Method for automatically applying and releasing machine based on cloud rendering platform and sky-wing cloud

Abstract

The invention discloses a method for automatically applying and releasing a machine based on a cloud rendering platform and a sky-wing cloud. The method comprises the following steps: S1, task monitoring; s2, applying and distributing machine resources; s3, executing and monitoring a rendering task; and S4, task completion and resource release. According to the invention, machines can be released when there is no task, and automatic application is realized when there are many tasks and the number of machines is limited. According to the method, the problems of tedious operation, resource waste and long task response time of a current cloud rendering platform are solved, and the resource utilization rate and the task rendering efficiency are improved. Meanwhile, the method has flexibility, intelligence and scalability, and an innovative solution is provided for the field of cloud rendering.

Description

Technical Field

[0001] The present invention relates to the field of cloud rendering technology, and in particular to a method for automatically applying for and releasing machines based on a cloud rendering platform and Tianyi Cloud. Background Art

[0002] Currently, with the booming development and widespread application of cloud computing and cloud rendering technology, more and more companies and individuals are beginning to choose to deploy rendering tasks in the cloud. The cloud rendering platform provides users with powerful rendering capabilities and convenient resource management methods. Users can flexibly apply for and release rendering machines according to their needs, and flexibly set the priority and deadline of rendering tasks according to actual conditions.

[0003] However, there are some problems with the current cloud rendering platform. One is the tediousness and unintelligent nature of manual operations. Users need to manually log in to the platform, select appropriate machine resources, and apply for and release tasks according to certain rules. This manual operation takes a lot of time and effort, and is a tedious process for users.

[0004] The second is the waste of resources and task queuing caused by a fixed number of machines. Current cloud rendering platforms usually allocate a fixed number of machine resources. Regardless of whether there are tasks to be processed, these machine resources will continue to be charged, resulting in resource waste. At the same time, during the peak period of tasks, if the available machine resources are insufficient, tasks will be queued, thereby extending the response time of tasks.

[0005] Therefore, the prior art has defects and needs to be improved. Summary of the invention

[0006] The technical problem to be solved by the present invention is: to provide a method for automatically applying for and releasing machines based on a cloud rendering platform and Tianyi Cloud, utilizing the powerful rendering capabilities provided by the cloud rendering platform and the flexible resource management of Tianyi Cloud to realize intelligent application and release of machine resources.

[0007] The technical solution of the present invention is as follows: a method for automatically applying for and releasing machines based on a cloud rendering platform and Tianyi Cloud is provided, comprising the following steps:

[0008] S1: Task monitoring;

[0009] The cloud rendering platform regularly monitors the task queue to determine whether there are any pending rendering tasks;

[0010] If the task queue is empty, it means that there is no rendering task that needs to be processed temporarily; then determine whether there are machine resources currently in use, and if so, release some or all of the machine resources;

[0011] If the task queue is not empty, go to step S2;

[0012] S2: Machine resource application and allocation;

[0013] Automatically send machine resource application requests to Tianyi Cloud based on the number and priority of pending tasks and the resource utilization rules preset by the cloud rendering platform;

[0014] Tianyi Cloud matches and allocates resources based on the currently available machine resources, and returns the available machine resource information to the cloud rendering platform;

[0015] S3: rendering task execution and monitoring;

[0016] The cloud rendering platform assigns the task to the requested machine resources and starts executing the rendering task;

[0017] The cloud rendering platform monitors the execution status of rendering tasks in real time, including the progress and completion time of the tasks;

[0018] S4: Task completion and resource release;

[0019] When the rendering task is completed or the rendering task deadline is reached, the cloud rendering platform analyzes the rendering machine resources and releases the resources back to Tianyi Cloud for others to use; the released machine resources will be restored to an idle state and will be used for the next rendering task application.

[0020] For peak task processing, the following method is adopted: during the peak task period, the cloud rendering platform determines whether the current machine resources are sufficient based on the real-time load monitoring data; if the available machine resources are insufficient, the cloud rendering automatically sends an application request for additional machine resources to Tianyi Cloud; Tianyi Cloud provides additional machine resources based on the application situation and returns them to the cloud rendering platform to meet the peak task requirements.

