CN116700987A

CN116700987A - Computing power scheduling of cloud application and file processing method of cloud application and cloud computing platform

Info

Publication number: CN116700987A
Application number: CN202310780810.2A
Authority: CN
Inventors: 贺志成
Original assignee: Hangzhou Alibaba Feitian Information Technology Co ltd
Current assignee: Hangzhou Alibaba Feitian Information Technology Co ltd
Priority date: 2023-06-28
Filing date: 2023-06-28
Publication date: 2023-09-05

Abstract

The embodiment of the application provides a computing power scheduling method of cloud application, a file processing method of cloud application and a cloud computing platform. In the embodiment of the application, the cloud computing resources consumed by the cloud application in processing different files are considered to be different. For this reason, for the file processing scene of the cloud application, the file characteristic information of the file to be processed is automatically acquired, the computing power resource demand information required by the cloud application when processing the file is reasonably evaluated based on the file characteristic information of the file to be processed, and the cloud application is scheduled to the computing node in the cloud computing platform matched with the computing power resource demand information. Furthermore, computing power resources can be reasonably scheduled for cloud application, and the resource utilization rate of the cloud computing resources is improved. In this way, in the subsequent file processing stage, the cloud application is started on the computing node adapted to the file characteristic information of the file to be processed to perform file processing, so that the file processing efficiency of the cloud application is effectively improved, and the user experience of the cloud application is improved.

Description

Computing power scheduling of cloud application and file processing method of cloud application and cloud computing platform

Technical Field

The application relates to the technical field of cloud computing, in particular to a computing power scheduling of cloud application, a file processing method of the cloud application and a cloud computing platform.

Background

With the rapid development of cloud computing technology, more and more enterprises and individuals choose to deploy application programs to a cloud computing platform, and the computing capacity of terminal equipment is expanded through cloud computing resources provided by the cloud computing platform, so that more efficient data processing and application running capacity are realized. Applications deployed on cloud computing platforms, also referred to as cloud applications or cloud applications, may be used by users for a wide variety of file processing, such as file browsing, image processing, video editing, video compression or video transcoding, and so forth. In practical application, when the cloud application processes different files, the consumed cloud computing resources are different. In a file processing scene of a cloud application, reasonably scheduling cloud computing resources for the cloud application is always a research hotspot.

Disclosure of Invention

The aspects of the application provide a computing power scheduling of a cloud application, a file processing method of the cloud application and a cloud computing platform, which are used for reasonably scheduling computing power resources for the cloud application, so that the resource utilization rate of the cloud computing resources is improved, the file processing efficiency of the cloud application can be effectively improved in the subsequent file processing stage, and the user experience of the cloud application is improved.

The embodiment of the application provides a file processing method of cloud application, which comprises the following steps: responding to a file processing request which is sent by a user through a cloud application client and comprises a target file identifier, and acquiring file characteristic information of a target file corresponding to the target file identifier; determining computing power resource demand information of the cloud application according to file characteristic information of the target file; screening target computing nodes meeting the computing power resource demand information from all computing nodes of the cloud computing platform; and scheduling the cloud application to the target computing node, and starting the cloud application scheduled to the target computing node to process the target file.

The embodiment of the application also provides a computing power scheduling method of the cloud application, which comprises the following steps: acquiring a target file identifier sent by a user through a cloud application client, and acquiring file characteristic information of a target file corresponding to the target file identifier; determining computing power resource demand information of the cloud application according to file characteristic information of the target file; screening target computing nodes meeting the computing power resource demand information from all computing nodes of the cloud computing platform; the cloud application is scheduled to the target computing node.

The embodiment of the application also provides a cloud computing platform, which comprises: the cloud application management and control device, the cloud storage system and the computing power resource pool comprise a plurality of computing nodes; the cloud application management and control device is used for responding to a file processing request which is sent by a user through a cloud application client and comprises a target file identifier, and acquiring file characteristic information of a target file corresponding to the target file identifier; determining computing power resource demand information of the cloud application according to file characteristic information of the target file; sending a resource application request comprising resource demand information to a computing power resource pool, receiving resource application response information returned by the computing power resource pool, scheduling cloud application to a target computing node according to a node identifier of the target computing node in the resource application response information, starting the cloud application scheduled to the target computing node, loading file data of a target file from a cloud storage system by the cloud application after starting, and carrying out file processing on the file data of the target file; and the computing power resource pool is used for responding to the resource application request, screening target computing nodes meeting the computing power resource demand information from all the computing nodes, and returning resource application response information to the cloud application management and control device.

The embodiment of the application also provides a computer device, which comprises: a memory and a processor; a memory for storing a computer program; the processor is coupled to the memory for executing the computer program for executing steps in a file processing method of the cloud application.

Embodiments of the present application also provide a computer-readable storage medium storing a computer program, which when executed by a processor, causes the processor to implement steps in a file processing method of a cloud application.

In the embodiment of the application, the cloud computing resources consumed by the cloud application in processing different files are considered to be different. For this reason, for the file processing scene of cloud application, the file characteristic information of the file to be processed is automatically obtained, the computing power resource demand information required by the cloud application when processing the file is reasonably evaluated based on the file characteristic information of the file to be processed, and the cloud application is scheduled to the computing nodes in the cloud computing platform matched with the computing power resource demand information, so that the computing power resource can be reasonably scheduled for the cloud application, and the resource utilization rate of the cloud computing resource is improved. In this way, in the subsequent file processing stage, the cloud application is started on the computing node adapted to the file characteristic information of the file to be processed to perform file processing, so that the file processing efficiency of the cloud application is effectively improved, and the user experience of the cloud application is improved.

Drawings

The accompanying drawings, which are included to provide a further understanding of the application and are incorporated in and constitute a part of this specification, illustrate embodiments of the application and together with the description serve to explain the application and do not constitute a limitation on the application. In the drawings:

FIG. 1 is an exemplary application scenario diagram provided by an embodiment of the present application;

fig. 2 is a flowchart of a file processing method of a cloud application according to an embodiment of the present application;

fig. 3 is a flowchart of a computing power scheduling method of a cloud application according to an embodiment of the present application;

FIG. 4 is a schematic view of a scenario of computing power scheduling and file processing for an exemplary cloud application provided by an embodiment of the present application;

fig. 5 is a schematic structural diagram of a cloud computing platform according to an embodiment of the present application;

fig. 6 is a schematic structural diagram of a computer device according to an embodiment of the present application.

Detailed Description

In order to make the objects, technical solutions and advantages of the present application more apparent, the technical solutions of the present application will be clearly and completely described below with reference to specific embodiments of the present application and corresponding drawings. It will be apparent that the described embodiments are only some, but not all, embodiments of the application. All other embodiments, which can be made by those skilled in the art based on the embodiments of the application without making any inventive effort, are intended to be within the scope of the application.

In embodiments of the present application, "at least one" means one or more, and "a plurality" may mean two or more. "and/or" describes the access relationship of the associated objects, meaning that there may be three relationships, e.g., a and/or B, which may represent: there are three cases where a alone exists, where a and B exist together, and where B alone exists, where a, B may be singular or plural. In the text description of the present application, the character "/" generally indicates that the front-rear associated object is an or relationship. In addition, in the embodiments of the present application, "first", "second", "third", etc. are only for distinguishing contents of different objects, and have no other special meaning.

