CN116932230A - Video rendering method based on dynamic task scheduling - Google Patents
Video rendering method based on dynamic task scheduling Download PDFInfo
- Publication number
- CN116932230A CN116932230A CN202311191641.5A CN202311191641A CN116932230A CN 116932230 A CN116932230 A CN 116932230A CN 202311191641 A CN202311191641 A CN 202311191641A CN 116932230 A CN116932230 A CN 116932230A
- Authority
- CN
- China
- Prior art keywords
- rendering
- task
- resource
- adjustment
- monitoring
- 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.)
- Granted
Links
- 238000009877 rendering Methods 0.000 title claims abstract description 254
- 238000000034 method Methods 0.000 title claims abstract description 124
- 230000008569 process Effects 0.000 claims abstract description 101
- 238000012544 monitoring process Methods 0.000 claims abstract description 71
- 238000012797 qualification Methods 0.000 claims abstract description 45
- 230000010354 integration Effects 0.000 claims abstract description 7
- 238000005457 optimization Methods 0.000 claims description 15
- 230000002159 abnormal effect Effects 0.000 claims description 10
- 230000004048 modification Effects 0.000 description 4
- 238000012986 modification Methods 0.000 description 4
- 238000010586 diagram Methods 0.000 description 2
- 238000004519 manufacturing process Methods 0.000 description 2
- 238000013468 resource allocation Methods 0.000 description 2
- 238000006467 substitution reaction Methods 0.000 description 2
- 230000000903 blocking effect Effects 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 230000009467 reduction Effects 0.000 description 1
- 230000000007 visual effect Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F9/00—Arrangements for program control, e.g. control units
- G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
- G06F9/46—Multiprogramming arrangements
- G06F9/50—Allocation of resources, e.g. of the central processing unit [CPU]
- G06F9/5005—Allocation of resources, e.g. of the central processing unit [CPU] to service a request
- G06F9/5027—Allocation of resources, e.g. of the central processing unit [CPU] to service a request the resource being a machine, e.g. CPUs, Servers, Terminals
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F9/00—Arrangements for program control, e.g. control units
- G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
- G06F9/46—Multiprogramming arrangements
- G06F9/48—Program initiating; Program switching, e.g. by interrupt
- G06F9/4806—Task transfer initiation or dispatching
- G06F9/4843—Task transfer initiation or dispatching by program, e.g. task dispatcher, supervisor, operating system
- G06F9/4881—Scheduling strategies for dispatcher, e.g. round robin, multi-level priority queues
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F9/00—Arrangements for program control, e.g. control units
- G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
- G06F9/46—Multiprogramming arrangements
- G06F9/50—Allocation of resources, e.g. of the central processing unit [CPU]
- G06F9/5005—Allocation of resources, e.g. of the central processing unit [CPU] to service a request
- G06F9/5027—Allocation of resources, e.g. of the central processing unit [CPU] to service a request the resource being a machine, e.g. CPUs, Servers, Terminals
- G06F9/505—Allocation of resources, e.g. of the central processing unit [CPU] to service a request the resource being a machine, e.g. CPUs, Servers, Terminals considering the load
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02D—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
- Y02D10/00—Energy efficient computing, e.g. low power processors, power management or thermal management
Landscapes
- Engineering & Computer Science (AREA)
- Software Systems (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Processing Or Creating Images (AREA)
Abstract
The invention relates to the technical field of video animation rendering, in particular to a video rendering method based on dynamic task scheduling, which comprises the steps of obtaining system resource quantity, determining a plurality of rendering tasks corresponding to rendering requirements, and determining system resource occupancy rate required by the rendering tasks; determining a monitoring mode and a task scheduling mode of a rendering process according to the system resource occupancy rates required by a plurality of rendering tasks determined by the resource integration module; executing rendering tasks according to the monitoring mode and the task scheduling mode determined by the resource scheduling module; counting the qualification rate of task completion in a plurality of rendering tasks to determine whether the rendering tasks in a corresponding task scheduling mode are qualified or not; adjusting the rendering process of the rendering task under the condition that the rendering task is unqualified; optimizing the rendering process according to the model complexity; the invention solves the problem of insufficient control precision of the rendering process based on dynamic task scheduling.
Description
Technical Field
The invention relates to the technical field of video animation rendering, in particular to a video rendering method based on dynamic task scheduling.
Background
The video animation production process needs a great number of links, each link is a key node for successful video animation production, and model rendering is applied to the video animation field, so that the model rendering gives people a true and visual feeling, is a serious problem in the video animation field, and most of the existing model rendering realizes qualified utilization of system resources based on a task mobilization mode.
