CN109639486B - Live broadcast-based cloud host elastic expansion method - Google Patents
Live broadcast-based cloud host elastic expansion method Download PDFInfo
- Publication number
- CN109639486B CN109639486B CN201811526626.0A CN201811526626A CN109639486B CN 109639486 B CN109639486 B CN 109639486B CN 201811526626 A CN201811526626 A CN 201811526626A CN 109639486 B CN109639486 B CN 109639486B
- Authority
- CN
- China
- Prior art keywords
- cloud host
- live broadcast
- task
- module
- cloud
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Images
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L41/00—Arrangements for maintenance, administration or management of data switching networks, e.g. of packet switching networks
- H04L41/08—Configuration management of networks or network elements
- H04L41/0896—Bandwidth or capacity management, i.e. automatically increasing or decreasing capacities
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N21/00—Selective content distribution, e.g. interactive television or video on demand [VOD]
- H04N21/20—Servers specifically adapted for the distribution of content, e.g. VOD servers; Operations thereof
- H04N21/21—Server components or server architectures
- H04N21/218—Source of audio or video content, e.g. local disk arrays
- H04N21/2187—Live feed
Abstract
The invention discloses a live broadcast-based cloud host elastic expansion method. The system comprises a first calculation module, a first configuration module, a first filtering module, a second calculation module and an integration module, wherein the first calculation module is used for analyzing and calculating the performance range of the live broadcast to obtain a performance value; the method comprises the steps that cloud host configuration information used for live broadcasting is defined through a first configuration module, and the cloud host configuration in the first configuration module is screened through a first filtering module to obtain the most appropriate cloud host required by the live broadcasting; calculating the expansion or reduction quantity of the cloud host through a second calculation module according to the information obtained by the first configuration module and the first filtering module; the integration module integrates the cloud host sdk to expand or reduce the specific types of cloud hosts and send corresponding live broadcast tasks. The invention has the beneficial effects that: the concurrency upper limit of the live broadcast task can be effectively improved, and the resource utilization rate of the cloud host is maximized.
Description
Technical Field
The invention relates to the technical field related to live broadcast processing, in particular to a live broadcast-based cloud host elastic expansion method.
Background
The live broadcast task usually consumes resources, the fixed number of hosts cannot cope with the emergency situation of the live broadcast task, and if the cloud host runs without the task, great resource waste is caused.
The elastic expansion of the existing cloud host is only based on the hardware use condition, and the flexible expansion can not be carried out aiming at the specific live broadcast task.
Disclosure of Invention
The invention provides a live broadcast-based cloud host elastic expansion method capable of effectively improving the resource utilization rate of a cloud host in order to overcome the defects in the prior art.
In order to achieve the purpose, the invention adopts the following technical scheme:
a live broadcast-based elastic expansion method for a cloud host comprises a first computing module, a first configuration module, a first filtering module, a second computing module and an integration module, and the specific process is as follows:
(1) analyzing and calculating the performance range of the live broadcast by a first calculation module to obtain a performance value;
(2) the method comprises the steps that cloud host configuration information used for live broadcasting is defined through a first configuration module, and the cloud host configuration in the first configuration module is screened through a first filtering module to obtain the most appropriate cloud host type required by the live broadcasting;
(3) after the most appropriate cloud host type required by the live broadcast is obtained, detecting whether the current cloud host type has an idle cloud host, if so, issuing a task to the idle cloud host, if not, ending the process, asynchronously expanding the capacity, and entering the step (5);
(4) detecting whether the task is successfully issued, and if the task is successfully issued, ending the process; if the task fails to be issued, adding one to the number of continuous errors of the current cloud host, recording the number of continuous errors of the current cloud host, removing when the number of continuous errors reaches a fixed number, and returning to the step (3); meanwhile, whether capacity expansion is needed or not is obtained through the second calculation module, and if the capacity expansion is not needed, the process is ended; if the capacity expansion is needed, entering the next step;
(5) calling the cloud host sdk for application through the integration module, and detecting whether the application is successful;
(6) if the application is unsuccessful, returning to the step (5); if the application is successful, adding cloud host information to the resource pool, selecting whether the task needs to be automatically issued, and if so, finishing the process after the task is established on the newly established cloud host; if not, the process ends.