The following description of some words related to embodiments of the present application is provided:

cloud application (also referred to as cloud application): the method is a novel application for changing the use mode of the traditional software of local installation and local operation into the service of 'instant taking and instant using', connecting and controlling a remote service cluster through the Internet and completing technical logic or operation tasks. The cloud application is installed on the cloud computing platform without the need for a user to install on the local device, so that the computing power resources and the storage pressure of the local device are greatly released. Cloud applications include, for example, but are not limited to: an instant messaging cloud-like application, a cloud gaming application, a cloud rendering application, a cloud desktop application, a live broadcast cloud-like application, or a video play cloud-like application, and the like.

Cloud application client: refers to a client of a cloud application, responsible for interacting with a user, requesting access to the cloud application, and so forth.

Cloud computing platform (also simply referred to as cloud): is an information system integrating various hardware resources and software resources based on cloud computing technology. Hardware resources include, for example, but are not limited to: server devices, storage devices, network devices, etc. Software resources include, for example, but are not limited to: operating systems, integrated development environments, middleware, application software, and the like. Cloud computing platforms have characteristics of high concurrency, large user volume, etc., and end users can meet the demands of computing resources, storage resources, or other infrastructure of application programs through the cloud computing platform.

Cloud storage systems refer to storage systems deployed in cloud computing platforms, including, for example, but not limited to: a distributed file storage system deployed in a cloud computing platform.

Cloud application management and control device: the cloud application system provides full life cycle management capability of cloud application development, release, deployment and the like, full life cycle management capability of resources such as purchase creation, elastic expansion and contraction capacity, dynamic scheduling and the like of cloud resources, and security monitoring management capability of access control and the like.

And (3) calculating a power resource pool: the method is a shared pool for centralizing cloud computing resources, so that the resources are conveniently and efficiently distributed, and the resource utilization rate and scheduling performance can be improved.

A distributed file storage system: the file can be stored in a plurality of nodes in a scattered way, the requirements of large-scale data storage and access can be processed, a high-degree data redundancy and fault tolerance mechanism is provided, and the safety of data is ensured.

In practical application, when the cloud application processes different files, the consumed cloud computing resources are different. In a file processing scene of a cloud application, reasonably scheduling cloud computing resources for the cloud application is always a research hotspot.

Therefore, the embodiment of the application provides a computing power scheduling method of cloud application and a file processing method of cloud application and a cloud computing platform. In the embodiment of the application, the cloud computing resources consumed by the cloud application in processing different files are considered to be different. For this reason, for the file processing scene of cloud application, the file characteristic information of the file to be processed is automatically obtained, the computing power resource demand information required by the cloud application when processing the file is reasonably evaluated based on the file characteristic information of the file to be processed, and the cloud application is scheduled to the computing nodes in the cloud computing platform matched with the computing power resource demand information, so that the computing power resource can be reasonably scheduled for the cloud application, and the resource utilization rate of the cloud computing resource is improved. In this way, in the subsequent file processing stage, the cloud application is started on the computing node adapted to the file characteristic information of the file to be processed to perform file processing, so that the file processing efficiency of the cloud application is effectively improved, and the user experience of the cloud application is improved.

Fig. 1 is an exemplary application scenario diagram provided in an embodiment of the present application. Referring to fig. 1, a user may access a cloud application in a cloud computing platform through a cloud application client in a terminal device, for example, the user requests the cloud application to perform file processing through the cloud application client. Referring to fig. 1, a cloud computing platform may include cloud application management and control devices, cloud storage systems (e.g., as a distributed file storage system), and a pool of computing resources, among others.

In practical application, referring to (1) in fig. 1, file data of a file stored locally in a terminal device can be synchronized to a cloud storage system, so that a cloud application can access the file data of a file to be processed nearby when the file is processed, that is, access to the cloud storage system can obtain the file data of the file to be processed, the file data of the file to be processed in the terminal device is not required to be accessed, and file processing efficiency of the cloud application is accelerated. The file stored locally in the terminal device is referred to as a user local file, and the file in the cloud storage system is referred to as a user cloud file. In practical applications, no requirement is made on the time for synchronizing file data in the local file of the user to the file on the user cloud, for example, the synchronization time may be before, during or after the user requests the cloud application to process the file through the application client, which is not limited.

Referring to (2) in fig. 1, when a user has a need to use a cloud application to process a file, the user triggers a cloud application client to send a file processing request to a cloud computing platform, where the file processing request includes a file identifier of a file to be processed. And searching file characteristic information of each file stored in advance according to the file identification of the file to be processed by a cloud application management and control device in the cloud computing platform, and if the file characteristic information of the file to be processed is searched, executing computing power resource scheduling operation. If the file characteristic information of the file to be processed is not found, the file characteristic information of the file to be processed is extracted through a file characteristic extraction module in the cloud application management and control device, and then the computing resource scheduling operation is executed. Specifically, referring to (3) in fig. 1, the file characteristic extraction module obtains file metadata of a file to be processed from the cloud storage system according to a file identifier of the file to be processed; referring to (4) in fig. 1, the file characteristic extraction module performs characteristic extraction on file data of a file to be processed in the cloud storage system to obtain key file characteristic information of the file to be processed. Referring to (5) in fig. 1, the file characteristic extraction module takes file metadata and key file characteristic information of the file to be processed as file characteristic information of the file to be processed and stores the file metadata and the key file characteristic information.

Referring to (6) in fig. 1, in the case that the file characteristic information of the file to be processed is obtained, the cloud application management and control device invokes the computing power resource scheduling module to analyze the file characteristic information of the file to be processed, and determines computing power resource requirement information of the cloud application to process the file to be processed, where the computing power resource requirement information includes, for example, but not limited to: the utilization of the central processing unit (Central Processing Unit, CPU) of the required compute node, the utilization of the graphics processor (graphics processing unit, GPU), the remaining space of the memory, and the remaining space of the disk. It can be understood that the usage rate of the central processor of the computing node, the usage rate of the graphics processor, the remaining space of the memory or the remaining space of the disk are different, and the specifications of the corresponding computing nodes are different.

Referring to (7) in fig. 1, the computing power resource scheduling module sends a resource application request including resource demand information to the algorithm resource pool, and the computing power resource scheduling module allocates computing nodes matched with the resource demand information to the cloud application in response to the resource application request.

After the resource allocation is finished, referring to (8) in fig. 1, the computing power resource scheduling module sends a cloud application distribution instruction to a cloud application distribution module in the cloud application management and control device. The cloud application distribution module is responsible for tasks such as cloud application mirror image distribution, application deployment, configuration, start-stop and the like. Referring to (9) in fig. 1, the cloud application distribution module distributes the cloud application to the computing nodes matched with the resource demand information in response to the application distribution instruction, and starts the cloud application on the corresponding computing nodes.

In the file processing stage, referring to a graphic display in fig. 1, the cloud application after being started loads file data of the files to be processed from the cloud storage system, and the number of the files to be processedAnd (5) processing the file according to the data. See fig. 1As shown, data acquisition such as ASP (Adaptive Streaming Protocol) generated in the cloud application file processing process is performed to obtain Streaming data of the cloud application protocol, and the Streaming data of the cloud application protocol is transmitted to a cloud application client through Gateway (Gateway) service. The ASP protocol runs on a TCP (Transmission Control Protocol )/UDP (User Datagram Protocol, user datagram protocol)) network protocol, and by means of the current efficient codec technology, the graphical interactive interface on the cloud is encoded and pushed to the client in a Streaming (Streaming) manner, and the client performs decoding and displays the graphical interface.

