a block diagram of a task scheduling system for a plurality of media encoders according to an embodiment of the present invention.

1, a task scheduling system for a plurality of media encoders according to an embodiment of the present invention includes a task scheduler server 1000, an encoder server 2000, a shared storage 3000, and a network 4000 . The task scheduler server 1000 includes a task ranking unit 1100, a task distribution unit 1200 and a transmission / reception unit 1300. The encoder server 2000 includes an encoding unit 2110, an encoder task queue 2120 And a control unit 2130. The encoder 2100 includes a plurality of encoders 2100,

The task scheduler server 1000 determines the task ranking of the content data to be encoded and distributes the encoding task for each encoder of the encoder server 2000.

The task ranking determining unit 1100 determines a task ranking of content data that needs to be encoded. The task ranking determining unit 1100 can efficiently determine the task ranking based on the priority and the task allocation waiting time for each content.

The task allocation waiting time is a time for waiting for distribution from the task scheduler server 1000 to the encoder server 2000 for the encoding of the content data, and the task allocation wait time can be measured for each content.

The task distributor 1200 distributes the task to each encoder of the encoder server 2000 so that the task can proceed according to the determined task ranking. The task distributor 1200 monitors the workload of the encoder server according to a predetermined criterion and predicts the work waiting time for all the encoders in the encoder server when there is an encoder satisfying a preset criterion.

The job wait time refers to a time for which the distributed job is to be encoded in the encoder 2100 when the content encoding job is distributed to the encoder 2100. [

The transmission / reception unit 1300 transmits / receives various information necessary for job ranking and work distribution to / from the encoder server 2000 and the shared storage 3000.

The plurality of encoders 2100 of the encoder server 2000 each receive the encoding work instruction of the content data from the job scheduler server 1000 and encode the content data.

The encoding unit 2110 encodes the content data according to the operation sequence stored in the encoder work queue 2120 to be described later.

The encoder task queue 2120 receives the task sequence of the content data to be encoded by the encoding unit 2110 from the task scheduler server 1000 and stores the same.

In the embodiment of the present invention, the encoder server 2000 includes a plurality of encoders 2100. However, the present invention is not limited thereto and may further include an apparatus such as a transcoder.

The control unit 2130 provides the job scheduler server 1000 with information necessary for distributing the encoding task of the content data and causes the encoding unit 2110 to perform the encoding operation according to the job order stored in the encoder job queue 2120. [

The shared storage 3000 stores content data requiring an encoding operation, and provides the content data stored in each encoder 2100.

The network 4000 includes a dedicated line, a LAN, a QAN, a VAN, an intranet, a private telephone network, a public telephone network, a PSTN network, and a mutual combination thereof. , And includes a wired Internet, a wireless Internet, and a mobile wireless communication network.

Hereinafter, the detailed configuration of the task ranking determiner 1100 according to an embodiment of the present invention will be described with reference to FIG.

FIG. 2 is a detailed configuration diagram of a task ranking determining unit 1100 according to an embodiment of the present invention.

2, the task ranking determining unit 1100 according to an embodiment of the present invention includes a priority setting unit 1110, a waiting time measuring unit 1120, a weight setting unit 1130, (S) 1140 and a mast work queue (s) 1150.

The priority setting unit 1110 sets the priority level of the content data to be processed based on the type and characteristics of the content data. For example, the priority level setting unit 1110 can set the priority level to 1 to 5, and sets the priority level of each content based on the genre of content data, whether it is a paid service, Can be set.

The waiting time measuring unit 1120 measures the waiting time of the task allocation of the content data waiting to be distributed to the encoder server 2000 for the encoding task and provides the measured task allocation waiting time to the ranking calculating unit 1140.

The weight setting unit 1130 sets a waiting time weight corresponding to the priority weight and the task allocation waiting time corresponding to the priority. The weight setting unit 1130 may set the weights to correspond to the entire priority level and the entire waiting time.

In addition, the weight setting unit 1130 can adjust the priority weighting and the waiting time weighting, and can flexibly adjust the operation order with respect to various requests for the encoding operation.

The ranking calculation unit 1140 calculates the ranking of the content data based on the priority level, the priority weight, the job waiting time, and the waiting time weight.

The ranking calculation unit 1140 may calculate a value (priority level x priority weight) + (job waiting time x wait time weight), for example, and determine a work order based on the calculated value.

The mast work queue 1150 sequentially stores the work list of the content data according to the calculated work order.