According to the live broadcast-based cloud host elastic expansion scheme provided by the invention, the concurrency upper limit of a live broadcast task can be effectively improved, and the resource utilization rate of the cloud host is maximized.
Preferably, in step (1), specifically, the performance index is defined according to parameters affecting live broadcast transcoding, including video height, video frame rate, and a performance value of the live broadcast is generated by a rule, where the performance value calculation formula is: video high video frame rate/(reference high reference wide).
Preferably, in the step (2), the cloud host configuration information includes a minimum performance value borne by the cloud host, a maximum live channel number of the cloud host, a number of free resource pools of the cloud host, and an upper limit of a number of capacity expansion machines of the cloud host; specifically, the cloud hosts are divided into a plurality of stages from low to high, each type of cloud host is tested respectively, and the width, the height and the frame rate are output in an incremental mode until a live broadcast picture is not smooth and the hardware utilization rate reaches a peak value, so that the maximum and minimum performance values of each type of cloud host are obtained; aiming at different services, different idle resource pools are set for each type of cloud host, namely the cloud hosts in the idle resource pools are also in operation when no task exists, so that the problem that the processing speed of live broadcast tasks is reduced due to time consumption caused by capacity expansion machines is solved, and meanwhile, the whole live broadcast volume can be controlled by setting the quantity of the capacity expansion machines of each type of cloud host, so that resource waste caused by malicious live broadcast is prevented; in order to guarantee the accuracy of the live broadcast tasks, the maximum continuous error times of each type of cloud host are increased, and when the number of live broadcast task processing errors of the cloud host reaches the value, cloud host release is carried out.
Preferably, in step (2), specifically, the filtering rule of the first filtering module is: the performance value of the current live broadcast task is between the minimum performance value and the maximum performance value of the defined cloud host; the current running task number of the cloud host does not reach the maximum live broadcast path number of the cloud host in the configuration module; the cloud host does not have the execution error record of the current live broadcast task.
Preferably, in the step (4), the second computing module is configured to compute an expansion or reduction amount of the cloud host according to the information obtained by the first configuration module and the first filtering module; specifically, another appropriate cloud host is obtained from the first configuration module and the first filtering module, and the process is circulated until the success; wherein: and if the task is successful or fails, the resource pool of the cloud hosts is adjusted and calculated to obtain the quantity of the cloud hosts needing capacity expansion or elimination.
Optionally, in the step (5), the integration module is configured to integrate the cloud host sdk to perform capacity expansion or reduce specific types of cloud hosts and issue corresponding live broadcast tasks; specifically, based on the cloud host sdk, the machine is expanded or rejected, whether a task is waiting for expansion needs to be processed, and if the task is completed, the task needs to be processed; secondary screening is needed during elimination, whether a running task exists on the cloud host to be eliminated currently is judged, and if the running task exists, the elimination operation is ignored; abnormal retry is carried out no matter expansion or elimination, and expansion failure caused by insufficient resources of cloud host manufacturers is prevented.
The invention has the beneficial effects that: the concurrency upper limit of the live broadcast task can be effectively improved, and the resource utilization rate of the cloud host is maximized.
Drawings
FIG. 1 is a flow chart of the method of the present invention.
Detailed Description
The invention is further described with reference to the following figures and detailed description.