See fig. 1The cloud application client analyzes the streaming data of the cloud application protocol to obtain a picture in the cloud application file processing process and update the displayed file process picture in real time. So far, the whole file processing process is ended.

It should be noted that, the application scenario shown in fig. 1 is only an exemplary application scenario, and the embodiment of the present application is not limited to the application scenario. The embodiment of the present application does not limit the devices included in fig. 1, nor does it limit the positional relationship between the devices in fig. 1.

In the embodiment of the application, the terminal equipment can interact with the cloud computing platform through a wired network or a wireless network. For example, the wired network may include coaxial cable, twisted pair, optical fiber, and the like, and the wireless network may be a 2G (2 generation ) network, a 3G (3 generation ) network, a 4G (4 generation ) network, or a 5G (5 generation ) network, a wireless fidelity (Wireless Fidelity, abbreviated WIFI) network, and the like. The application is not limited to the specific type or specific form of interaction, as long as the interaction function between the terminal equipment and the cloud computing platform can be realized. In addition, the terminal device may be hardware or software. When the terminal device is hardware, the terminal device is, for example, a mobile phone, a tablet computer, a desktop computer, a wearable intelligent device, an intelligent home device, or the like. When the terminal device is software, it may be installed in the above-listed hardware device, and in this case, the user device is, for example, a plurality of software modules or a single software module, etc., and the embodiment of the present application is not limited.

The following describes in detail the technical solutions provided by the embodiments of the present application with reference to the accompanying drawings.

Fig. 2 is a flowchart of a file processing method of a cloud application according to an embodiment of the present application. The method is applied to a cloud computing platform, see fig. 2, and can comprise the following steps:

201. and responding to a file processing request which is sent by a user through the cloud application client and comprises the target file identification, and acquiring file characteristic information of the target file corresponding to the target file identification.

202. And determining computing power resource demand information of the cloud application according to the file characteristic information of the target file.

203. And screening target computing nodes meeting the computing power resource demand information from all computing nodes of the cloud computing platform.

204. And scheduling the cloud application to the target computing node, and starting the cloud application scheduled to the target computing node to process the target file.

In this embodiment, the user may request, through the cloud application client, the cloud application to perform file processing. The file to be subjected to file processing is referred to herein as a target file, which may be a plain text file or a rich text file, without limitation. In the case where the target file is a rich text file, the target file includes, for example, but is not limited to: video format file, picture format file, PDF (portable document format ) format file, PPT (Power Point) format file, wherein PPT format file is a presentation.

When a user has a file processing requirement, the user can trigger a file processing request through the cloud application client, wherein the file processing request is used for requesting the cloud application to perform file processing on a target file, and the file processing request comprises a target file identifier (namely, the file identifier of the target file). In some scenarios, the number of cloud applications of the cloud computing platform may be multiple, and for this case, an application identifier of the cloud application may be carried in the file processing request to explicitly indicate that file processing is performed with a certain cloud application.

In this embodiment, the cloud computing platform responds to a file processing request sent by a user through a cloud application client to obtain file characteristic information of a target file corresponding to a target file identifier. The document property information may be capable of characterizing representation information of the document, including, for example, but not limited to: file Metadata (Metadata) describing characteristics of a file and key file characteristic information extracted by performing characteristics on file data of the file. File metadata includes, for example, but is not limited to: file type, file size, picture size, image resolution, encoding format, code rate or frame rate, etc.

In this embodiment, when the cloud computing platform obtains the file characteristic information of the target file corresponding to the target file identifier, the cloud computing platform may obtain the file metadata information of the target file from the cloud storage system, and use the file metadata information of the target file as the file characteristic information of the target file. Or the cloud computing platform performs feature extraction on the file data of the target file cached in the cloud storage system to obtain key file feature information of the target file, and takes the key file feature information as file feature information of the target file. Alternatively, in order to better characterize the portrait information of the file, the cloud computing platform may use file metadata information and key file feature information of the target file as file characteristic information of the target file.

In practical application, various feature extraction algorithms can be adopted to extract features of file data of the file. Feature extraction algorithms include, for example, but are not limited to: LBP (Local Binary Pattern ) feature extraction algorithm, SIFT (Scale Invariant Feature Transform, scale-invariant feature transform matching) feature extraction algorithm, TF-IDF (term frequency-inverse document frequency) feature extraction algorithm, mutual information (Mutual Information) based feature extraction algorithm, and the like.

In this embodiment, a plurality of feature extraction algorithms of different algorithm capabilities may be provided, including, for example, but not limited to: accuracy, extraction efficiency, etc. Further optionally, in order to perform feature extraction better, feature extraction is performed on file data of the target file cached in the cloud storage system, and an optional implementation manner for obtaining key file feature information of the target file is as follows: selecting a target feature extraction algorithm with algorithm performance matched with file metadata information of a target file from a plurality of feature extraction algorithms; and extracting the characteristics of the file data of the target file cached in the cloud storage system by using a target characteristic extraction algorithm to obtain the key file characteristic information of the target file.

Specifically, the file metadata information applicable to the algorithm performance of each feature extraction algorithm can be flexibly set in advance as required, so that in the algorithm selection stage, a feature extraction algorithm matched with the file metadata information of the target file is selected as a target feature extraction algorithm, and feature extraction is performed by using the target feature extraction algorithm. Or in the algorithm selection stage, the algorithm performance adapted to the target file can be estimated according to the file metadata of the target file, and a feature extraction algorithm corresponding to the algorithm performance adapted to the target file is selected for feature extraction. For example, a plain text file may have lower accuracy and feature extraction efficiency than a rich text file, or a large file may have higher accuracy and feature extraction efficiency than a small file, and so on.

Further optionally, in order to improve file processing efficiency, after the cloud computing platform acquires the file characteristic information of the file, the file characteristic information of the file can be stored in the cloud computing platform, so that the file characteristic information of the required file can be quickly acquired from the file characteristic information of each stored file.

Based on the above, as an example, when the cloud computing platform obtains the file characteristic information of the target file corresponding to the target file identifier, the cloud computing platform may search the file characteristic information of the target file from the file characteristic information of each file already stored in the cloud computing platform according to the target file identifier; if the file characteristic information of the target file is not found in the cloud computing platform, acquiring file metadata information of the target file from a cloud storage system; taking file metadata information of the target file as file characteristic information of the target file, and associating and storing target file identification and file characteristic information of the target file in a cloud computing platform. Of course, if the file characteristic information of the target file is found in the cloud computing platform, a subsequent step of determining the computing power resource requirement information may be performed.

As another example, when the cloud computing platform obtains the file characteristic information of the target file corresponding to the target file identifier, the cloud computing platform may search the file characteristic information of the target file from the file characteristic information of each file already stored in the cloud computing platform according to the target file identifier; if the file characteristic information of the target file is not found in the cloud computing platform, acquiring file metadata information of the target file from the cloud storage system, and extracting features of file data of the target file cached in the cloud storage system to obtain key file feature information of the target file; taking file metadata information and key file characteristic information of the target file as file characteristic information of the target file, and storing target file identification and file characteristic information of the target file in a cloud computing platform in a correlation manner. Of course, if the file characteristic information of the target file is found in the cloud computing platform, a subsequent step of determining the computing power resource requirement information may be performed.

In this embodiment, computing power resource requirement information of the cloud application is determined according to file characteristic information of the target file. The computing power resource requirement information describes resource information of computing nodes required by the cloud application to process the target file, and includes, for example, but is not limited to: the CPU utilization rate of the required computing node, the GPU utilization rate, the residual space of the memory and the residual space of the disk. Optionally, the computing power resource demand information may also indicate whether the demanded computing node is a graphics class computing node with GPU or a normal computing node without GPU. The graphic class computing node refers to a computing node with a GPU, and the common computing node refers to a computing node without the GPU. For example, if the file type of the target file is a rich text type, the file characteristic information of the target file includes at least one of the following: file size, picture size, image resolution, encoding format, code rate, frame rate, and key file feature information, the computing power resource requirement information indicates that the graphics class compute node with GPU is the target compute node for the requirement. If the file type of the target file is a plain text type, the file characteristic information of the target file comprises at least one of the following: file size and key file feature information, the computing power resource requirement information indicates that a common computing node without a GPU is taken as a target computing node for the requirement.

In practical application, the computing power resource demand information corresponding to each feature data in the file characteristic information can be analyzed according to expert experience, so that the computing power resource demand information corresponding to each feature data in the file characteristic information is weighted and summed to determine the computing power resource demand information corresponding to the cloud application processing target file. For example, the file characteristic information includes: the method comprises the steps of analyzing file type, file size, picture size, image resolution, coding format, code rate, frame rate and key file characteristic information obtained by characteristic extraction of file data according to expert experience, calculating power resource demand information corresponding to each characteristic data such as file type, file size, picture size, image resolution, coding format, code rate, frame rate and key file characteristic information, weighting and summing the calculating power resource demand information corresponding to each characteristic data, and obtaining calculating power resource demand information corresponding to a cloud application processing target file.

In practical application, the method can collect a plurality of file characteristic information which is subjected to file processing and the marked computing power resource information thereof, and perform model training by utilizing the file characteristic information which is subjected to file processing and the marked computing power resource information thereof to obtain a machine learning model with the computing power resource requirement information determining function. In this way, the machine learning model can be utilized to determine computing power resource demand information corresponding to the cloud application processing target file.

Of course, in practical application, computing power resource demand information of the cloud application can be flexibly determined according to file characteristic information of the target file, and the method is not limited. For example, for a file of rich text type, computing power resource requirement information of the cloud application is determined according to one or more characteristic information such as file size, picture size, image resolution, encoding format, code rate, frame rate, key file characteristic information, and the like. For example, for a file of a plain text type, computing power resource requirement information of the cloud application is determined according to one or more feature information such as file size and key file feature information. Further optionally, in order to efficiently and accurately determine the computing power resource requirement information, an optional implementation manner of determining the computing power resource requirement information of the cloud application according to the file characteristic information of the target file is as follows: determining the calculation complexity of the target file according to the file characteristic information of the target file; and determining computing power resource demand information of the cloud application according to the computing complexity of the target file.

In practical application, the calculation complexity corresponding to each feature data in the file characteristic information can be analyzed according to expert experience, and weighted summation is carried out on the calculation complexity corresponding to each feature data in the file characteristic information, so that the calculation complexity of the target file is obtained.

In practical application, the calculation complexity of the characteristic information and the labels of the plurality of files can be collected, and model training is performed by using the calculation complexity of the characteristic information and the labels of the plurality of files, so that a machine learning model with a calculation complexity determining function is obtained. In this way, the computational complexity of the target file may be determined using a machine learning model.

Further optionally, in order to efficiently and accurately determine the computational complexity of the file, an alternative implementation manner of determining the computational complexity of the target file according to the file characteristic information of the target file is: determining a target calculation complexity determination mode matched with the file type of the target file from calculation complexity determination modes corresponding to the file types according to the file types in the file characteristic information of the target file; and processing the file characteristic information of the target file by using a target calculation complexity determination mode so as to determine the calculation complexity of the target file.

In practical application, the calculation complexity determination mode of each file type can be flexibly set according to the requirement. For example, for a file of plain text type, the computational complexity determination is: determining the calculation complexity corresponding to the file size and the calculation complexity corresponding to the key file characteristic information respectively; and carrying out weighted summation on the calculation complexity corresponding to the file size and the calculation complexity corresponding to the key file characteristic information to obtain the calculation complexity of the whole file. Wherein, the larger the file size, the greater the computational complexity; the more abundant the key file feature information, the greater the computational complexity.

For example, for a file of rich text type, the computational complexity determination is: determining the calculation complexity corresponding to one or more characteristic information such as file size, picture size, image resolution, coding format, code rate, frame rate, key file characteristic information and the like respectively; and carrying out weighted summation on the computation complexity corresponding to each piece of characteristic information to obtain the computation complexity of the whole file.

For example, for a file of plain text type, the computational complexity determination is: determining the calculation complexity corresponding to one or more feature information such as file size, key file feature information and the like respectively; and carrying out weighted summation on the computation complexity corresponding to each piece of characteristic information to obtain the computation complexity of the whole file.

It should be noted that the file types can be flexibly divided as required. For example, file types are classified into plain text types and rich text types. For another example, the file is divided into a large file and a small file according to the file size division. For another example, the division may be made into h.264 and h.265 according to coding formats. Wherein, H.264: is a new video compression coding standard that has increased motion picture compression technology to a higher level, providing high quality image transmission over lower bandwidths. The H.264 coding saves more code stream, has stronger error code resistance, and can be suitable for video transmission in wireless channels with high packet loss rate and serious interference, thereby obtaining stable image quality. 265: the method is a new video compression coding standard, H.265 surrounds the video coding standard H.264, certain original technologies are reserved, and certain aspects such as code stream, coding quality, delay and the like are improved and optimized by using the new technologies, so that the compression efficiency is improved, the robustness and error recovery capability are enhanced, the real-time delay is reduced, the complexity is reduced and the like.

In this embodiment, the file characteristic information of the target file is processed by using a target calculation complexity determination method matched with the file type of the target file, so as to determine the calculation complexity of the target file. And after the calculation complexity of the target file is determined, determining the calculation power resource demand information of the cloud application according to the calculation complexity of the target file.

In practical application, the computing power resource requirement information required by the computing complexity of the target file can be analyzed according to expert experience. The method can also collect the computational complexity of the files subjected to file processing and the marked computational complexity resource information of the files subjected to file processing, and perform model training by utilizing the computational complexity of the files subjected to file processing and the marked computational complexity resource information of the files subjected to file processing, so as to obtain a machine learning model with a computational power resource demand information determining function. In this way, the machine learning model can be utilized to determine computing power resource demand information corresponding to the cloud application processing target file.

Further optionally, in order to efficiently and accurately determine the computing power resource requirement information, a computing complexity range corresponding to each computing complexity level may be preset, and a corresponding relationship between the computing complexity level and the computing power resource requirement information may be established. In this way, when determining the computing power resource demand information corresponding to the target file processed by the cloud application, the computing complexity level of the target file can be determined according to the computing complexity range in which the computing complexity of the target file falls, and the computing power resource demand information corresponding to the target file processed by the cloud application can be determined according to the corresponding relation between the computing complexity level of the target file and the computing power resource demand information.

Further optionally, in order to efficiently and accurately determine the computing power resource requirement information of the cloud application, an optional implementation manner of determining the computing power resource requirement information of the cloud application according to the computing complexity of the target file is as follows: and determining computing power resource demand information of the cloud application according to the computing complexity of the target file and node information of each computing node of the cloud computing platform, wherein the node information comprises load pressure and/or residual resource information. For example, the CPU load pressure, GPU load pressure, memory load pressure or disk load pressure, the remaining utilization of the CPU, the remaining utilization of the GPU, the remaining space of memory and the remaining space of disk, etc. of the compute node.