Hereinafter, the detailed configuration of the work distribution unit 1200 according to an embodiment of the present invention will be described with reference to FIG.

3 is a detailed configuration diagram of the job distribution unit 1200 according to an embodiment of the present invention.

3, the task distributor 1200 according to an exemplary embodiment of the present invention includes a monitoring unit 1210, a task status information receiving unit 1220, a waiting time predicting unit 1230, and a task transmitting unit 1240 ).

The monitoring unit 1210 monitors the encoder 2100 whose number of pieces of content data in operation or waiting for work is less than or equal to a preset threshold value.

The monitoring unit 1210 can receive the number of pieces of content data in operation or waiting for a job from the encoder 2100. For example, it is possible to determine how many encoders having the number of content data of 1 or less exist. In addition, the monitoring unit 1210 provides the monitored information to the waiting time predicting unit 1230. [

The work situation information receiving unit 1220 receives information related to the encoding operation from the encoders 2100. [ The work situation information receiving unit 1220 can receive information on the total time of the unprocessed work for each encoder 2100, the input resolution, the output resolution, the codec, the encoder performance, and the like.

The waiting time predicting unit 1230 predicts the waiting time of the jobs of the encoders 2100 usable in the encoder server 2000 connected to the task scheduler server 1000. [

The waiting time predicting unit 1230 predicts the waiting time of the encoded data received from the working status information receiving unit 1220 when the number of pieces of content data in operation or waiting jobs is equal to or less than a predetermined threshold value Based on the information associated with the task, the waiting time for work can be predicted for each encoder 2100. [

In addition, the waiting time predicting unit 1230 can estimate the waiting time for the work for all the encoders 2100 that can perform the encoding work in the encoder server 2000.

The waiting time predicting unit 1230 estimates the remaining time of the content data (total time when the operation is not performed) x input resolution weight x output resolution weight x codec The complexity weight x encoder performance weight 'value, and summing the calculated values to predict the waiting time value for each encoder 2100. Also, in this case, the resolution weight can take the horizontal resolution x the vertical resolution x the frame rate x the video weight, in the case of video data, the bit resolution x the sample rate x the number of channels x audio weight have. Furthermore, each of the weights can be pre-determined by a ratio of the encoding time using the sample set of content data.

The job transferring unit 1240 distributes the encoding job of the content data to the respective encoders 2100 based on the estimated waiting time. The task transferring unit 1240 may distribute the tasks having the higher priority to the encoder 2100 having a shorter waiting time and store the tasks provided in the task transferring unit 1240 in order to each of the encoder work queues 2120 .

Accordingly, the job scheduler server 1000 can estimate the waiting time only when the number of jobs is equal to or less than the predetermined number, and can reduce the amount of calculation required to estimate the waiting time of the encoder 2100, It is possible to accurately determine an encoder capable of quickly processing data irrespective of a waiting job by predicting the waiting time for all usable encoders 2100. [

In addition, the encoder server 2000 can accurately perform encoding in accordance with the order of operations determined by the job scheduler server 1000, efficiently distributes contents to be processed next before each of the encoders 2100 finishes the job, It is possible to solve the work efficiency problem that occurs when the encoder server 2000 accesses data from the external shared storage 3000. [

Hereinafter, a method of determining a work order of a job scheduling system for a plurality of media encoders according to an embodiment of the present invention will be described with reference to FIG.

4 is a detailed flowchart of a method of determining a work order of a job scheduling system for a plurality of media encoders according to an exemplary embodiment of the present invention.

Step S400 is a step of setting a priority level for each content data. In step S400, the level of priority of the content data to be operated is set based on the type and characteristics of the content data. For example, the priority level setting unit 1110 can set the priority level to 1 to 5, and sets the priority level of each content based on the genre of content data, whether it is a paid service, It can be set to a value from 1 to 5.

Step S402 is a step of measuring the work allocation waiting time for each content data. In step S402, the task allocation waiting time of the content data waiting to be distributed to the encoder server 2000 for the encoding operation can be measured.

Step S404 is a step of calculating a job ranking. In step S404, the job ranking of the content data can be calculated based on the priority level, the priority weight, the job wait time, and the wait time weight. For example, (priority level x priority weight) + Time x wait time weight value), and determine the work order based on the calculated value.

Step S406 is a step of determining whether or not the weight corresponding to the priority and the task allocation waiting time is changed.

If it is determined that the weight is changed, the job ranking is calculated based on the changed weight (step S404). If the weight is not changed, step S408 is performed.