In the embodiment shown in fig. 1, a live broadcast-based elastic expansion method for a cloud host includes a first computing module, a first configuration module, a first filtering module, a second computing module, and an integration module, and the specific process includes:
(1) analyzing and calculating the performance range of the live broadcast by a first calculation module to obtain a performance value;
specifically, the performance index is defined according to parameters affecting live broadcast transcoding, including video height, video width and video frame rate, and a performance value of the live broadcast is generated through a rule, wherein a performance value calculation formula is as follows: video high video frame rate/(reference high video frame rate); the reference height width is defined as 640 x 360, that is, the performance value of the live broadcast with the video width being 640 x 360 and the frame rate being 1 is 1, then the performance value of the normal high-definition live broadcast (the video width being 1280 x 720 and the frame rate being 25) is 100 according to the formula;
(2) the method comprises the steps that cloud host configuration information used for live broadcasting is defined through a first configuration module, and the cloud host configuration in the first configuration module is screened through a first filtering module to obtain the most appropriate cloud host type required by the live broadcasting;
the cloud host configuration information comprises a performance value borne by the cloud host in the minimum, a performance value borne by the cloud host in the maximum, the maximum live broadcast path number of the cloud host, the number of the idle resource pools of the cloud host and the upper limit of the number of the capacity expansion machines of the cloud host; specifically, the cloud host is divided into a plurality of gears from low to high, for example: the method comprises the following steps that 4 cores 8G, 8 cores 16G, 12 cores 24G, 16 cores 32G, 24 cores 48G, 32 cores 64G and the like are respectively tested, and the width, the height and the frame rate are incrementally output until a live broadcast picture is not smooth and the hardware utilization rate reaches a peak value, so that the maximum and minimum performance value of each type of cloud host is obtained; aiming at different services, different idle resource pools are set for each type of cloud host, namely the cloud hosts in the idle resource pools are also in operation when no task exists, so that the problem that the processing speed of live broadcast tasks is reduced due to time consumption caused by capacity expansion machines is solved, and meanwhile, the whole live broadcast volume can be controlled by setting the quantity of the capacity expansion machines of each type of cloud host, so that resource waste caused by malicious live broadcast is prevented; in order to guarantee the accuracy of the live broadcast tasks, the maximum continuous error times of each type of cloud host are increased, and when the number of live broadcast task processing errors of the cloud host reaches the value, cloud host release is carried out;
specifically, the filtering rule of the first filtering module is as follows: the performance value of the current live broadcast task is between the minimum performance value and the maximum performance value of the defined cloud host; the current running task number of the cloud host does not reach the maximum live broadcast path number of the cloud host in the configuration module; the cloud host does not have the execution error record of the current live broadcast task;
(3) after the most appropriate cloud host type required by the live broadcast is obtained, detecting whether the current cloud host type has an idle cloud host, if so, issuing a task to the idle cloud host, if not, ending the process, asynchronously expanding the capacity, and entering the step (5);
(4) detecting whether the task is successfully issued, and if the task is successfully issued, ending the process; if the task fails to be issued, adding one to the number of continuous errors of the current cloud host, recording the number of continuous errors of the current cloud host, removing when the number of continuous errors reaches a fixed number, and returning to the step (3); meanwhile, whether capacity expansion is needed or not is obtained through the second calculation module, and if the capacity expansion is not needed, the process is ended; if the capacity expansion is needed, entering the next step;
the second computing module is used for computing the expansion or reduction quantity of the cloud host according to the information obtained by the first configuration module and the first filtering module; specifically, another appropriate cloud host is obtained from the first configuration module and the first filtering module, and the process is circulated until the success; wherein: the task success or failure can carry out resource pool adjustment calculation of the cloud hosts to obtain the quantity of the cloud hosts needing capacity expansion or elimination;
(5) calling the cloud host sdk for application through the integration module, and detecting whether the application is successful;
the integration module is used for integrating the cloud host sdk to perform capacity expansion or reduce specific types of cloud hosts and corresponding live broadcast task issuing; specifically, based on the cloud host sdk, the machine is expanded or rejected, whether a task is waiting for expansion needs to be processed, and if the task is completed, the task needs to be processed; secondary screening is needed during elimination, whether a running task exists on the cloud host to be eliminated currently is judged, and if the running task exists, the elimination operation is ignored; abnormal retry is carried out no matter expansion or elimination, so that expansion failure caused by insufficient resources of cloud host manufacturers is prevented;
(6) if the application is unsuccessful, returning to the step (5); if the application is successful, adding cloud host information to the resource pool, selecting whether the task needs to be automatically issued, and if so, finishing the process after the task is established on the newly established cloud host; if not, the process ends.
In conclusion, according to the live broadcast-based cloud host elastic expansion method provided by the invention, the concurrency upper limit of the live broadcast task can be effectively improved, and the resource utilization rate of the cloud host can be maximized. Has been applied to the rainbow cloud live broadcast. At present, various live broadcasts from standard definition to 4K are supported, real-time expansion of a cloud host based on live broadcast tasks is realized, and the concurrence upper limit of the number of live channels in the rainbow cloud is obviously improved.