In practical application, the computing complexity of the target file and the node information of each computing node of the cloud computing platform can be comprehensively analyzed to determine the computing power resource demand information of the cloud application. The initial computing power resource demand information can be determined based on the computing complexity of the target file, and then the node information of each computing node is utilized to optimize the initial computing power resource demand information, so that the final computing power resource demand information of the cloud application is obtained. And the training data comprises the calculation complexity of the file, the node information of each calculation node of the cloud calculation platform and the marked computing power resource information, and the model training is carried out by utilizing the training data to obtain a machine learning model with the computing power resource requirement information determining function. In this way, the machine learning model can be utilized to determine computing power resource demand information corresponding to the cloud application processing target file. Of course, in practical application, computing power resource demand information of the cloud application can be flexibly determined according to the computing complexity of the target file and node information of each computing node of the cloud computing platform, and the computing power resource demand information is not limited.

In this embodiment, after determining computing power resource demand information required when the cloud application processes the target file, a target computing node satisfying the computing power resource demand information is selected from all computing nodes of the cloud computing platform; and scheduling the cloud application to the target computing node, and starting the cloud application scheduled to the target computing node to process the target file.

As an example, screening out target computing nodes that satisfy computing power resource demand information from respective computing nodes of a cloud computing platform includes: if the file type of the target file is rich text type, the file characteristic information of the target file comprises at least one of the following: the method comprises the steps that file size, picture size, image resolution, coding format, code rate, frame rate and key file characteristic information, and the computing power resource demand information indicates that a graphic class computing node with a GPU is used as a target computing node of demand, and then the graphic class computing node with the GPU meeting the computing power resource demand information is screened out from all computing nodes of a cloud computing platform to be used as the target computing node; if the file type of the target file is a plain text type, the file characteristic information of the target file comprises at least one of the following: and the file size and the key file characteristic information, wherein the computing power resource demand information indicates that a common computing node without GPU is taken as a target computing node for demand, and the common computing node without GPU which meets the computing power resource demand information is screened out from all computing nodes of the cloud computing platform to be taken as the target computing node.

Further optionally, in order to improve the resource utilization rate and the scheduling performance, when the target computing node meeting the computing power resource requirement information is screened out from the computing nodes of the cloud computing platform, a resource application request may be sent to the computing power resource pool in the cloud computing platform, so that the computing power resource pool screens out the target computing node meeting the computing power resource requirement information from the computing nodes therein.

Further optionally, in order to improve the resource utilization rate, after the cloud application finishes processing the target file, resource recovery may also be performed. Specifically, in response to a closing request of a cloud application sent by a user through a cloud application client, closing the cloud application, and releasing computing resources occupied by the cloud application in a target computing node.

According to the technical scheme provided by the embodiment of the application, the fact that cloud computing resources consumed by cloud applications when processing different files are different is considered. For this reason, for the file processing scene of the cloud application, the file characteristic information of the file to be processed is automatically obtained, the computing power resource demand information required by the cloud application when processing the file is reasonably evaluated based on the file characteristic information of the file to be processed, the cloud application is scheduled to the computing nodes in the cloud computing platform matched with the computing power resource demand information, and the cloud application is started to process the file, so that the file processing efficiency of the cloud application and the resource utilization rate of the cloud computing resource are effectively improved, and the user experience of the cloud application is improved.

Fig. 3 is a flowchart of a computing power scheduling method of a cloud application according to an embodiment of the present application. The method is applied to a cloud computing platform, see fig. 3, and can comprise the following steps:

301. and acquiring a target file identifier sent by a user through the cloud application client, and acquiring file characteristic information of a target file corresponding to the target file identifier.

302. And determining computing power resource demand information of the cloud application according to the file characteristic information of the target file.

303. And screening target computing nodes meeting the computing power resource demand information from all computing nodes of the cloud computing platform.

304. The cloud application is scheduled to the target computing node.

In practical application, a user can send a corresponding target file identifier of a target file to the cloud computing platform through the cloud application client at any time, for example, when the user has a processing requirement on the target file, the user sends the target file identifier corresponding to the target file to the cloud computing platform through the cloud application client, which is not limited. For more description of determining file characteristic information of the target file, computing power resource requirement information of the cloud application, screening the target computing nodes meeting the computing power resource requirement information, and cloud application scheduling, reference is made to the foregoing, and details are not repeated here.

Further optionally, determining the computing power resource requirement information of the cloud application according to the file characteristic information of the target file includes: determining the calculation complexity of the target file according to the file characteristic information of the target file; and determining computing power resource demand information of the cloud application according to the computing complexity of the target file.

Further optionally, determining computing power resource requirement information of the cloud application according to the computing complexity of the target file includes: and determining computing power resource demand information of the cloud application according to the computing complexity of the target file and node information of each computing node of the cloud computing platform, wherein the node information comprises load pressure and/or residual resource information.

Further optionally, determining the computational complexity of the target file according to the file characteristic information of the target file includes: determining a target calculation complexity determination mode matched with the file type of the target file from calculation complexity determination modes corresponding to the file types according to the file types in the file characteristic information of the target file; and processing the file characteristic information of the target file by using a target calculation complexity determination mode so as to determine the calculation complexity of the target file.

Further optionally, obtaining the file characteristic information of the target file corresponding to the target file identifier includes: searching file characteristic information of the target file from file characteristic information of each file stored by the cloud computing platform according to the target file identification; if the file characteristic information of the target file is not found in the cloud computing platform, acquiring file metadata information of the target file from a cloud storage system; taking file metadata information of the target file as file characteristic information of the target file, and associating and storing target file identification and file characteristic information of the target file in a cloud computing platform.

Further optionally, before taking the file metadata information of the target file as the file characteristic information of the target file, the method further includes: extracting features of file data of the target file cached in the cloud storage system to obtain key file feature information of the target file; accordingly, taking the file metadata information of the target file as the file characteristic information of the target file includes: and taking the file metadata information and the key file characteristic information of the target file as file characteristic information of the target file.

Further optionally, feature extraction is performed on file data of the target file cached in the cloud storage system to obtain key file feature information of the target file, including: selecting a target feature extraction algorithm with algorithm performance matched with file metadata information of a target file from a plurality of feature extraction algorithms; and extracting the characteristics of the file data of the target file cached in the cloud storage system by using a target characteristic extraction algorithm to obtain the key file characteristic information of the target file.

Further optionally, after the cloud application finishes processing the target file, the method further includes: and responding to a closing request of the cloud application sent by a user through the cloud application client, closing the cloud application, and releasing the computing resources occupied by the cloud application in the target computing node.

Further optionally, selecting a target computing node that meets the computing power resource requirement information from the computing nodes of the cloud computing platform includes: and sending a resource application request to a computing power resource pool in the cloud computing platform so that the computing power resource pool screens out target computing nodes meeting computing power resource demand information from all computing nodes in the computing power resource pool.

Further optionally, selecting a target computing node that meets the computing power resource requirement information from the computing nodes of the cloud computing platform includes: if the file type of the target file is rich text type, the file characteristic information of the target file comprises at least one of the following: file size, picture size, image resolution, coding format, code rate, frame rate and key file characteristic information, wherein the computing power resource demand information indicates that a graphic class computing node with a GPU is taken as the target computing node for demand, and then the graphic class computing node with the GPU meeting the computing power resource demand information is screened out from all computing nodes of a cloud computing platform to be taken as the target computing node; if the file type of the target file is a plain text type, the file characteristic information of the target file comprises at least one of the following: and the computing power resource demand information indicates that the common computing node without the GPU is taken as the target computing node for demand, and the common computing node without the GPU meeting the computing power resource demand information is screened out from all computing nodes of the cloud computing platform to be taken as the target computing node.

For the implementation of each step in the embodiment shown in fig. 3, reference may be made to the implementation of each step in the embodiment shown in fig. 2, which is not described herein.

According to the computing power scheduling method of the cloud application, aiming at a file processing scene of the cloud application, file characteristic information of a file to be processed is automatically obtained, computing power resource demand information required by the cloud application when the file is processed is reasonably evaluated based on the file characteristic information of the file to be processed, and the cloud application is scheduled to a computing node in a cloud computing platform matched with the computing power resource demand information. Furthermore, computing power resources can be reasonably scheduled for cloud application, and the resource utilization rate of the cloud computing resources is improved.

Fig. 4 is a schematic view of a scenario of computing power scheduling and file processing of an exemplary cloud application according to an embodiment of the present application.

In practical application, the cloud computing platform comprises a cloud application management and control device, a cloud storage system and a computing resource pool, wherein the cloud application management and control device comprises a file characteristic extraction module, a computing resource scheduling module and other modules. For an introduction to a cloud computing platform, reference may be made to the relevant content of the foregoing embodiments.

S1, receiving a file access request.

When a user has a file processing requirement, an access request can be sent to the cloud application management and control device through the cloud application client, and the access request can comprise a file access path of a file to be accessed. When the file access path indicates to access a file on the cloud, step S2 is executed.

S2, judging whether the file characteristic information is extracted.

And the cloud application management and control device judges whether the file characteristic information of the file to be processed is already extracted, if not, the file characteristic extraction module is called to extract the file characteristic information on line and extract the file characteristic information off line, the file metadata can be quickly extracted by extracting the file characteristic information on line, and the file characteristic information can be extracted by extracting the file characteristic information off line through an intelligent algorithm. In addition, the file characteristic information that has been extracted is cached, and step S3 is performed. If yes, step S3 is also performed.

It should be noted that, the online extraction of the file characteristic information may also be performed by using a lightweight intelligent algorithm (which is less time-consuming to perform), which is not limited. When the file characteristic information is extracted offline, an intelligent algorithm with higher precision can be selected for extracting the file characteristic.

In addition, it can be appreciated that the file characteristic information is cached in advance for direct use when the user accesses, so that the influence on the application starting speed can be further reduced.

S3, obtaining the cached file characteristic information.

S4, analyzing the file characteristic information.

The cloud application management and control device calls the computing power resource scheduling module to analyze file characteristic information.

S5, judging whether the file type is a rich text file. If yes, go to step S6, if no, go to step S7.

Different text characteristic information-based automatic calculation force decision logics are arranged for different file types, and in addition, multiple pieces of characteristic information participating in calculation complexity calculation can be subjected to weighted analysis so as to improve the accuracy of calculation force decision.

S6, determining the calculation complexity based on the file characteristic information such as the image resolution, the coding format, the video frame rate and the like, and executing the step S8.

And S7, determining the calculation complexity based on file characteristic information such as file size or key file characteristic information, and executing the step S8.

S8, calculating force resource requirement information required by the process of deciding and processing the file.

The computing power resource requirement information is, for example, a graphic class computing node with a GPU or a general computing node without a GPU, and specification information of the required computing node, such as high specification, standard specification and low specification. The method comprises the steps of sorting available computing resources from small to large, and sequentially arranging low-specification computing nodes, standard-specification computing nodes and high-specification computing nodes. The graphic class computing node refers to a computing node with a GPU, and the common computing node refers to a computing node without the GPU. For example, a file of the rich text type requires computing nodes using graphics classes with GPUs, while a file of the plain text type selects normal computing nodes without GPUs.

S9, corresponding computing power resources are scheduled.

S10, completing the distribution of the computing power resources.

And the computing power resource scheduling module applies the required computing power resources to the computing power resource pool according to the required computing power resource demand information and finishes computing power resource allocation.

S11, starting the cloud application.

For example, a cloud application distribution module is invoked to perform cloud application image distribution, application deployment, configuration and start-stop.

S12, loading file data.

S13, file processing is carried out.

After the cloud application is started, file data of the files to be processed can be loaded to process the files. In addition, a user can browse and edit the file at the cloud end through the cloud application client, and can update the file processing process picture in real time through streaming data of the cloud application protocol. After the user finishes the file processing task, the user can directly close the cloud application, and corresponding cloud data caching and computing power resources are automatically released to finish resource recovery.

Fig. 5 is a schematic structural diagram of a cloud computing platform according to an embodiment of the present application. Referring to fig. 5, the cloud computing platform may include: the cloud application management and control device 51, the cloud storage system 52 and the computing power resource pool 53, wherein the computing power resource pool 53 comprises a plurality of computing nodes;

The cloud application management and control device 51 is configured to obtain file characteristic information of a target file corresponding to a target file identifier in response to a file processing request including the target file identifier sent by a user through a cloud application client; determining computing power resource demand information of the cloud application according to file characteristic information of the target file; sending a resource application request comprising resource demand information to a computing power resource pool, receiving resource application response information returned by the computing power resource pool, scheduling cloud application to a target computing node according to a node identifier of the target computing node in the resource application response information, starting the cloud application scheduled to the target computing node, loading file data of a target file by the started cloud application from a cloud storage system 52, and carrying out file processing on the file data of the target file;

the computing power resource pool 53 is configured to respond to the resource application request, screen out target computing nodes that meet the computing power resource demand information from the computing nodes, and return resource application response information to the cloud application management and control device.

Further optionally, when the computing power resource pool 53 screens out target computing nodes that meet the computing power resource requirement information from all computing nodes of the cloud computing platform, the computing power resource pool is specifically configured to: if the file type of the target file is rich text type, the file characteristic information of the target file comprises at least one of the following: file size, picture size, image resolution, coding format, code rate, frame rate and key file characteristic information, wherein the computing power resource demand information indicates that a graphic class computing node with a GPU is taken as the target computing node for demand, and then the graphic class computing node with the GPU meeting the computing power resource demand information is screened out from all computing nodes of a cloud computing platform to be taken as the target computing node; if the file type of the target file is a plain text type, the file characteristic information of the target file comprises at least one of the following: and the computing power resource demand information indicates that the common computing node without the GPU is taken as the target computing node for demand, and the common computing node without the GPU meeting the computing power resource demand information is screened out from all computing nodes of the cloud computing platform to be taken as the target computing node.

Further optionally, when the cloud application management and control device 51 determines the computing power resource requirement information of the cloud application according to the file characteristic information of the target file, the method is specifically used for: determining the calculation complexity of the target file according to the file characteristic information of the target file; and determining computing power resource demand information of the cloud application according to the computing complexity of the target file.

Further optionally, when the cloud application management and control device 51 determines the computing power resource requirement information of the cloud application according to the computing complexity of the target file, the method is specifically used for: and determining computing power resource demand information of the cloud application according to the computing complexity of the target file and node information of each computing node of the cloud computing platform, wherein the node information comprises load pressure and/or residual resource information.

