An integrated management method, device and system of artificial intelligence applications

Abstract

The present disclosure provides an integrated management method, device and system of artificial intelligence applications, the method comprising: obtaining a predetermined standard format installation file of a first artificial intelligence application, installing the first artificial intelligence application based on the predetermined standard format installation file; obtaining construction configuration information of a second artificial intelligence application, constructing the second artificial intelligence application based on the construction configuration information; and managing the installed first artificial intelligence application and the constructed second artificial intelligence application.

Description

Technical Field

[0001] This disclosure relates to the field of artificial intelligence technology, and more specifically, to an integrated management method for artificial intelligence applications, an integrated management device for artificial intelligence applications, an integrated management system for artificial intelligence applications including at least one computing device and at least one storage device, and a computer-readable storage medium. Background Technology

[0002] With the development and popularization of Artificial Intelligence (AI) technology, an increasing number of industry scenarios require the application of AI. Currently, due to the diverse business problems faced by enterprises and other users, and the fact that most solutions on the market can only address specific domain issues, the vast majority of AI applications are highly customized for specific business problems. Constrained by the complexity of users' own business problems and issues of human efficiency, users often choose to purchase AI tools and software from multiple vendors to meet their business needs when a single vendor's customized AI application cannot solve all their problems. As user demands for software surge, the lack of data interoperability, inconsistent standards, and complex operation and maintenance among various AI applications lead to significant management costs for users, which is highly detrimental to the AI-driven intelligent transformation of enterprises and other users. Summary of the Invention

[0003] One objective of this disclosure is to provide a new technical solution for the integrated management of AI applications.

[0004] According to a first aspect of this disclosure, an integrated management method for AI applications is provided, comprising:

[0005] Obtain an installation file in a predetermined standard format for the first AI application, and install the first AI application based on the installation file in the predetermined standard format;

[0006] Obtain the construction configuration information of the second AI application, and construct the second AI application based on the construction configuration information;

[0007] The first AI application installed and the second AI application constructed are managed.

[0008] Optionally, obtaining the installation file of the first AI application in a predetermined standard format includes:

[0009] Obtain the original installation file of the target AI application and convert the format of the original installation file into the predetermined standard format.

[0010] Optionally, the predetermined standard format includes the Docker standard format.

[0011] Optionally, the construction configuration information includes storage configuration information and computation configuration information;

[0012] The construction of the second AI application based on the configuration information includes:

[0013] The second AI application is constructed based on the storage configuration information and the computing configuration information.

[0014] Optionally, the method further includes:

[0015] Provide an app store that offers at least one first AI application;

[0016] The acquisition of the installation file of the first AI application in a predetermined standard format includes:

[0017] Download the installation file of the first AI application in a predetermined standard format from the app store; or,

[0018] Download the original installation file of the first AI application from the app store, and convert the format of the original installation file to the predetermined standard format.

[0019] Optionally, the management of the installed first AI application and the constructed second AI application includes at least one of the following:

[0020] Acquire the data generated by the first AI application and the second AI application during operation, and store the generated data;

[0021] Collect the operation data of the first AI application and the second AI application to obtain the operation logs of the first AI application and the second AI application;

[0022] Monitor at least one of the currently running processes, operators, workflows, and running phases of the first AI application and the second AI application to obtain the resources occupied by at least one of the processes, operators, workflows, and running phases;

[0023] Monitor user actions on the first AI application and the second AI application, and obtain audit records of the first AI application and the second AI application;

[0024] The system monitors specified metrics of the first AI application and the second AI application during operation, and issues an alarm notification when the specified metrics are abnormal.

[0025] In response to the operation of running the first AI application and the second AI application, the user's permission to operate the first AI application and the second AI application is verified, and if the user has the permission to operate the first AI application and the second AI application, the first AI application and the second AI application are run.

[0026] The system accepts and stores user-uploaded data, and then provides the uploaded data to the first AI application and the second AI application for consumption.

[0027] Optionally, after storing the generated data, the method further includes: displaying the generated data in response to an operation of viewing the generated data;

[0028] or,

[0029] After obtaining the audit records of the first AI application and the second AI application, the method further includes: displaying the audit records in response to an operation of viewing the audit records;

[0030] or,

[0031] After obtaining the operation logs of the first AI application and the second AI application, the method further includes: displaying the operation logs in response to the operation of viewing the operation logs;

[0032] or,

[0033] After acquiring the resources occupied by at least one of the process, the operator, the workflow, and the running phase, the method further includes:

[0034] In response to an operation to view the occupied resources, the occupied resources are displayed.

[0035] Optionally, providing the uploaded data to the first AI application and the second AI application for consumption includes:

[0036] AI data related to the first AI application and the second AI application is obtained from the uploaded data, and the AI ​​data is provided to the first AI application and the second AI application for consumption;

[0037] or,

[0038] AI data related to the first AI application and the second AI application is obtained from the uploaded data. Data matching the application scenarios corresponding to the first AI application and the second AI application is filtered out from the AI ​​data and provided to the first AI application and the second AI application for consumption.