[0021] In step S2, the specific steps are:

[0022] S21: Determine whether there is an idle machine.

[0023] If not, go to step S22; if so, determine whether the number of idle machines is greater than the number of waiting tasks. If so, directly assign the tasks to the machines to perform rendering tasks, and go to step S3; if not, assign the corresponding number of rendering tasks to the idle machines, and the remaining tasks enter the waiting state, and go to step S22;

[0024] S22: Determine whether the total number of existing machines has reached the limit of the number of machines. If not, call Tianyi Cloud API to apply for a machine, create a machine, and enter step S23; if yes, the remaining tasks continue to wait for an idle machine to appear, and enter step S21;

[0025] S23: Calculate the difference between the waiting tasks and the machines being created;

[0026] S24: Determine whether the number of machines being created can meet the waiting tasks. If so, assign the waiting tasks to the machines being created and go to step S3; if not, assign the corresponding number of rendering tasks to the machines being created, and the remaining tasks wait for an idle machine to appear, and go to step S21.

[0027] In step S4, the specific steps are:

[0028] S41: monitoring the status of the machine of the current cloud rendering platform through a timer;

[0029] S42: Determine whether there are machines with idle time greater than 1 minute. If not, it means that all machines are executing rendering tasks, and the process ends. If yes, proceed to step S43.

[0030] S43: determining whether there are tasks waiting;

[0031] If yes, determine whether the idle number is greater than the number of waiting tasks, if not, then end, if yes, this goes to step S45; if not, go to step S44;

[0032] S44: Determine whether the number of idle machines is greater than the set minimum number of machines. If not, end the process. If yes, proceed to step S45.

[0033] S45: Calculate the difference between the number of idle machines and the minimum threshold + the number of waiting tasks;

[0034] S55: Call Tianyi Cloud API to release the machines with the difference quantity selected from the idle machines, and end.

[0035] The timer uses the open source distributed task scheduling platform xxl-job to implement visual timing execution, routing management, and logging functions.

[0036] The present invention can release machines when there are no tasks, and automatically apply for them when there are many tasks but the number of machines is limited. This method solves the problems of cumbersome operations, waste of resources and long task response time in the current cloud rendering platform, and improves resource utilization and efficiency of rendering tasks. At the same time, the present invention has flexibility, intelligence and scalability, bringing an innovative solution to the field of cloud rendering.

[0037] By adopting the above scheme, the present invention provides a method for automatically applying for and releasing machines based on a cloud rendering platform and Tianyi Cloud, which has the following technical effects:

[0038] Task monitoring and machine resource management: By regularly monitoring the task queue and machine resource utilization, the machine can be released when there are no tasks, and machine resources can be automatically requested when there are tasks to be processed.

[0039] Machine resource application and allocation: Based on the number of tasks, priority and resource utilization rules, machine resource application requests are sent to Tianyi Cloud, and machine resources are allocated based on Tianyi Cloud's resource allocation.

[0040] Rendering task execution and monitoring: Assign tasks to the requested machine resources to execute rendering tasks, and monitor the progress and completion time of rendering tasks in real time.

[0041] Peak task processing: During peak task periods, the system determines whether machine resources are sufficient based on real-time load monitoring data. If insufficient, it automatically applies for additional machine resources to meet peak task requirements.

[0042] Task completion and resource release: After the task is completed or the deadline is reached, the rendering machine resources are analyzed and released back to Tianyi Cloud. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] Figure 1 The present invention is a flow chart of the method. DETAILED DESCRIPTION

[0044] The present invention is described in detail below in conjunction with the accompanying drawings and specific embodiments.