Step S408 is a step of storing the determined task priority. In step S408, the task list of the content data according to the calculated task sequence may be stored in order.

According to an embodiment of the present invention, when a job wait time is measured for each content data and a different weight is assigned to the measured wait time, when a job ranking is determined based on a priority level for each content data, It is possible to prevent the problem that the work of the low content data is pushed backward indefinitely, and furthermore, the waiting time weight can be flexibly adjusted to effectively process the content with a large waiting time.

Hereinafter, with reference to FIG. 5, a method of allocating work in a task scheduling system for a plurality of media encoders according to an embodiment of the present invention will be described.

5 is a detailed flowchart of a task allocation method of a task scheduling system for a plurality of media encoders according to an embodiment of the present invention.

Step S500 is a step of monitoring the operation status of the encoder 2100. In step S500, it is possible to monitor the number of the encoders 2100 whose number of pieces of content data during operation or waiting for work is equal to or less than a preset threshold value. In step S500, it is possible to receive from the encoder 2100 the number of pieces of content data that is being processed or waiting for a job. For example, it is possible to determine how many encoders having the number of content data of 1 or less exist.

Step S502 is a step of determining whether or not there is an encoder whose work amount is equal to or less than a predetermined number. In step 502, it may be determined whether or not a predetermined number or less of the encoder 2100 exists.

If it is determined in step S502 that there is no encoder having a predetermined number of workings or less, step S500 is performed. If there is an encoder having a predetermined number of workings or less, step S504 is performed.

Step S504 is a step of receiving the job status information from the encoder 2100 in the encoder server 2000. [ In step S504, it is possible to receive from the encoders 2100 information information related to the encoding operation, such as the total time of the unprocessed operation, the input resolution, the output resolution, the codec, and the encoder performance.

Step S506 is a step of estimating the waiting time of the available encoders 2100 in the encoder server 2000. [ In step S506, the job waiting time of the encoders 2100 usable in the encoder server 2000 connected to the job scheduler server 1000 can be predicted based on the information received in step S504. In step S506, the encoder server 2000 can predict the waiting time for all the encoders 2100 that can perform the encoding operation.

In step S506, the remaining time of content data (total time if the operation is not performed) x input resolution weight x output resolution weight x codec complexity weight x encoder performance Quot; weight " value, and calculate the waiting time value for each encoder 2100 by summing the calculated values. Also, in this case, the resolution weight can take the horizontal resolution x the vertical resolution x the frame rate x the video weight, in the case of video data, the bit resolution x the sample rate x the number of channels x audio weight have.

Step S508 is a step of distributing the task based on the estimated waiting time. In step S508, based on the estimated waiting time, tasks with higher priority can be allocated to the shortened waiting time encoder 2100, and tasks provided in the task transmitting unit 1240 can be distributed to the respective encoder task queues 2120, As shown in FIG.

Therefore, according to the job scheduling method according to the embodiment of the present invention, the waiting time is predicted only when the encoder 2100 whose number of jobs is equal to or less than the predetermined number is present, thereby reducing the amount of calculation required to estimate the waiting time of the encoder 2100 In predicting the waiting time, it is possible to accurately determine the encoder that can actually process the data quickly regardless of the waiting job by predicting the waiting time for all the available encoders 2100.

In addition, the encoder server 2000 can accurately perform encoding in accordance with the order of operations determined by the job scheduler server 1000, efficiently distributes contents to be processed next before each of the encoders 2100 finishes the job, It is possible to solve the work efficiency problem that occurs when the encoder server 2000 accesses data from the external shared storage 3000. [

One embodiment of the present invention may also be embodied in the form of a recording medium including instructions executable by a computer, such as program modules, being executed by a computer. Computer readable media can be any available media that can be accessed by a computer and includes both volatile and nonvolatile media, removable and non-removable media. In addition, the computer-readable medium can include both computer storage media and communication media. Computer storage media includes both volatile and nonvolatile, removable and non-removable media implemented in any method or technology for storage of information such as computer readable instructions, data structures, program modules or other data. Communication media typically includes any information delivery media, including computer readable instructions, data structures, program modules, or other data in a modulated data signal such as a carrier wave, or other transport mechanism.

It will be understood by those skilled in the art that the foregoing description of the present invention is for illustrative purposes only and that those of ordinary skill in the art can readily understand that various changes and modifications may be made without departing from the spirit or essential characteristics of the present invention. will be. It is therefore to be understood that the above-described embodiments are illustrative in all aspects and not restrictive. For example, each component described as a single entity may be distributed and implemented, and components described as being distributed may also be implemented in a combined form.