Chinese patent application publication No.: CN114201282a discloses a dynamic task scheduling method for real-time rendering computation, which analyzes whether available resources in current residual resources of a server can meet CPU demand, memory demand and GPU demand of the resource scheduling demand after receiving the resource scheduling demand instruction sent by a user application, marks the server meeting the conditions as an available server, and invokes the available server to execute the resource scheduling demand instruction; therefore, the dynamic matching of the resource requirement of the application end and the available resource of the service end is realized, the resource utilization rate of the service end can be improved, and the problems of image blocking and frame rendering rate reduction of the application end caused by insufficient resources can be effectively avoided; therefore, the problem of low rendering efficiency caused by insufficient control precision of the rendering process based on dynamic task scheduling exists in the prior art.
Disclosure of Invention
Therefore, the invention provides a video rendering method based on dynamic task scheduling, which is used for solving the problem of low rendering efficiency caused by insufficient control precision of a rendering process based on dynamic task scheduling in the prior art.
In order to achieve the above object, the present invention provides a video rendering method based on dynamic task scheduling, including:
step S1, a resource integration module acquires system resource amount, determines a plurality of rendering tasks corresponding to rendering requirements, and determines system resource occupancy rate required by the rendering tasks;
s2, a resource scheduling module determines a monitoring mode and a task scheduling mode of a rendering process according to the system resource occupancy rate required by a plurality of rendering tasks determined by the resource integration module;
step S3, an execution determining module executes rendering tasks according to the monitoring mode and the task scheduling mode determined by the resource scheduling module;
step S4, a task error statistics module counts the qualification rate of task completion in a plurality of rendering tasks to determine whether the rendering tasks in a corresponding task scheduling mode are qualified or not;
step S5, a task adjusting module adjusts the rendering process of the rendering task under the condition that the rendering task is unqualified;
s6, optimizing the rendering process by a task optimization module according to the complexity of the model;
when the task adjusting module adjusts the rendering process of the rendering task, an adjusting mode of the rendering process is determined according to the qualification rate of the task set of the rendering process;
and when the task optimization module determines that the adjusted rendering process is unqualified, determining to optimize the rendering process according to the model complexity in the rendering task.
Further, when the resource scheduling module determines the monitoring mode and the task scheduling mode, the resource scheduling module determines a plurality of monitoring modes for the rendering process according to the comparison result of the system resource occupancy rate required by the rendering tasks and the preset resource occupancy rate, wherein the monitoring modes comprise a first monitoring mode under the condition that the system resource occupancy rate is less than or equal to the preset resource occupancy rate and a second monitoring mode under the condition that the system resource occupancy rate is greater than the preset resource occupancy rate.
Further, when the resource scheduling module determines that the rendering process is monitored in a first monitoring mode, the monitoring unit of the resource scheduling module monitors abnormal points of the rendering unit distributed with the rendering tasks, and when the abnormal points exist in the rendering process, the resource scheduling amount is determined according to the comparison result of the execution time length of the rendering tasks in the rendering unit and the preset time length.
Further, when the resource scheduling module determines to monitor the rendering process in a second monitoring mode, the monitoring unit monitors each rendering node of the rendering task, compares the rendering time length of each rendering node with the standard rendering time length, determines that resources need to be scheduled when the rendering time length is longer than the standard rendering time length, and determines the resource scheduling amount.
Further, when the resource scheduling module determines that resources need to be scheduled, the resource scheduling module calculates a time length difference value between the rendering time length and the standard rendering time length, and determines a resource scheduling amount according to a comparison result of the time length difference value and a preset time length difference value.
Further, when the execution determining module executes the rendering task in the determined monitoring mode and the task scheduling mode, the task error counting module counts the number of error frames of a task set to determine the qualification rate of the task set when the execution of the task set is completed, and determines to adjust the rendering process of the rendering task when the qualification rate is less than or equal to a preset qualification rate.
Further, when the task error statistics module determines to adjust the rendering process of the rendering task, a qualification rate difference value between the qualification rate and a preset qualification rate is calculated, and the task adjustment module determines a plurality of adjustment modes of the rendering process according to a comparison result of the qualification rate difference value and the preset qualification rate difference value, wherein the plurality of adjustment modes comprise a first adjustment mode for increasing the resource adjustment amount and a second adjustment mode for increasing the monitoring force.