Further optionally, when the cloud application management and control device 51 determines the computational complexity of the target file according to the file characteristic information of the target file, the method is specifically used for: determining a target calculation complexity determination mode matched with the file type of the target file from calculation complexity determination modes corresponding to the file types according to the file types in the file characteristic information of the target file; and processing the file characteristic information of the target file by using a target calculation complexity determination mode so as to determine the calculation complexity of the target file.

Further optionally, when the cloud application management and control device 51 obtains the file characteristic information of the target file corresponding to the target file identifier, the method is specifically used for: searching file characteristic information of the target file from file characteristic information of each file stored by the cloud computing platform according to the target file identification;

if the file characteristic information of the target file is not found in the cloud computing platform, acquiring file metadata information of the target file from a cloud storage system; taking file metadata information of the target file as file characteristic information of the target file, and associating and storing target file identification and file characteristic information of the target file in a cloud computing platform.

Further optionally, before the cloud application management and control device 51 uses the file metadata information of the target file as the file characteristic information of the target file, the cloud application management and control device is further configured to: extracting features of file data of the target file cached in the cloud storage system to obtain key file feature information of the target file;

accordingly, when the cloud application management and control device 51 uses the file metadata information of the target file as the file characteristic information of the target file, the cloud application management and control device is specifically configured to: and taking the file metadata information and the key file characteristic information of the target file as file characteristic information of the target file.

Further optionally, the cloud application management and control device 51 performs feature extraction on file data of the target file cached in the cloud storage system, and is specifically configured to: selecting a target feature extraction algorithm with algorithm performance matched with file metadata information of a target file from a plurality of feature extraction algorithms; and extracting the characteristics of the file data of the target file cached in the cloud storage system by using a target characteristic extraction algorithm to obtain the key file characteristic information of the target file.

Further optionally, the cloud application management and control device 51 is further configured to, after the cloud application finishes processing the target file: and responding to a closing request of the cloud application sent by a user through the cloud application client, closing the cloud application, and releasing the computing resources occupied by the cloud application in the target computing node.

The specific manner in which the cloud application management and control device 51 performs the operations in the above embodiments has been described in detail in the embodiments of the related methods, and will not be described in detail herein.

It should be noted that, the execution subjects of each step of the method provided in the above embodiment may be the same device, or the method may also be executed by different devices. For example, the execution subject of steps 201 to 205 may be device a; for another example, the execution subject of steps 201 and 202 may be device a and the execution subject of steps 203 to 205 may be device B; etc.

In addition, in some of the flows described in the above embodiments and the drawings, a plurality of operations appearing in a specific order are included, but it should be clearly understood that the operations may be performed out of the order in which they appear herein or performed in parallel, the sequence numbers of the operations such as 201, 202, etc. are merely used to distinguish between the various operations, and the sequence numbers themselves do not represent any order of execution. In addition, the flows may include more or fewer operations, and the operations may be performed sequentially or in parallel. It should be noted that, the descriptions of "first" and "second" herein are used to distinguish different messages, devices, modules, etc., and do not represent a sequence, and are not limited to the "first" and the "second" being different types.

It should be noted that, the user information (including but not limited to user equipment information, user personal information, etc.) and the data (including but not limited to data for analysis, stored data, presented data, etc.) related to the present application are information and data authorized by the user or fully authorized by each party, and the collection, use and processing of the related data need to comply with the related laws and regulations and standards of the related country and region, and provide corresponding operation entries for the user to select authorization or rejection.

Fig. 6 is a schematic structural diagram of a computer device according to an embodiment of the present application. As shown in fig. 6, the computer device includes: a memory 61 and a processor 62;

memory 61 is used to store computer programs and may be configured to store various other data to support operations on the computing platform. Examples of such data include instructions for any application or method operating on a computing platform, contact data, phonebook data, messages, pictures, videos, and the like.

The Memory 61 may be implemented by any type or combination of volatile or non-volatile Memory devices, such as Static Random access Memory (Static Random-AccessMemory, SRAM), electrically erasable programmable Read-Only Memory (Electrically Erasable Programmable Read Only Memory, EEPROM), erasable programmable Read-Only Memory (Erasable Programmable Read Only Memory, EPROM), programmable Read-Only Memory (Programmable Read-Only Memory, PROM), read-Only Memory (ROM), magnetic Memory, flash Memory, magnetic or optical disk.

A processor 62 coupled to the memory 61 for executing a computer program in the memory 61 for executing steps in a file processing method of a cloud application or a computing power scheduling method of the cloud application.

Further, as shown in fig. 6, the computer apparatus further includes: communication component 63, display 64, power component 65, audio component 66, and other components. Only some of the components are schematically shown in fig. 6, which does not mean that the computer device only comprises the components shown in fig. 6. In addition, the components within the dashed box in FIG. 6 are optional components, and not necessarily optional components, depending on the product form of the computer device.

The detailed implementation process of each action performed by the processor may be referred to in the foregoing embodiments, and will not be described herein.

Accordingly, embodiments of the present application also provide a computer-readable storage medium storing a computer program, where the computer program is executed to implement the steps executable by a computer device in the above-described method embodiments.

Accordingly, embodiments of the present application also provide a computer program product comprising a computer program/instructions which, when executed by a processor, cause the processor to carry out the steps of the above-described method embodiments that are executable by a computer device.

The communication component is configured to facilitate wired or wireless communication between the device in which the communication component is located and other devices. The device where the communication component is located may access a wireless network based on a communication standard, such as a mobile communication network of WiFi (Wireless Fidelity ), 2G (2 generation,2 generation), 3G (3 generation ), 4G (4 generation,4 generation)/LTE (long Term Evolution ), 5G (5 generation,5 generation), or a combination thereof. In one exemplary embodiment, the communication component receives a broadcast signal or broadcast-related information from an external broadcast management system via a broadcast channel. In one exemplary embodiment, the communication component further includes a near field communication (Near Field Communication, NFC) module to facilitate short range communications. For example, the NFC module may be implemented based on radio frequency identification (Radio Frequency Identification, RFID) technology, infrared data association (The Infrared Data Association, irDA) technology, ultra Wideband (UWB) technology, bluetooth (BT) technology, and other technologies.

The display includes a screen, which may include a liquid crystal display (Liquid Crystal Display, LCD) and a Touch Panel (TP). If the screen includes a touch panel, the screen may be implemented as a touch screen to receive input signals from a user. The touch panel includes one or more touch sensors to sense touches, swipes, and gestures on the touch panel. The touch sensor may sense not only the boundary of a touch or sliding action, but also the duration and pressure associated with the touch or sliding operation.

The power supply component provides power for various components of equipment where the power supply component is located. The power components may include a power management system, one or more power sources, and other components associated with generating, managing, and distributing power for the devices in which the power components are located.

The audio component described above may be configured to output and/or input an audio signal. For example, the audio component includes a Microphone (MIC) configured to receive external audio signals when the device in which the audio component is located is in an operational mode, such as a call mode, a recording mode, and a voice recognition mode. The received audio signal may be further stored in a memory or transmitted via a communication component. In some embodiments, the audio assembly further comprises a speaker for outputting audio signals.

It will be appreciated by those skilled in the art that embodiments of the present application may be provided as a method, system, or computer program product. Accordingly, the present application may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the present application may take the form of a computer program product embodied on one or more computer-readable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, and the like) having computer-usable program code embodied therein.