Claims (4)
1. A live broadcast-based elastic expansion method of a cloud host is characterized by comprising a first computing module, a first configuration module, a first filtering module, a second computing module and an integration module, and the specific flow is as follows:
(1) analyzing and calculating the performance range of the live broadcast by a first calculation module to obtain a performance value;
(2) the method comprises the steps that cloud host configuration information used for live broadcasting is defined through a first configuration module, and the cloud host configuration in the first configuration module is screened through a first filtering module to obtain the most appropriate cloud host type required by the live broadcasting; the cloud host configuration information comprises a performance value borne by the cloud host in the minimum, a performance value borne by the cloud host in the maximum, the maximum live broadcast path number of the cloud host, the number of the idle resource pools of the cloud host and the upper limit of the number of the capacity expansion machines of the cloud host; specifically, the cloud hosts are divided into a plurality of stages from low to high, each type of cloud host is tested respectively, and the width, the height and the frame rate are output in an incremental mode until a live broadcast picture is not smooth and the hardware utilization rate reaches a peak value, so that the maximum and minimum performance values of each type of cloud host are obtained; aiming at different services, different idle resource pools are set for each type of cloud host, namely the cloud hosts in the idle resource pools are also in operation when no task exists, so that the problem that the processing speed of live broadcast tasks is reduced due to time consumption caused by capacity expansion machines is solved, and meanwhile, the whole live broadcast volume can be controlled by setting the quantity of the capacity expansion machines of each type of cloud host, so that resource waste caused by malicious live broadcast is prevented; in order to guarantee the accuracy of the live broadcast tasks, the maximum continuous error times of each type of cloud host are increased, and when the number of live broadcast task processing errors of the cloud host reaches the value, cloud host release is carried out; specifically, the filtering rule of the first filtering module is as follows: the performance value of the current live broadcast task is between the minimum performance value and the maximum performance value of the defined cloud host; the current running task number of the cloud host does not reach the maximum live broadcast path number of the cloud host in the configuration module; the cloud host does not have the execution error record of the current live broadcast task;
(3) after the most appropriate cloud host type required by the live broadcast is obtained, detecting whether the current cloud host type has an idle cloud host, if so, issuing a task to the idle cloud host, if not, ending the process, asynchronously expanding the capacity, and entering the step (5);
(4) detecting whether the task is successfully issued, and if the task is successfully issued, ending the process; if the task fails to be issued, adding one to the number of continuous errors of the current cloud host, recording the number of continuous errors of the current cloud host, removing when the number of continuous errors reaches a fixed number, and returning to the step (3); meanwhile, whether capacity expansion is needed or not is obtained through the second calculation module, and if the capacity expansion is not needed, the process is ended; if the capacity expansion is needed, entering the next step;
(5) calling the cloud host sdk for application through the integration module, and detecting whether the application is successful;
(6) if the application is unsuccessful, returning to the step (5); if the application is successful, adding cloud host information to the resource pool, selecting whether the task needs to be automatically issued, and if so, finishing the process after the task is established on the newly established cloud host; if not, the process ends.
2. The method as claimed in claim 1, wherein in step (1), specifically, the performance index is defined according to parameters affecting live broadcast transcoding, including video height, video frame rate, and the performance value of the live broadcast is generated by a rule, and the performance value calculation formula is as follows: video high video frame rate/(reference high reference wide).
3. The live broadcast-based cloud host elastic expansion and contraction method according to claim 1, wherein in the step (4), the second calculation module is used for calculating the expansion or reduction quantity of the cloud host according to the information obtained by the first configuration module and the first filtering module; specifically, another appropriate cloud host is obtained from the first configuration module and the first filtering module, and the process is circulated until the success; wherein: and if the task is successful or fails, the resource pool of the cloud hosts is adjusted and calculated to obtain the quantity of the cloud hosts needing capacity expansion or elimination.