Further, when the task adjustment module determines to adjust the rendering process in a first adjustment manner, a ratio of the qualification rate difference to a preset qualification rate difference is calculated to determine an increased resource adjustment scale according to a comparison result of the ratio and the preset ratio.
Further, when the task adjustment module determines to adjust the rendering process in the second adjustment mode, determining the resource occupation amount of each model in the rendering task, and determining a mode of increasing the monitoring strength according to a comparison result of the resource occupation amount and the preset resource occupation amount, wherein the mode of increasing the monitoring strength comprises adding monitoring nodes in the corresponding monitoring mode at a first time interval or adding the monitoring nodes in the corresponding monitoring mode at a second time interval.
Further, when the task adjustment module completes adjustment of the rendering process, the task optimization module determines whether the adjusted rendering process is qualified, if not, the task optimization module determines to optimize the rendering process according to a comparison result of the model complexity and the preset complexity, and the optimization includes adjusting a proportion or a time interval of a resource adjustment amount in a corresponding adjustment mode by a first adjustment coefficient or adjusting a proportion or a time interval of the resource adjustment amount in the corresponding adjustment mode by a second adjustment coefficient.
Compared with the prior art, the method has the advantages that the monitoring mode and the task scheduling mode of the rendering process are determined according to the system resource occupancy rate required by rendering, so that the monitoring and the real-time dynamic task scheduling of the rendering process are realized, the usability of the system in the rendering process is improved by carrying out load balancing control on a server executing the rendering, and the resource scheduling amount under the corresponding monitoring mode and task scheduling mode is set, so that the accurate control of the rendering process is realized, and the rendering efficiency is further improved.
Furthermore, the invention controls the dynamic task scheduling of the rendering process flexibly by setting a plurality of monitoring modes and task scheduling modes, monitors the abnormal points in the rendering process in the first monitoring mode, determines the resource scheduling amount according to the execution time length of the abnormal points, improves the control precision of the rendering process and further improves the rendering efficiency.
Further, in the invention, the rendering nodes are monitored in the second monitoring mode, so that whether to schedule resources is determined according to the rendering time length of each monitored rendering node, and when the scheduling is determined to be required, the resource scheduling amount is determined according to the time length difference, thereby further improving the control precision of the rendering process and further improving the rendering efficiency.
Further, the invention determines the qualification rate of the task set when one task set is determined to be completed in the rendering process, and determines whether to adjust the rendering process according to the qualification rate, thereby further improving the control precision of the rendering process and the rendering efficiency.
Further, when the rendering process is determined to be finished, a plurality of adjustment modes for the rendering process are determined according to the qualification rate difference, and different adjustment modes are set, so that the flexibility of task scheduling in the rendering process is improved, the rendering process is adjusted in the adjustment modes for increasing the resource scheduling amount and the monitoring strength, and the control precision of the rendering process is further improved.
Drawings
FIG. 1 is a flow chart of a video rendering method based on dynamic task scheduling according to an embodiment of the present invention;
fig. 2 is a schematic structural diagram of an execution system of a video rendering method based on dynamic task scheduling according to an embodiment of the present invention.
Detailed Description
In order that the objects and advantages of the invention will become more apparent, the invention will be further described with reference to the following examples; it should be understood that the specific embodiments described herein are for purposes of illustration only and are not intended to limit the scope of the invention.
Preferred embodiments of the present invention are described below with reference to the accompanying drawings. It should be understood by those skilled in the art that these embodiments are merely for explaining the technical principles of the present invention, and are not intended to limit the scope of the present invention.
It should be noted that, in the description of the present invention, terms such as "upper," "lower," "left," "right," "inner," "outer," and the like indicate directions or positional relationships based on the directions or positional relationships shown in the drawings, which are merely for convenience of description, and do not indicate or imply that the apparatus or elements must have a specific orientation, be constructed and operated in a specific orientation, and thus should not be construed as limiting the present invention.
Furthermore, it should be noted that, in the description of the present invention, unless explicitly specified and limited otherwise, the terms "mounted," "connected," and "connected" are to be construed broadly, and may be either fixedly connected, detachably connected, or integrally connected, for example; can be mechanically or electrically connected; can be directly connected or indirectly connected through an intermediate medium, and can be communication between two elements. The specific meaning of the above terms in the present invention can be understood by those skilled in the art according to the specific circumstances.
Referring to fig. 1 and 2, fig. 1 is a flowchart of a video rendering method based on dynamic task scheduling according to an embodiment of the present invention; fig. 2 is a schematic structural diagram of an execution system of a video rendering method based on dynamic task scheduling according to an embodiment of the present invention.
The video rendering method based on dynamic task scheduling in the embodiment of the invention comprises the following steps:
step S1, a resource integration module acquires system resource amount, determines a plurality of rendering tasks corresponding to rendering requirements, and determines system resource occupancy rate required by the rendering tasks;
s2, a resource scheduling module determines a monitoring mode and a task scheduling mode of a rendering process according to the system resource occupancy rate required by a plurality of rendering tasks determined by the resource integration module;
step S3, an execution determining module executes rendering tasks according to the monitoring mode and the task scheduling mode determined by the resource scheduling module;
step S4, a task error statistics module counts the qualification rate of task completion in a plurality of rendering tasks to determine whether the rendering tasks in a corresponding task scheduling mode are qualified or not;
step S5, a task adjusting module adjusts the rendering process of the rendering task under the condition that the rendering task is unqualified;
and S6, optimizing the rendering process by a task optimization module according to the model complexity.
Specifically, the invention realizes the monitoring and real-time dynamic task scheduling of the rendering process by determining the system resource occupancy rate required by the rendering to determine the monitoring mode and the task scheduling mode of the rendering process according to the system resource occupancy rate, thereby realizing the load balancing control of the server for executing the rendering to improve the availability of the system in the rendering process, and setting the resource scheduling amount under the corresponding monitoring mode and task scheduling mode to realize the accurate control of the rendering process so as to improve the rendering efficiency.
Specifically, when the resource scheduling module determines the monitoring mode and the task scheduling mode, the resource scheduling module determines the monitoring mode and the task scheduling mode of the rendering process according to the comparison results of the system resource occupancy rate P and the preset resource occupancy rate P0 required by a plurality of rendering tasks;
when P is less than or equal to P0, the resource scheduling module determines to monitor the rendering task in a first monitoring mode;
when P > P0, the resource scheduling module determines to monitor the rendering task in a second monitoring manner.
In the embodiment of the invention, the preset resource occupancy rate is 50%.
Specifically, when the resource scheduling module determines that the rendering process is monitored in a first monitoring mode, the monitoring unit of the resource scheduling module monitors abnormal points of the rendering unit distributed with the rendering tasks, and when the abnormal points exist in the rendering process, the resource scheduling amount is determined according to the comparison result of the execution time t of the rendering tasks in the rendering unit and the preset time t 0;
when t is less than or equal to t0, the resource scheduling module determines the resource scheduling amount as a first resource scheduling amount;
when t > t0, the resource scheduling module determines the resource scheduling amount as a second resource scheduling amount.
In the embodiment of the present invention, the preset time period t0 is an average time period required for executing a plurality of identical rendering tasks in history, the value of the first resource allocation amount is 2% of the system resource amount, and the value of the second resource allocation amount is 5% of the system resource amount.
Specifically, the invention flexibly controls the dynamic task scheduling of the rendering process by setting a plurality of monitoring modes and task scheduling modes, and monitors the abnormal points in the rendering process in the first monitoring mode to determine the resource scheduling amount according to the execution time length of the abnormal points, thereby improving the control precision of the rendering process and further improving the rendering efficiency.
Specifically, when the resource scheduling module determines to monitor the rendering process in a second monitoring mode, the monitoring unit monitors each rendering node of the rendering task, and compares the rendering time length T of each rendering node with the standard rendering time length T0 according to the rendering time length T of each rendering node so as to determine whether to perform resource scheduling according to the comparison result;
when T is less than or equal to T0, the resource scheduling module determines that scheduling of resources is not needed;
when T > T0, the resource scheduling module determines that resources need to be scheduled, and determines the resource scheduling amount.
In the embodiment of the invention, the value of the standard rendering time length of the single rendering node is 1 second.
Specifically, when the resource scheduling module determines that resources need to be scheduled, the resource scheduling module calculates a time length difference delta T between the rendering time length T and a standard rendering time length T0, and determines a resource scheduling amount according to a comparison result of the time length difference delta T and a preset time length difference delta T0;
when delta T is less than or equal to delta T0, the resource scheduling module determines the resource scheduling amount as a first resource scheduling amount;
when ΔT > ΔT0, the resource scheduling module determines the resource scheduling amount as a second resource scheduling amount.
Specifically, the invention monitors the rendering nodes in the second monitoring mode, so as to determine whether to schedule resources according to the monitored rendering time length of each rendering node, and determine the resource scheduling amount according to the time length difference when determining that scheduling is required, thereby further improving the control precision of the rendering process and further improving the rendering efficiency.
Specifically, when the execution determining module executes a rendering task in a determined monitoring mode and a task scheduling mode, the task error counting module counts the number of error frames of a task set to determine the qualification rate of the task set when the execution of the task set is completed, and determines whether to adjust the rendering process of the rendering task according to the comparison result of the qualification rate Y and a preset qualification rate Y0;
when Y is less than or equal to Y0, the task error statistics module determines to adjust the rendering process of the rendering task;
when Y > Y0, the task error statistics module determines that no adjustments are made to the rendering process of the rendering task.
In the embodiment of the invention, the preset qualification rate Y0 has a value of 97%, and one task set is a plurality of video frames in one continuous video in a rendering task.
Specifically, the invention determines the qualification rate of the task set when one task set is determined to be completed in the rendering process, determines whether to adjust the rendering process according to the qualification rate, further improves the control precision of the rendering process and improves the rendering efficiency.
Specifically, when the task error statistics module determines to adjust the rendering process of the rendering task, a qualification rate difference delta Y between the qualification rate Y and a preset qualification rate Y0 is calculated, and the task adjustment module determines an adjustment mode of the rendering process according to a comparison result of the qualification rate difference delta Y and the preset qualification rate difference delta Y0;
when delta Y is less than or equal to delta Y0, the task adjustment module determines to adjust the rendering process in a first adjustment mode;
when ΔY > ΔY0, the task adjustment module determines to adjust the rendering process in a second adjustment manner;
the first adjustment mode is to increase the resource adjustment amount, and the second adjustment mode is to increase the monitoring strength.
In the embodiment of the invention, the preset qualification rate difference delta Y0 is 2 percent.
Specifically, when the rendering process is determined to be finished, a plurality of adjustment modes for the rendering process are determined according to the qualification rate difference value, and the flexibility of task scheduling in the rendering process is improved by setting different adjustment modes, and the rendering process is adjusted in the adjustment modes for increasing the resource scheduling amount and the monitoring strength, so that the control precision of the rendering process is further improved.
Specifically, when the task adjustment module determines to adjust the rendering process in a first adjustment manner, calculating a ratio B of the qualification rate difference Δy to a preset qualification rate difference Δy0, so as to determine an increased resource adjustment ratio according to a comparison result of the ratio B and the preset ratio B0;
when B is less than or equal to B1, the task adjustment module determines to increase the resource adjustment amount of the first proportion;
when B > B1, the task adjustment module determines to increase the amount of resource adjustment by a second proportion.
In the embodiment of the invention, the value of the preset ratio is 0.5, the first ratio is 2%, and the second ratio is 3%.
Specifically, when the task adjustment module determines to adjust the rendering process in a second adjustment manner, determining a resource occupation amount U of each model in the rendering task, and determining a manner of increasing monitoring strength according to a comparison result of the resource occupation amount U and a preset resource occupation amount U0;
when U is less than or equal to U0, the task adjusting module determines that the rendering unit increases monitoring nodes in a corresponding monitoring mode at a first time interval;
when U is larger than U0, the task adjusting module determines that the rendering unit increases monitoring nodes in a corresponding monitoring mode at a second time interval.
In the embodiment of the invention, the system resource amount with the value of 10% of the resource occupation amount is preset, the first time interval is 50ms, and the second time interval is 20ms.
Specifically, when the task adjustment module completes adjustment of the rendering process, the task optimization module determines whether the adjusted rendering process is qualified, and if not, the task optimization module determines to optimize the rendering process according to a comparison result of the model complexity F and the preset complexity F0;
when F is less than or equal to F0, the task optimization module determines to adjust the proportion or the time interval of the resource adjustment amount in the corresponding adjustment mode by a first adjustment coefficient;
when F > F0, the task optimization module determines to adjust the proportion or the time interval of the resource adjustment amount in the corresponding adjustment mode by the second adjustment coefficient.
In the embodiment of the invention, the value of the preset complexity F is 0.70, the value of the first adjusting coefficient is 1.2, and the value of the second adjusting coefficient is 1.4.
In an embodiment of the present invention, the model complexity is determined from the sum of the ratio of the surface shading area of the model to the total surface area of the model and the ratio of the mixed color area of the model surface to the total surface area of the model.
Specifically, the qualification of the adjusted rendering process is judged, so that the adjusted parameters of the corresponding adjusting mode are determined to be adjusted according to the model complexity under the condition that the adjusted rendering process is unqualified, the rendering process is further optimized, the control precision of the rendering process is further improved, and the rendering efficiency is improved.
Thus far, the technical solution of the present invention has been described in connection with the preferred embodiments shown in the drawings, but it is easily understood by those skilled in the art that the scope of protection of the present invention is not limited to these specific embodiments. Equivalent modifications and substitutions for related technical features may be made by those skilled in the art without departing from the principles of the present invention, and such modifications and substitutions will be within the scope of the present invention.
The foregoing description is only of the preferred embodiments of the invention and is not intended to limit the invention; various modifications and variations of the present invention will be apparent to those skilled in the art. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention should be included in the protection scope of the present invention.
Claims (10)
1. A video rendering method based on dynamic task scheduling, comprising:
step S1, a resource integration module acquires system resource amount, determines a plurality of rendering tasks corresponding to rendering requirements, and determines system resource occupancy rate required by the rendering tasks;
s2, a resource scheduling module determines a monitoring mode and a task scheduling mode of a rendering process according to the system resource occupancy rate required by a plurality of rendering tasks determined by the resource integration module;
step S3, an execution determining module executes rendering tasks according to the monitoring mode and the task scheduling mode determined by the resource scheduling module;
step S4, a task error statistics module counts the qualification rate of task completion in a plurality of rendering tasks to determine whether the rendering tasks in a corresponding task scheduling mode are qualified or not;
step S5, a task adjusting module adjusts the rendering process of the rendering task under the condition that the rendering task is unqualified;
s6, optimizing the rendering process by a task optimization module according to the complexity of the model;
when the task adjusting module adjusts the rendering process of the rendering task, an adjusting mode of the rendering process is determined according to the qualification rate of the task set of the rendering process;
and when the task optimization module determines that the adjusted rendering process is unqualified, determining to optimize the rendering process according to the model complexity in the rendering task.
2. The video rendering method according to claim 1, wherein when the resource scheduling module determines the monitoring mode and the task scheduling mode, the resource scheduling module determines a plurality of monitoring modes for the rendering process according to a comparison result of system resource occupancy rates and preset resource occupancy rates required by a plurality of rendering tasks, and the plurality of monitoring modes include a first monitoring mode under a condition that the system resource occupancy rate is less than or equal to the preset resource occupancy rate and a second monitoring mode under a condition that the system resource occupancy rate is greater than the preset resource occupancy rate.
3. The video rendering method based on dynamic task scheduling according to claim 2, wherein when the resource scheduling module determines to monitor the rendering process in a first monitoring manner, the monitoring unit of the resource scheduling module monitors abnormal points of the rendering units to which the rendering tasks are distributed, and determines the resource scheduling amount according to a comparison result of the execution time length of the rendering tasks in the rendering units and a preset time length when the abnormal points exist in the rendering process.
4. The video rendering method based on dynamic task scheduling according to claim 3, wherein when the resource scheduling module determines to monitor the rendering process in a second monitoring manner, the monitoring unit monitors each rendering node of the rendering task, compares the rendering time length of a single rendering node with a standard rendering time length, and determines that resources need to be scheduled when the rendering time length is longer than the standard rendering time length, and determines the resource scheduling amount.
5. The video rendering method according to claim 4, wherein when the resource scheduling module determines that a resource needs to be scheduled, the resource scheduling module calculates a time length difference between the rendering time length and a standard rendering time length, and determines a resource scheduling amount according to a comparison result of the time length difference and a preset time length difference.
6. The video rendering method according to claim 5, wherein when the execution determining module performs the rendering task in the determined monitoring manner and the task scheduling manner, the task error counting module counts the number of error frames of a task set to determine a qualification rate of the task set when the execution of the task set is completed, and determines to adjust the rendering process of the rendering task when the qualification rate is less than or equal to a preset qualification rate.
7. The video rendering method according to claim 6, wherein when the task error statistics module determines to adjust the rendering process of the rendering task, a qualification rate difference between the qualification rate and a preset qualification rate is calculated, and the task adjustment module determines a plurality of adjustment modes for the rendering process according to a comparison result of the qualification rate difference and the preset qualification rate difference, wherein the plurality of adjustment modes includes a first adjustment mode for increasing the resource adjustment amount and a second adjustment mode for increasing the monitoring force.
8. The video rendering method of claim 7, wherein when the task adjustment module determines to adjust the rendering process in a first adjustment manner, a ratio of the qualification rate difference to a preset qualification rate difference is calculated to determine an increased resource adjustment scale according to a comparison of the ratio and the preset ratio.