The present application is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems) and computer program products according to embodiments of the application. It will be understood that each flow and/or block of the flowchart illustrations and/or block diagrams, and combinations of flows and/or blocks in the flowchart illustrations and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.

These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks.

These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.

In one typical configuration, a computing device includes one or more processors (Central Processing Unit, CPUs), input/output interfaces, network interfaces, and memory.

The Memory may include non-volatile Memory in a computer readable medium, random access Memory (Random Access Memory, RAM) and/or non-volatile Memory, etc., such as Read Only Memory (ROM) or flash RAM. Memory is an example of computer-readable media.

Computer readable media, including both non-transitory and non-transitory, removable and non-removable media, may implement information storage by any method or technology. The information may be computer readable instructions, data structures, modules of a program, or other data. Examples of storage media for a computer include, but are not limited to, phase Change RAM (PRAM), static Random-Access Memory (SRAM), dynamic Random-Access Memory (Dynamic Random Access Memory, DRAM), other types of Random-Access Memory (Random Access Memory, RAM), read Only Memory (ROM), electrically Erasable Programmable Read Only Memory (EEPROM), flash Memory or other Memory technology, compact disc Read Only Memory (CD-ROM), digital versatile disc (Digital versatile disc, DVD) or other optical storage, magnetic cassettes, magnetic tape storage or other magnetic storage devices, or any other non-transmission medium, operable to store information that may be accessed by the computing device. Computer-readable media, as defined herein, does not include transitory computer-readable media (transmission media), such as modulated data signals and carrier waves.

It should also be noted that the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising one … …" does not exclude the presence of other like elements in a process, method, article or apparatus that comprises an element.

The foregoing is merely exemplary of the present application and is not intended to limit the present application. Various modifications and variations of the present application will be apparent to those skilled in the art. Any modification, equivalent replacement, improvement, etc. which come within the spirit and principles of the application are to be included in the scope of the claims of the present application.

Claims

1. A method for processing a file of a cloud application, comprising:

responding to a file processing request which is sent by a user through a cloud application client and comprises a target file identifier, and acquiring file characteristic information of a target file corresponding to the target file identifier;

Determining computing power resource demand information of the cloud application according to the file characteristic information of the target file;

screening target computing nodes meeting the computing power resource demand information from all computing nodes of a cloud computing platform;

and scheduling the cloud application to the target computing node, and starting the cloud application scheduled to the target computing node to process the target file.

2. The method of claim 1, wherein determining computing power resource requirement information of the cloud application from file characteristic information of the target file comprises:

determining the calculation complexity of the target file according to the file characteristic information of the target file;

and determining computing power resource demand information of the cloud application according to the computing complexity of the target file.

3. The method of claim 2, wherein determining computing power resource requirement information of the cloud application according to the computational complexity of the target file comprises:

and determining computing power resource demand information of the cloud application according to the computing complexity of the target file and node information of each computing node of the cloud computing platform, wherein the node information comprises load pressure and/or residual resource information.

4. The method of claim 2, wherein determining the computational complexity of the target file based on file characteristic information of the target file comprises:

determining a target calculation complexity determination mode matched with the file type of the target file from calculation complexity determination modes corresponding to the file types according to the file types in the file characteristic information of the target file;

and processing the file characteristic information of the target file by using the target calculation complexity determining mode so as to determine the calculation complexity of the target file.

5. The method of claim 1, wherein obtaining file characteristic information of the target file corresponding to the target file identifier comprises:

searching file characteristic information of the target file from file characteristic information of each file stored by the cloud computing platform according to the target file identification;

if the file characteristic information of the target file is not found in the cloud computing platform, acquiring file metadata information of the target file from a cloud storage system;

and taking the file metadata information of the target file as the file characteristic information of the target file, and storing the target file identification and the file characteristic information of the target file in a correlated manner in the cloud computing platform.

6. The method of claim 5, further comprising, prior to taking file metadata information of the target file as file characteristic information of the target file:

extracting features of file data of the target file cached in the cloud storage system to obtain key file feature information of the target file;

accordingly, taking the file metadata information of the target file as the file characteristic information of the target file includes: and taking the file metadata information and the key file characteristic information of the target file as the file characteristic information of the target file.

7. The method of claim 6, wherein extracting features from the file data of the target file cached in the cloud storage system to obtain key file feature information of the target file, includes:

selecting a target feature extraction algorithm with algorithm performance matched with file metadata information of the target file from a plurality of feature extraction algorithms;

and extracting the characteristics of the file data of the target file cached in the cloud storage system by using the target characteristic extraction algorithm to obtain the key file characteristic information of the target file.

8. The method of any one of claims 1 to 7, further comprising, after the cloud application finishes processing the target file:

and responding to a closing request of the cloud application sent by the user through a cloud application client, closing the cloud application, and releasing computing resources occupied by the cloud application in the target computing node.

9. The method of any one of claims 1 to 7, wherein screening out target computing nodes from the computing nodes of the cloud computing platform that meet the computing power resource demand information comprises:

and sending a resource application request to a computing power resource pool in the cloud computing platform, so that the computing power resource pool screens out target computing nodes meeting the computing power resource demand information from all computing nodes in the computing power resource pool.

10. The method of any one of claims 1 to 7, wherein screening out target computing nodes from the computing nodes of the cloud computing platform that meet the computing power resource demand information comprises:

if the file type of the target file is rich text type, the file characteristic information of the target file comprises at least one of the following: file size, picture size, image resolution, coding format, code rate, frame rate and key file characteristic information, wherein the computing power resource demand information indicates that a graphic class computing node with a GPU is taken as the target computing node for demand, and then the graphic class computing node with the GPU meeting the computing power resource demand information is screened out from all computing nodes of a cloud computing platform to be taken as the target computing node;

If the file type of the target file is a plain text type, the file characteristic information of the target file comprises at least one of the following: and the computing power resource demand information indicates that the common computing node without the GPU is taken as the target computing node for demand, and the common computing node without the GPU meeting the computing power resource demand information is screened out from all computing nodes of the cloud computing platform to be taken as the target computing node.

11. The computing power scheduling method for the cloud application is characterized by comprising the following steps of:

acquiring a target file identifier sent by a user through a cloud application client, and acquiring file characteristic information of a target file corresponding to the target file identifier;

and scheduling the cloud application to the target computing node.

12. A cloud computing platform, comprising: the cloud application management and control device, the cloud storage system and the computing power resource pool comprise a plurality of computing nodes;

The cloud application management and control device is used for responding to a file processing request which is sent by a user through a cloud application client and comprises a target file identifier, and acquiring file characteristic information of a target file corresponding to the target file identifier; determining computing power resource demand information of the cloud application according to the file characteristic information of the target file; sending a resource application request comprising the resource demand information to the computing power resource pool, receiving resource application response information returned by the computing power resource pool, scheduling the cloud application to a target computing node according to a node identifier of the target computing node in the resource application response information, starting the cloud application scheduled to the target computing node, loading file data of the target file from the cloud storage system by the cloud application after starting, and carrying out file processing on the file data of the target file;

and the computing power resource pool is used for responding to the resource application request, screening target computing nodes meeting the computing power resource demand information from all computing nodes, and returning resource application response information to the cloud application management and control device.

13. A computer device, comprising: a memory and a processor; the memory is used for storing a computer program; the processor is coupled to the memory for executing the computer program for performing the steps in the method of any of claims 1-11.

14. A computer readable storage medium storing a computer program, which, when executed by a processor, causes the processor to carry out the steps of the method according to any one of claims 1-11.