[0045] See also Figure 1 The present invention provides a method for automatically applying for and releasing machines based on a cloud rendering platform and Tianyi Cloud, comprising the following steps:

[0046] S1: Task monitoring;

[0047] The cloud rendering platform regularly monitors the task queue to determine whether there are any pending rendering tasks;

[0048] If the task queue is empty, it means that there is no rendering task that needs to be processed temporarily; then determine whether there are machine resources currently in use, and if so, release some or all of the machine resources;

[0049] If the task queue is not empty, go to step S2;

[0050] S2: Machine resource application and allocation;

[0051] Automatically send machine resource application requests to Tianyi Cloud based on the number and priority of pending tasks and the resource utilization rules preset by the cloud rendering platform;

[0052] Tianyi Cloud matches and allocates resources based on the currently available machine resources, and returns the available machine resource information to the cloud rendering platform;

[0053] S3: rendering task execution and monitoring;

[0054] The cloud rendering platform assigns the task to the requested machine resources and starts executing the rendering task;

[0055] The cloud rendering platform monitors the execution status of rendering tasks in real time, including the progress and completion time of the tasks;

[0056] S4: Task completion and resource release;

[0057] When the rendering task is completed or the rendering task deadline is reached, the cloud rendering platform analyzes the rendering machine resources and releases the resources back to Tianyi Cloud for others to use; the released machine resources will be restored to an idle state and will be used for the next rendering task application.

[0058] For peak task processing, the following method is adopted: during the peak task period, the cloud rendering platform determines whether the current machine resources are sufficient based on the real-time load monitoring data; if the available machine resources are insufficient, the cloud rendering automatically sends an application request for additional machine resources to Tianyi Cloud; Tianyi Cloud provides additional machine resources based on the application situation and returns them to the cloud rendering platform to meet the peak task requirements.

[0059] By introducing automatic application and release of machines, the current problems of resource waste and long task response time can be solved, which has the following advantages:

[0060] Automated management: Through automated machine resource management, user manual operations are reduced and operational efficiency is improved.

[0061] Intelligent resource utilization: Automatically select, apply for, and release machine resources based on the platform's preset resource utilization rules and task requirements to improve resource utilization.

[0062] Elastic Scaling: Automatically apply for additional machine resources to meet task requirements based on changes in task volume, effectively solving the problem of insufficient resources during peak hours.

[0063] Through the above technical solutions, the automatic application and release machine mode based on the cloud rendering platform and Tianyi Cloud can release machines when there are no tasks, and automatically apply when there are many tasks but the number of machines is limited, thereby improving resource utilization, reducing users' manual operation costs, and responding to task requirements in a timely manner, improving the efficiency and response time of rendering tasks.

[0064] In step S2, the specific steps are:

[0065] S21: Determine whether there is an idle machine.

[0066] If not, go to step S22; if so, determine whether the number of idle machines is greater than the number of waiting tasks. If so, directly assign the tasks to the machines to perform rendering tasks, and go to step S3; if not, assign the corresponding number of rendering tasks to the idle machines, and the remaining tasks enter the waiting state, and go to step S22;

[0067] S22: Determine whether the total number of existing machines has reached the limit of the number of machines. If not, call Tianyi Cloud API to apply for a machine, create a machine, and enter step S23; if yes, the remaining tasks continue to wait for an idle machine to appear, and enter step S21;

[0068] S23: Calculate the difference between the waiting tasks and the machines being created;

[0069] S24: Determine whether the number of machines being created can meet the waiting tasks. If so, assign the waiting tasks to the machines being created and go to step S3; if not, assign the corresponding number of rendering tasks to the machines being created, and the remaining tasks wait for an idle machine to appear, and go to step S21.

[0070] In step S4, the specific steps are:

[0071] S41: monitoring the status of the machine of the current cloud rendering platform through a timer;

[0072] S42: Determine whether there are machines with idle time greater than 1 minute. If not, it means that all machines are executing rendering tasks, and the process ends. If yes, proceed to step S43.

[0073] S43: determining whether there are tasks waiting;

[0074] If yes, determine whether the idle number is greater than the number of waiting tasks, if not, then end, if yes, this goes to step S45; if not, go to step S44;

[0075] S44: Determine whether the number of idle machines is greater than the set minimum number of machines. If not, end the process. If yes, proceed to step S45.

[0076] S45: Calculate the difference between the number of idle machines and the minimum threshold + the number of waiting tasks;

[0077] S55: Call Tianyi Cloud API to release the machines with the difference quantity selected from the idle machines, and end.