9. The video rendering method according to claim 8, wherein when the task adjustment module determines to adjust the rendering process in the second adjustment manner, determining a resource occupation amount of each model in the rendering task, and determining a manner of increasing the monitoring effort according to a comparison result of the resource occupation amount and a preset resource occupation amount, wherein the manner of increasing the monitoring effort includes increasing the monitoring node in the corresponding monitoring manner at a first time interval or increasing the monitoring node in the corresponding monitoring manner at a second time interval.
10. The video rendering method according to claim 9, wherein when the task adjustment module completes the adjustment of the rendering process, the task optimization module determines whether the adjusted rendering process is qualified, and if not, the task optimization module determines to optimize the rendering process according to a comparison result of a model complexity and a preset complexity, and the optimization includes adjusting a proportion or a time interval of a resource adjustment amount in a corresponding adjustment manner by a first adjustment coefficient or adjusting a proportion or a time interval of a resource adjustment amount in a corresponding adjustment manner by a second adjustment coefficient.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311191641.5A CN116932230B (en) | 2023-09-15 | 2023-09-15 | Video rendering method based on dynamic task scheduling |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311191641.5A CN116932230B (en) | 2023-09-15 | 2023-09-15 | Video rendering method based on dynamic task scheduling |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN116932230A true CN116932230A (en) | 2023-10-24 |
| CN116932230B CN116932230B (en) | 2023-12-08 |
Family
ID=88375719
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202311191641.5A Active CN116932230B (en) | 2023-09-15 | 2023-09-15 | Video rendering method based on dynamic task scheduling |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116932230B (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117560459A (en) * | 2024-01-11 | 2024-02-13 | 深圳市志泽科技有限公司 | Audio/video conversion method based on conversion wire |
Citations (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20120001925A1 (en) * | 2010-06-30 | 2012-01-05 | Ati Technologies, Ulc | Dynamic Feedback Load Balancing |
| CN108415766A (en) * | 2018-03-06 | 2018-08-17 | 洛阳理工学院 | A kind of rendering task dynamic dispatching method |
| CN110532100A (en) * | 2019-09-02 | 2019-12-03 | Oppo广东移动通信有限公司 | Dispatch method, apparatus, terminal and the storage medium of resource |
| CN111930479A (en) * | 2020-06-22 | 2020-11-13 | 北京视博云信息技术有限公司 | Resource scheduling method and device, readable storage medium and equipment |
| US20210200656A1 (en) * | 2019-12-27 | 2021-07-01 | Intel Corporation | Apparatus and method for adaptively scheduling work on heterogeneous processing resources |
| CN113064695A (en) * | 2021-03-25 | 2021-07-02 | 西安万像电子科技有限公司 | Processing method and device of video memory resources |
| CN113961331A (en) * | 2021-11-05 | 2022-01-21 | 深圳市思迪信息技术股份有限公司 | Video rendering task scheduling method and device |
| CN114968521A (en) * | 2022-05-20 | 2022-08-30 | 每平每屋(上海)科技有限公司 | Distributed rendering method and device |
| KR20220161070A (en) * | 2021-05-28 | 2022-12-06 | 주식회사 엘지유플러스 | Container-based gpu resource sharing mtehod and apparatus in cloud environment |
| CN115934365A (en) * | 2023-03-10 | 2023-04-07 | 深圳华龙讯达信息技术股份有限公司 | Task scheduling coupling system applied to CPU |
| CN116263701A (en) * | 2022-11-29 | 2023-06-16 | 中移(苏州)软件技术有限公司 | Computing power network task scheduling method and device, computer equipment and storage medium |
| WO2023122954A1 (en) * | 2021-12-28 | 2023-07-06 | 深圳大学 | Multi-rendering task scheduling method for quality of service perception of interactive application |
| CN116521504A (en) * | 2023-05-10 | 2023-08-01 | 武汉烽火技术服务有限公司 | Method and device for managing and measuring calculation force |
-
2023
- 2023-09-15 CN CN202311191641.5A patent/CN116932230B/en active Active
Patent Citations (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20120001925A1 (en) * | 2010-06-30 | 2012-01-05 | Ati Technologies, Ulc | Dynamic Feedback Load Balancing |
| CN108415766A (en) * | 2018-03-06 | 2018-08-17 | 洛阳理工学院 | A kind of rendering task dynamic dispatching method |