The scope of the present invention is defined by the appended claims rather than the detailed description and all changes or modifications derived from the meaning and scope of the claims and their equivalents are to be construed as being included within the scope of the present invention do.

1 is a configuration diagram of a task scheduling system for a plurality of media encoders according to an embodiment of the present invention;

2 is a detailed configuration diagram of a task ranking determining unit 1100 according to an embodiment of the present invention.

FIG. 3 is a detailed configuration diagram of a task distribution unit 1200 according to an embodiment of the present invention.

4 is a detailed flowchart of a method of determining a work order of a job scheduling system for a plurality of media encoders according to an embodiment of the present invention.

5 is a detailed flowchart of a task allocation method of a task scheduling system for a plurality of media encoders according to an embodiment of the present invention.

Claims (11)

1. A job scheduling system for a plurality of media encoders, A work order determination unit for determining a work order of a plurality of pieces of content data, A monitoring unit for monitoring the number of encoders whose number of content data in operation or waiting for work is equal to or less than a predetermined threshold, A task status information receiving unit for receiving information related to an encoding task from available encoders, A waiting time predicting unit for predicting the waiting time of the available encoders based on the number of the monitored encoders and the information related to the received encoding task, And a work distribution unit for distributing the plurality of contents data to the encoder in accordance with the determined work order, based on the estimated waiting time of the encoders, , ≪ / RTI & Wherein the job distributing unit distributes the content data having the high operation order to the encoder having the shortened waiting time for the job. The method according to claim 1, The task ranking determination unit A priority setting unit for setting a priority level of the plurality of contents data to be operated, A waiting time measurement unit for measuring a distribution waiting time to the encoder for each of the plurality of contents data, Based on the set priority level and the measured distribution waiting time, a job ranking calculation unit Wherein the job scheduling system comprises: 3. The method of claim 2, Wherein the task priority determination unit further includes a weight setting unit that sets a weight corresponding to the priority level and the distribution waiting time and adjusts the set weight, Wherein the job ranking calculation unit calculates a job ranking of the plurality of pieces of content data based on the priority level, the distribution waiting time, and the weight, Wherein the weights correspond to the entire priority level and the entire wait time, respectively. The method according to claim 1, Wherein the monitoring unit monitors whether or not the number of the encoders whose number of pieces of content data during the operation or waiting for the job is equal to or less than a predetermined threshold is equal to or greater than a predetermined number. 5. The method of claim 4, Wherein the wait time predicting unit predicts a work wait time for all the encoders when the number of the content data in the job or waiting job is equal to or greater than a predetermined number. The method according to claim 1, The work situation information receiving unit From the available encoders, information about the total time of the unprocessed operation, the input resolution, the output resolution, the codec, or the encoder performance. A method for scheduling jobs for a plurality of media encoders, Determining a work order of a plurality of pieces of content data, Monitoring whether the number of content data in operation or waiting for work is below a predetermined threshold value for each available encoder; Receiving operational status information relating to an encoding operation from the available encoders, Predicting the waiting time of the available encoders based on the number of the contents data under the operation or the number of waiting jobs and the number of the encoders below the predetermined threshold value and the received working condition information, Distributing the plurality of pieces of content data to the encoder based on the predicted job waiting time in the determined work order , ≪ / RTI & Wherein the step of distributing to the encoder distributes the content data having a high work order to the encoder having the shortened waiting time for the job. 8. The method of claim 7, Wherein the step of determining the work order comprises: Setting a priority level of the plurality of contents data to be operated, Measuring a distribution waiting time for each of the plurality of contents data to the encoder; and Calculating a job ranking of the plurality of pieces of content data based on the set priority level and the measured distribution waiting time The method further comprising: 9. The method of claim 8, Wherein the step of determining the work order comprises: Further comprising setting a weight corresponding to the priority level and the distribution waiting time and adjusting the set weight, Wherein the calculating the job ranking calculates a job ranking of the plurality of pieces of content data based on the priority level, the distribution waiting time, and the weight, Wherein the weights correspond to the entire priority level and the entire wait time, respectively. 8. The method of claim 7, Wherein the step of predicting the waiting time includes predicting a waiting time for a job for all of the encoders when the number of the content data in the job or waiting jobs is equal to or greater than a predetermined threshold, . 8. The method of claim 7, The step of receiving the job status information And receiving at least one of information about the total time of the unprocessed job, the input resolution, the output resolution, the codec, or the encoder performance from the available encoders.

Images