[0078] The timer uses the open source distributed task scheduling platform xxl-job to implement visual timing execution, routing management, and logging functions.

[0079] This embodiment has the following advantages:

[0080] 1. Automated operation process: With the technical solution of the present invention, users only need to set task requirements, and the system will automatically handle the application and release process of machine resources without manual operation. This automated operation process greatly simplifies the user's operation process and improves the user's operation convenience and efficiency.

[0081] 2. Improve resource utilization: The technical solution of the present invention can apply for and release machine resources according to the number and priority of tasks through intelligent machine resource management. By releasing idle machine resources and automatically applying for additional machine resources, the system can effectively improve resource utilization and reduce resource waste. This can maximize the use of available machine resources, improve the efficiency of rendering tasks and the performance of the overall system.

[0082] 3. Shorten task response time: With the technical solution of the present invention, the system can respond quickly when tasks are pending, and automatically apply for and allocate machine resources. At the same time, Tianyi Cloud uses cloud disk startup technology to greatly shorten the time for machine initialization and application, shortening machine startup from one or two hours to one or two minutes, and putting it into use in time. From the user's perspective, it can effectively shorten the task response time and improve the processing efficiency of rendering tasks. Users do not need to wait in line for resources, and tasks can be quickly started, saving valuable time.

[0083] 4. Saving resource costs: Since the technical solution of the present invention can release machine resources in time after the task is completed, it can reduce resource occupation when there are no tasks to be processed, thereby avoiding the waste of resources caused by continuous charging of machine resources that are not used by tasks. This can save resource costs and improve the economic benefits of the overall system.

[0084] 5. Elastic scaling for peak period processing: During the peak period of tasks, the technical solution of the present invention can determine whether the machine resources are sufficient based on real-time load monitoring data. When the available machine resources are insufficient, the system can automatically apply for additional machine resources from Tianyi Cloud. This elastic scaling mechanism ensures that the demand can still be met during the peak period of tasks, avoiding task delays and inefficiency caused by insufficient machines.

[0085] The automatic application and release of machines in the present invention can realize automated operation, improve resource utilization, shorten task response time, save resource costs, and handle peak hours with elastic scaling. These effects make the present invention have broad application prospects and market competitive advantages in cloud rendering platforms and Tianyi Cloud.

[0086] The present invention can release machines when there are no tasks, and automatically apply for them when there are many tasks but the number of machines is limited. This method solves the problems of cumbersome operations, waste of resources and long task response time in the current cloud rendering platform, and improves resource utilization and efficiency of rendering tasks. At the same time, the present invention has flexibility, intelligence and scalability, bringing an innovative solution to the field of cloud rendering.

[0087] In summary, the present invention provides a method for automatically applying for and releasing machines based on a cloud rendering platform and Tianyi Cloud, which has the following technical effects:

[0088] Task monitoring and machine resource management: By regularly monitoring the task queue and machine resource utilization, the machine can be released when there are no tasks, and machine resources can be automatically requested when there are tasks to be processed.

[0089] Machine resource application and allocation: Based on the number of tasks, priority and resource utilization rules, machine resource application requests are sent to Tianyi Cloud, and machine resources are allocated based on Tianyi Cloud's resource allocation.

[0090] Rendering task execution and monitoring: Assign tasks to the requested machine resources to execute rendering tasks, and monitor the progress and completion time of rendering tasks in real time.

[0091] Peak task processing: During peak task periods, the system determines whether machine resources are sufficient based on real-time load monitoring data. If insufficient, it automatically applies for additional machine resources to meet peak task requirements.

[0092] Task completion and resource release: After the task is completed or the deadline is reached, the rendering machine resources are analyzed and released back to Tianyi Cloud.

[0093] The above are only preferred embodiments of the present invention and are not intended to limit the present invention. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. A method for automatically applying for and releasing machines based on a cloud rendering platform and Tianyi Cloud, characterized in that: The following steps are involved: S1: Task monitoring; The cloud rendering platform regularly monitors the task queue to determine whether there are any pending rendering tasks; If the task queue is empty, it means there is no rendering task that needs to be processed at the moment; If the task queue is not empty, go to step S2; S2: Machine resource application and allocation; Automatically send machine resource application requests to Tianyi Cloud based on the number and priority of pending tasks and the resource utilization rules preset by the cloud rendering platform; Tianyi Cloud matches and allocates resources based on the currently available machine resources, and returns the available machine resource information to the cloud rendering platform; S3: rendering task execution and monitoring; The cloud rendering platform assigns the task to the requested machine resources and starts executing the rendering task; The cloud rendering platform monitors the execution status of rendering tasks in real time; S4: Task completion and resource release; When the rendering task is completed or the rendering task deadline is reached, the cloud rendering platform analyzes the rendering machine resources and releases the resources back to Tianyi Cloud; the released machine resources will be restored to an idle state and will be used for the next rendering task application.

2. The method for automatically applying for and releasing machines based on the cloud rendering platform and Tianyi Cloud according to claim 1 is characterized in that: In step S2, the specific steps are: S21: Determine whether there is an idle machine. If not, proceed to step S22; If yes, determine whether the number of idle machines is greater than the number of waiting tasks. If yes, directly assign the tasks to the machines to perform rendering tasks, and go to step S3. If no, assign the corresponding number of rendering tasks to the idle machines, and the remaining tasks enter the waiting state, and go to step S22. S22: Determine whether the total number of existing machines has reached the limit of the number of machines. If not, call Tianyi Cloud API to apply for a machine, create a machine, and enter step S23; if yes, the remaining tasks continue to wait for an idle machine to appear, and enter step S21; S23: Calculate the difference between the waiting tasks and the machines being created; S24: Determine whether the number of machines being created can meet the waiting tasks. If so, assign the waiting tasks to the machines being created and go to step S3; if not, assign the corresponding number of rendering tasks to the machines being created, and the remaining tasks wait for an idle machine to appear and go to step S21.

3. The method for automatically applying for and releasing machines based on the cloud rendering platform and Tianyi Cloud according to claim 1 is characterized in that: In step S4, the specific steps are: S41: monitoring the status of the machine of the current cloud rendering platform through a timer; S42: Determine whether there are machines with idle time greater than 1 minute. If not, it means that all machines are executing rendering tasks, and the process ends. If yes, proceed to step S43. S43: judging whether there are tasks waiting; If yes, determine whether the idle number is greater than the number of waiting tasks, if no, then end, if yes, this enters step S45; If not, proceed to step S44; S44: Determine whether the number of idle machines is greater than the set minimum number of machines. If not, end the process. If yes, proceed to step S45. S45: Calculate the difference between the number of idle machines and the minimum threshold + the number of waiting tasks; S55: Call Tianyi Cloud API to release the machines with the difference quantity selected from the idle machines, and end.

4. The method for automatically applying for and releasing machines based on the cloud rendering platform and Tianyi Cloud according to claim 1 is characterized in that: The timer uses the open source distributed task scheduling platform xxl-job to implement visual timing execution, routing management, and logging functions.

5. The method for automatically applying for and releasing machines based on the cloud rendering platform and Tianyi Cloud according to claim 1 is characterized in that: In step S3, the cloud rendering platform monitors the execution status of the rendering task in real time, including the progress and completion time of the task.

6. The method for automatically applying for and releasing machines based on the cloud rendering platform and Tianyi Cloud according to claim 1 is characterized in that: In step S1, if the task queue is empty, it means that there is no rendering task that needs to be processed temporarily; then it is determined whether there are machine resources currently in use, and if so, some or all of the machine resources are released.

7. The method for automatically applying for and releasing machines based on the cloud rendering platform and Tianyi Cloud according to claim 1 is characterized in that: For peak task processing, the following methods are adopted: During the peak period of tasks, the cloud rendering platform determines whether the current machine resources are sufficient based on real-time load monitoring data; If the available machine resources are insufficient, Cloud Rendering automatically sends a request for additional machine resources to Tianyi Cloud; Tianyi Cloud provides additional machine resources based on the application situation and returns them to the cloud rendering platform to meet peak task demands.

Images