| CN110532100A (en) * | 2019-09-02 | 2019-12-03 | Oppo广东移动通信有限公司 | Dispatch method, apparatus, terminal and the storage medium of resource |
| US20210200656A1 (en) * | 2019-12-27 | 2021-07-01 | Intel Corporation | Apparatus and method for adaptively scheduling work on heterogeneous processing resources |
| CN111930479A (en) * | 2020-06-22 | 2020-11-13 | 北京视博云信息技术有限公司 | Resource scheduling method and device, readable storage medium and equipment |
| CN113064695A (en) * | 2021-03-25 | 2021-07-02 | 西安万像电子科技有限公司 | Processing method and device of video memory resources |
| KR20220161070A (en) * | 2021-05-28 | 2022-12-06 | 주식회사 엘지유플러스 | Container-based gpu resource sharing mtehod and apparatus in cloud environment |
| CN113961331A (en) * | 2021-11-05 | 2022-01-21 | 深圳市思迪信息技术股份有限公司 | Video rendering task scheduling method and device |
| WO2023122954A1 (en) * | 2021-12-28 | 2023-07-06 | 深圳大学 | Multi-rendering task scheduling method for quality of service perception of interactive application |
| CN114968521A (en) * | 2022-05-20 | 2022-08-30 | 每平每屋(上海)科技有限公司 | Distributed rendering method and device |
| CN116263701A (en) * | 2022-11-29 | 2023-06-16 | 中移(苏州)软件技术有限公司 | Computing power network task scheduling method and device, computer equipment and storage medium |
| CN115934365A (en) * | 2023-03-10 | 2023-04-07 | 深圳华龙讯达信息技术股份有限公司 | Task scheduling coupling system applied to CPU |
| CN116521504A (en) * | 2023-05-10 | 2023-08-01 | 武汉烽火技术服务有限公司 | Method and device for managing and measuring calculation force |
Non-Patent Citations (1)
| Title |
|---|
| 侯守明;张玉珍;: "基于时间负载均衡蚁群算法的云任务调度优化", 测控技术, no. 07, pages 31 - 35 * |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117560459A (en) * | 2024-01-11 | 2024-02-13 | 深圳市志泽科技有限公司 | Audio/video conversion method based on conversion wire |
| CN117560459B (en) * | 2024-01-11 | 2024-04-16 | 深圳市志泽科技有限公司 | Audio/video conversion method based on conversion wire |
Also Published As
| Publication number | Publication date |
|---|---|
| CN116932230B (en) | 2023-12-08 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN116932230B (en) | Video rendering method based on dynamic task scheduling | |
| AU777032B2 (en) | System and method for persistence-vector-based rate assignment | |
| US7193992B2 (en) | Method of radio resource management for integrated voice and data CDMA networks | |
| EP1645076A1 (en) | Method and apparatus for providing a delay guarantee for a wireless network | |
| CN104427625A (en) | Network resource scheduling method and system based on user experience | |
| CN110198274A (en) | A kind of control method and its system of group flow | |
| CN114444739A (en) | Digital smart power grid region management system and method | |
| US7298719B2 (en) | Method for scheduling wireless downlink transmissions subject to rate constraints | |
| CN115334084A (en) | Cloud platform based on cloud computing and internet | |
| CN103081537B (en) | The attainable product of communication equipment, method and computer for admissions control | |
| CN104770036A (en) | System and methods to achieve optimum efficiency-Jain fairness in wireless systems | |
| CN112261120A (en) | Cloud-side cooperative task unloading method and device for power distribution internet of things | |
| US20110222425A1 (en) | Controlling network traffic | |
| CN109167737A (en) | The CAN dispatching method of fused data reduction and dynamic priority queue | |
| CN102404778B (en) | Load estimation method | |
| EP2658082A1 (en) | Electricity management apparatus and electricity management method | |
| CN102932885B (en) | Mobile intelligent terminal third-generation (3G) communication energy consumption and user performance experience balancing and scheduling scheme | |
| CN114185674A (en) | CPU task real-time scheduling method, device, equipment and storage medium | |
| CN111954239B (en) | Cloud computing-based basic communication network optimization system for public security | |
| CN102056095B (en) | Method, system and device for controlling short message transmitting flow rate | |
| CN116302578B (en) | QoS (quality of service) constraint stream application delay ensuring method and system | |
| CN110855570A (en) | SDN (software defined network) packet loss judgment method, device and system and multi-network control system | |
| CN116302582A (en) | Stock exchange platform load balancing control system | |
| US20020124041A1 (en) | System and method for managing real-time processing | |
| CN105682102A (en) | Unlicensed channel competition parameter generation method and device |
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 |