4. The live broadcast-based cloud host elastic expansion method according to claim 1, wherein in the step (5), the integration module is used for integrating the cloud host sdk to perform capacity expansion or reduce specific types of cloud hosts and corresponding live broadcast task issuing; specifically, based on the cloud host sdk, the machine is expanded or rejected, whether a task is waiting for expansion needs to be processed, and if the task is completed, the task needs to be processed; secondary screening is needed during elimination, whether a running task exists on the cloud host to be eliminated currently is judged, and if the running task exists, the elimination operation is ignored; abnormal retry is carried out no matter expansion or elimination, and expansion failure caused by insufficient resources of cloud host manufacturers is prevented.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811526626.0A CN109639486B (en) | 2018-12-13 | 2018-12-13 | Live broadcast-based cloud host elastic expansion method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811526626.0A CN109639486B (en) | 2018-12-13 | 2018-12-13 | Live broadcast-based cloud host elastic expansion method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109639486A CN109639486A (en) | 2019-04-16 |
| CN109639486B true CN109639486B (en) | 2021-10-15 |
Family
ID=66073727
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811526626.0A Active CN109639486B (en) | 2018-12-13 | 2018-12-13 | Live broadcast-based cloud host elastic expansion method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109639486B (en) |
Families Citing this family (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111479134A (en) * | 2020-04-02 | 2020-07-31 | 亦非云互联网技术（上海）有限公司 | Distributed cloud transcoding method, system and server |
| CN114168325A (en) * | 2021-11-26 | 2022-03-11 | 山东浪潮科学研究院有限公司 | Elastic expansion method and device based on edge environment |
Family Cites Families (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103248659B (en) * | 2012-02-13 | 2016-04-20 | 北京华胜天成科技股份有限公司 | A kind of cloud computing resource scheduling method and system |
| WO2013131189A1 (en) * | 2012-03-08 | 2013-09-12 | Iwatchlife Inc. | Cloud-based video analytics with post-processing at the video source-end |
| CN105407056B (en) * | 2014-09-16 | 2019-04-26 | 中国电信股份有限公司 | Business chain method for building up and system in a kind of software defined network |
| CN104243998B (en) * | 2014-09-29 | 2018-01-09 | 广州华多网络科技有限公司 | A kind of data processing method, device and associated server |
| CN105426228B (en) * | 2015-10-29 | 2018-07-27 | 西安交通大学 | A kind of OpenStack virtual machine placement methods towards live streaming media and video code conversion |
| CN106201661B (en) * | 2016-07-20 | 2018-09-14 | 北京百度网讯科技有限公司 | Method and apparatus for elastic telescopic cluster virtual machine |
-
2018
- 2018-12-13 CN CN201811526626.0A patent/CN109639486B/en active Active
Also Published As
| Publication number | Publication date |
|---|---|
| CN109639486A (en) | 2019-04-16 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109639486B (en) | Live broadcast-based cloud host elastic expansion method | |
| CN103873852A (en) | Multi-mode parallel video quality fault detection method and device | |
| JP3227717U (en) | Collision check data processing device | |
| EP4170514A1 (en) | Data association query method and apparatus, and device and storage medium | |
| CN107402851B (en) | Data recovery control method and device | |
| CN104182348A (en) | Software test method and device | |
| CN109218728B (en) | Scene switching detection method and system | |
| CN112182289A (en) | Data deduplication method and device based on Flink framework | |
| CN106547958A (en) | A kind of graphical analysis method and device of mobile unit data | |
| WO2016206241A1 (en) | Data analysis method and apparatus | |
| CN108090224B (en) | Cascade connection method and device | |
| CN110704223A (en) | Recovery system and method for single-node abnormity of database | |
| CN115525703A (en) | Data comparison method, data synchronization device and medium for heterogeneous database | |
| US20220318205A1 (en) | Machine station file processing methods and machine station file processing systems | |
| CN1684043A (en) | Real time monitoring system and method for computer files | |
| CN106874391B (en) | deadlock processing method and device | |
| CN113438417A (en) | Method, system, medium and device for capturing object to be identified by video | |
| CN105830437A (en) | Method and system for identifying background in monitoring system | |
| CN1758572A (en) | Method for detecting fault of receiving channel | |
| CN111476101A (en) | Video shot switching detection method and device and computer readable storage medium | |
| CN116939679B (en) | Multi-unmanned aerial vehicle distributed cluster construction method under unreliable network | |
| CN115018398B (en) | Animation project postponing prediction method, device and system | |
| WO2014127703A1 (en) | Self-optimizing system and method, and computer storage medium | |
| CN110377474B (en) | Loop test method and system for Mellanox network card | |
| EP4181476A1 (en) | Network data analysis method, network data analysis functional network element and communication system |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |