Cloud rendering scheduling method, system and storage medium suitable for application update period

Abstract

The present invention discloses a cloud rendering scheduling method, system and storage medium suitable for application update, including: receiving a scheduling request sent by a client, generating a plan according to the client computing resource demand carried in the scheduling request; determining the application operation / operation and maintenance status and initial value of the weight of each server in the server cluster, and updating the data of the server resource index according to the planned operation and maintenance needs; providing application scheduling services according to the new server resource index. The above scheme effectively solves the problem of not relying on a large number of redundant computing servers to implement a scheduling scheme without downtime during the update in a multi-application, hybrid storage scenario, while ensuring startup efficiency, which greatly saves costs to a certain extent.

Description

Technical Field

[0001] The present invention relates to computing resource scheduling technology used in the fields of cloud rendering and cloud computing, and in particular to a cloud rendering scheduling method, system and storage medium suitable for application update. Background Art

[0002] Currently in the field of cloud rendering, in order to solve the problem of service clusters for rendering applications (such as games) remaining available during updates, the following solutions are often used:

[0003] A-By temporarily calling other cluster servers or currently redundant idle servers to replace the machines that need to be updated, the update is completed in batches. The disadvantage of this method is that a certain number of redundant machines must be reserved to ensure that the overall cloud rendering computing power of the cluster is sufficient. If the application update frequency is high and the number of applications is large, more redundant machines are needed to reduce batches and complete the turnover.

[0004] B-In a cluster, each server instance stores multiple backups locally. If one of the backups is being updated, other backups are used to provide services. Although this method does not require additional server instance costs, if the number of applications is large, the storage cost of a single server will increase exponentially, or exceed the disk space limit of the cloud computing server itself, resulting in limited application scenarios of the solution itself.

[0005] C-Use the hybrid solution of local storage + network storage, which is common in cloud computing. Although this solution can solve the space limitation, the application being updated, or the local storage space limitation, can not store the rendering network storage, but the network storage solution (NFS or similar solutions) in some high-throughput scenarios, the cloud rendering application will run slower, or may not run normally. Therefore, it does not meet the requirements in actual use experience. Summary of the invention

[0006] In view of the defects in the prior art, the purpose of the present invention is to provide a cloud rendering scheduling method, system and storage medium suitable for application updates, which calls computing power to solve customers' various needs for graphics computing, AI computing and scientific computing. From the scheduling service level, an application evaluation mechanism is introduced to evaluate the scores of different applications on different servers. Based on the score calculation, the plan is generated based on the integration of past customer computing resource demand records, and the customer computing power demand is allocated to the most suitable instance to avoid insufficient cluster computing resources.

[0007] To achieve the above-mentioned object, in a first aspect, the present invention provides a cloud rendering scheduling method suitable for application update, comprising: receiving a scheduling request sent by a client, and generating a plan according to the client computing resource demand carried in the scheduling request;

[0008] Determine the application operation / maintenance status and initial weight of each server in the server cluster, and update the server resource index data according to the planned maintenance needs;

[0009] Provide application scheduling services based on the new server resource index.

[0010] Preferably, the data of updating the server resource index according to the planned operation and maintenance needs includes:

[0011] Obtain historical data on application usage status based on the client's computing resource requirements, reduce the weight of at least one application in the server based on the application's past usage, current usage, and update requirements, and issue update instructions to the server cluster;

[0012] When any server in the server cluster receives an application update task, the server locally updates the application scheduling weight and reports it to the planning server;

[0013] The planning server obtains the operation / operation and maintenance status and initial value of the weight of each application in the server cluster; records the historical data of the application usage status, calculates the scheduling weight of the application in the server resource index, and obtains the updated server resource index.

[0014] Furthermore, the calculation of the scheduling weight of the server resource index application includes:

[0015] When a server in the server cluster receives any application update task, the server is instantly prohibited from scheduling the scheduled service, and the scheduling is resumed in the next cycle, and the initial weight value is updated;

[0016] In combination with the current operation / maintenance situation of the server application and according to the requirements of the plan, the initial weight value of the application is recalculated to determine the scheduling weight of each of the server applications.

[0017] Furthermore, the data of the updating server resource index also includes: obtaining the current application update volume and calculating the update time; obtaining the current usage of the application and the historical usage over the same period and calculating the number of simultaneous updates;

[0018] Find the server that can be updated according to the server resource index and prepare temporary cache space for the update;

[0019] Calculate the maximum number of concurrently updateable resources and set higher weights for the updated application and server resource indexes.

[0020] Furthermore, providing application scheduling service according to the new server resource index includes: obtaining the server resource index; scheduling according to the application weight value in the index, the higher the weight value of the application in the current server, the higher the scheduling weight.

[0021] Preferably, obtaining the server resource index includes: recording the scheduling weights of different applications, and when receiving a scheduling request from a client, defining the scheduling priority according to the weight value of the application requested for scheduling; when the value is 0, scheduling the application on the server is prohibited.

[0022] In a second aspect, the present invention provides a cloud rendering scheduling system suitable for application update, the system comprising:

[0023] A planning service module is used to receive a scheduling request sent by a client and generate a plan according to the client computing resource demand carried in the scheduling request;

[0024] An update module is used to determine the application operation / operation and maintenance status and initial value of the weight of each server in the server cluster, and update the data of the server resource index according to the planned operation and maintenance needs;

[0025] The scheduling service module is used to provide application scheduling services according to the new server resource index.

[0026] In a third aspect, the present invention further provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the cloud rendering scheduling method applicable to application updates as described in any one of claims 1 to 6.

[0027] The beneficial effects of the present invention are embodied in:

[0028] The present invention provides a cloud rendering scheduling method, system and storage medium suitable for application update period, receiving a scheduling request sent by a client, generating a plan according to the client computing resource demand carried in the scheduling request; determining the application operation / operation and maintenance status and initial value of the weight of each server in the server cluster, updating the data of the server resource index according to the planned operation and maintenance needs; providing application scheduling service according to the new server resource index. From the scheduling service level, by introducing the application operation / operation and maintenance status and weight evaluation mechanism of the server, different applications on different servers are evaluated, and based on the evaluation results and the past customer computing resource demand records, the customer computing demand is allocated to the most suitable instance, thereby avoiding insufficient cluster computing resources.

[0029] The solution proposed in this invention effectively solves the scheduling solution that does not rely on a large number of redundant computing servers to implement the update without downtime in the scenario of multiple applications and hybrid storage, while ensuring the startup efficiency. In the case of using hybrid storage, a small amount of redundant resources, and limited local storage space, the cost of cloud computing resources is effectively controlled through the scheduling weight algorithm. BRIEF DESCRIPTION OF THE DRAWINGS

[0030] In order to more clearly illustrate the specific implementation of the present invention or the technical solution in the prior art, the drawings required for use in the specific implementation or the description of the prior art are briefly introduced below.

[0031] Figure 1 is a flow chart of a cloud rendering scheduling method applicable to an application update period provided by an embodiment of the present invention;

[0032] Figure 2 is a schematic diagram of a cloud rendering scheduling architecture applicable to an application update period provided by an embodiment of the present invention;

[0033] Figure 3 It is a schematic diagram of a method for calculating a server resource index application weight provided by an embodiment of the present invention;

[0034] Figure 4 is a flow chart of the execution of the calculation strategy for the update quantity provided by an embodiment of the present invention;

[0035] Figure 5 It is a structural block diagram of a cloud rendering scheduling system suitable for application update period provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0036] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0037] It should be noted that, unless otherwise specified, the technical terms or scientific terms used in this application should have the common meanings understood by those skilled in the art to which the present invention belongs.

[0038] Embodiment 1: Based on the technical concept, core protection points and technical problems solved by the present invention, this embodiment 1 provides a cloud rendering scheduling method suitable for application update period, such as Figure 1 As shown, the method specifically comprises the following steps:

[0039] S101, receiving a scheduling request sent by a client, and generating a plan according to the client computing resource demand carried in the scheduling request;

[0040] S102, determining the application operation / operation and maintenance status and initial value of the weight of each server in the server cluster, and updating the data of the server resource index according to the planned operation and maintenance needs;

[0041] S103: Provide application scheduling service according to the new server resource index.

[0042] In step S102, the data of the server resource index is updated according to the planned operation and maintenance needs, including:

[0043] Obtain historical data on application usage status based on the client's computing resource requirements, reduce the weight of at least one application in the server based on the application's past usage, current usage, and update requirements, and issue update instructions to the server cluster;

[0044] When any server in the server cluster receives an application update task, the server locally updates the application scheduling weight and reports it to the planning server;

[0045] The planning server obtains the operation / operation and maintenance status and initial value of the weight of each application in the server cluster; records the historical data of the application usage status, calculates the scheduling weight of the application in the server resource index, and obtains the updated server resource index.

[0046] Based on the architecture corresponding to the above steps S101 and S102, Figure 2 As shown. Among them, planning service: obtain the operation and maintenance status and initial value of weight of each software in the server cluster. Record the usage status of some applications in the past. Update the data of server resource index.

[0047] Server resource index: records the scheduling weights of different games / applications. When a game scheduling request is sent from the customer service, the one with a higher weight will be scheduled first. If the value is 0, the game / application will not be scheduled on the server.

[0048] Operation and maintenance / update service: Based on the past usage, current usage, and update requirements of the planning service, the planning server is informed to reduce the weight values ​​of certain applications in certain servers, and update instructions are issued to the server cluster.

[0049] Server cluster: If a single machine in the cluster receives an application update or operation and maintenance task, the server locally updates the application scheduling weight (initial value) and reports it to the planning service.

[0050] like Figure 3 As shown, the scheduling weight of the computing server resource index application includes:

[0051] When a server in the server cluster receives any application update task, the server is instantly prohibited from scheduling the scheduled service, and the scheduling is resumed in the next cycle, and the initial weight value is updated;

[0052] In combination with the current operation / maintenance of the server application and according to the requirements of the plan, the initial weight value of the application is recalculated, ie, x→F(x), to determine the scheduling weight of each of the server applications.

[0053] Furthermore, the updating of the data of the server resource index also includes calculating the update quantity, such as Figure 4 As shown, it specifically includes: obtaining the current application update volume and calculating the update time; obtaining the current usage of the application and the historical usage in the same period, and calculating the number of simultaneous updates;

[0054] Find the server that can be updated according to the server resource index and prepare temporary cache space for the update;

[0055] Calculate the maximum number of concurrently updateable resources and set higher weights for the updated application and server resource indexes.

[0056] While ensuring the update efficiency, calculate the maximum number of updates that can be made at the same time, and set the updated application + server resources to a higher weight value. If there is an update rate requirement, temporarily prepare temporary storage services in the public cloud or private cloud to ensure the number of service paths for specific applications.

[0057] Step S103, providing application scheduling service according to the new server resource index includes: obtaining the server resource index; scheduling according to the application weight value in the index, the higher the weight value of the application in the current server, the higher the scheduling weight.

[0058] Among them, obtaining the server resource index includes: recording the scheduling weights of different applications, and when receiving a scheduling request from the client, defining the scheduling priority according to the weight value of the application requested for scheduling; when the value is 0, scheduling the application on the server is prohibited.

[0059] In addition, based on the same technical concept, the specific embodiment of the present invention also provides a cloud rendering scheduling system suitable for application update period, such as Figure 5 As shown, the system comprises:

[0060] A planning service module is used to receive a scheduling request sent by a client and generate a plan according to the client computing resource demand carried in the scheduling request;

[0061] An update module is used to determine the application operation / operation and maintenance status and initial value of the weight of each server in the server cluster, and update the data of the server resource index according to the planned operation and maintenance needs;

[0062] The scheduling service module is used to provide application scheduling services according to the new server resource index.

[0063] Furthermore, a specific embodiment of the present invention also provides a computer-readable storage medium, which stores a computer program. When the computer program is executed by a processor, the steps of a cloud rendering scheduling method suitable for application updates as described in any one of steps S101-103 are implemented.

[0064] Specifically, the processor may be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor may be a microprocessor or any conventional processor, etc.

[0065] The code of the computer program may be in source code form, object code form, executable file or some intermediate form.

[0066] Computer-readable storage media may include cache, high-speed random access memory (RAM), such as the common double data rate synchronous dynamic random access memory (DDRSDRAM), and may also include non-volatile memory (NVRAM), such as one or more read-only memories (ROMs), disk storage devices, flash memory devices, or other non-volatile solid-state memory devices such as optical disks (CD-ROMs, DVD-ROMs), floppy disks or data tapes, etc.

[0067] Those of ordinary skill in the art will appreciate that the modules and steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of the two. In order to clearly illustrate the interchangeability of hardware and software, the composition and steps of each example have been generally described in the above description according to function. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of the present invention.

[0068] In the several embodiments provided in this application, it should be understood that the disclosed methods and systems can be implemented in other ways. For example, the system embodiments described above are only schematic, for example, the division of the units is only a logical function division, and there may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed.

[0069] The above description is only a specific implementation mode of the present invention, but the protection scope of the present invention is not limited thereto. Any technician familiar with the technical field can easily think of various equivalent modifications or substitutions within the technical scope disclosed by the present invention, and these modifications or substitutions should be included in the protection scope of the present invention.

Claims

1. A cloud rendering scheduling method suitable for application update, characterized in that: The method comprises: Receive a scheduling request sent by a client, and generate a plan based on the client computing resource requirements carried in the scheduling request; Determine the application running status and initial value of application scheduling weight of each server in the server cluster, and update the server resource index data according to the planned operation and maintenance needs; Provide application scheduling services based on the new server resource index; The data of updating the server resource index according to the planned operation and maintenance needs includes: Obtain historical data on application usage status based on the client's computing resource requirements, reduce the weight of at least one application in the server based on the application's past usage, current usage, and update requirements, and issue update instructions to the server cluster; When any server in the server cluster receives an application update task, the server locally updates the application scheduling weight and reports it to the planning server; The planning server obtains the running status of each application in the server cluster and the initial value of the application scheduling weight; records the historical data of the application usage status, calculates the application scheduling weight of the application in the server resource index, and obtains the updated server resource index; The application scheduling weights of the applications in the computing server resource index include: When a server in the server cluster receives any application update task, the server is instantly prohibited from scheduling the scheduled service, and the scheduling is resumed in the next cycle, and the initial value of the application scheduling weight is updated; In combination with the current running status of the server application and according to the requirements of the plan, the initial value of the application scheduling weight is recalculated to determine the application scheduling weight of each of the server applications; Providing the application scheduling service according to the new server resource index includes: obtaining the server resource index; scheduling according to the application scheduling weight value of the application in the index, the higher the weight value of the application on the current server, the higher the application scheduling weight; The obtaining of the server resource index comprises: Record the scheduling weights of different applications. When receiving a scheduling request from a client, define the scheduling priority according to the scheduling weight value of the application requested for scheduling; when the value is 0, the application is prohibited from being scheduled on the server.

2. The method according to claim 1, characterized in that The data of the updating server resource index also includes: obtaining the current application update volume and calculating the update time; obtaining the current usage of the application and the historical usage over the same period and calculating the number of simultaneous updates; Find the server that can be updated according to the server resource index and prepare temporary cache space for the update; Calculate the maximum number of applications that can be updated at the same time, and set the updated application and server resource indexes to higher application scheduling weight values.

3. A cloud rendering scheduling system suitable for application update, characterized in that: The system comprises: A planning service module is used to receive a scheduling request sent by a client and generate a plan according to the client computing resource demand carried in the scheduling request; The update module is used to determine the application running status and the initial value of the application scheduling weight of each server in the server cluster, and update the data of the server resource index according to the planned operation and maintenance needs; The scheduling service module is used to provide application scheduling services according to the new server resource index; The data of updating the server resource index according to the planned operation and maintenance needs includes: Obtain historical data on application usage status based on the client's computing resource requirements, reduce the application scheduling weight of at least one application in the server based on the application's past usage, current usage, and update requirements, and issue update instructions to the server cluster; When any server in the server cluster receives an application update task, the server locally updates the application scheduling weight and reports it to the planning server; The planning server obtains the running status of each application in the server cluster and the initial value of the application scheduling weight; records the historical data of the application usage status, calculates the application scheduling weight of the application in the server resource index, and obtains the updated server resource index; The application scheduling weights of the applications in the computing server resource index include: When a server in the server cluster receives any application update task, the server is instantly prohibited from scheduling the scheduled service, and the scheduling is resumed in the next cycle, and the initial value of the application scheduling weight is updated; In combination with the current running status of the server application and according to the requirements of the plan, the initial value of the application scheduling weight is recalculated to determine the application scheduling weight of each of the server applications; Providing the application scheduling service according to the new server resource index includes: obtaining the server resource index; scheduling according to the application scheduling weight value of the application in the index, the higher the weight value of the application on the current server, the higher the application scheduling weight; The obtaining of the server resource index comprises: Record the application scheduling weights of different applications. When receiving a scheduling request from a client, define the scheduling priority according to the scheduling weight of the application requested for scheduling; when the value is 0, the application is prohibited from being scheduled on the server.

4. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the cloud rendering scheduling method applicable to application update period as described in any one of claims 1 to 2 are implemented.

Images