[0039] Optionally, the method further includes:

[0040] Determine the user's operating permissions for the integrated AI application, and provide the corresponding AI application based on the operating permissions.

[0041] Optionally, the method further includes:

[0042] Provide a management desktop, on which shortcuts to some or all of the integrated AI applications and / or management tools and / or file shortcuts are provided;

[0043] In response to an editing operation on the desktop, the shortcuts provided on the desktop are edited.

[0044] Optionally, a portion of the storage configuration information and the computing configuration information are obtained through user-configured feature capabilities, wherein the feature capabilities refer to entities that have encapsulated available resources for direct invocation of relevant capabilities;

[0045] The step of constructing the second AI application based on the storage configuration information and the computing configuration information includes: calling corresponding feature capabilities based on a portion of the information in the storage configuration information and the computing configuration information to construct the second AI application.

[0046] Optionally, the storage configuration information includes at least one of input source information, application metadata, log and monitoring metric aggregation information, and storage declaration information;

[0047] The computing configuration information includes at least one of the following: service startup method information, dynamic resource information, and computing capacity information.

[0048] According to a second aspect of this disclosure, an integrated management device for AI applications is provided, comprising:

[0049] The first installation module is used to obtain an installation file of a predetermined standard format for the first AI application, and to install the first AI application based on the installation file of the predetermined standard format.

[0050] The second construction module is used to obtain the construction configuration information of the second AI application and construct the second AI application based on the construction configuration information.

[0051] The application management module is used to manage the installed first AI application and the constructed second AI application.

[0052] Optionally, the first installation module is used for:

[0053] Obtain the original installation file of the target AI application and convert the format of the original installation file into the predetermined standard format.

[0054] Optionally, the predetermined standard format includes the Docker standard format.

[0055] Optionally, the construction configuration information includes storage configuration information and computation configuration information;

[0056] The second building module is used for:

[0057] The second AI application is constructed based on the storage configuration information and the computing configuration information.

[0058] Optionally, the integrated management device further includes:

[0059] The app store provides a module in which at least one first AI application is provided;

[0060] The first installation module is used for:

[0061] Download the installation file of the first AI application in a predetermined standard format from the app store; or

[0062] Download the original installation file of the first AI application from the app store, and convert the format of the original installation file to the predetermined standard format.

[0063] Optionally, the application management module is used to perform at least one of the following:

[0064] Acquire the data generated by the first AI application and the second AI application during operation, and store the generated data;

[0065] Collect the operation data of the first AI application and the second AI application to obtain the operation logs of the first AI application and the second AI application;

[0066] Monitor at least one of the currently running processes, operators, workflows, and running phases of the first AI application and the second AI application to obtain the resources occupied by at least one of the processes, operators, workflows, and running phases;

[0067] Monitor user actions on the first AI application and the second AI application, and obtain audit records of the first AI application and the second AI application;

[0068] The system monitors specified metrics of the first AI application and the second AI application during operation, and issues an alarm notification when the specified metrics are abnormal.

[0069] In response to the operation of running the first AI application and the second AI application, the user's permission to operate the first AI application and the second AI application is verified, and if the user has the permission to operate the first AI application and the second AI application, the first AI application and the second AI application are run.

[0070] The system accepts and stores user-uploaded data, and then provides the uploaded data to the first AI application and the second AI application for consumption.

[0071] Optionally, a display module is also included for:

[0072] After storing the generated data, the generated data is displayed in response to the operation of viewing the generated data;

[0073] or,

[0074] After obtaining the audit records of the first AI application and the second AI application, in response to the operation of viewing the audit records, the audit records are displayed;

[0075] or,

[0076] After obtaining the operation logs of the first AI application and the second AI application, the operation logs are displayed in response to the operation of viewing the operation logs;

[0077] or,

[0078] After obtaining the resources occupied by at least one of the process, the operator, the workflow, and the running phase, the occupied resources are displayed in response to the operation of viewing the occupied resources.

[0079] Optionally, the application management module is used for:

[0080] AI data related to the first AI application and the second AI application is obtained from the uploaded data, and the AI ​​data is provided to the first AI application and the second AI application for consumption;

[0081] or,

[0082] AI data related to the first AI application and the second AI application is obtained from the uploaded data. Data matching the application scenarios corresponding to the first AI application and the second AI application is filtered out from the AI ​​data and provided to the first AI application and the second AI application for consumption.

[0083] Optionally, the integrated management device further includes:

[0084] The permission management module is used to determine the user's operation permissions for the integrated AI applications, and provide the corresponding AI applications based on the operation permissions.

[0085] Optionally, the integrated management device further includes:

[0086] The desktop management module provides a management desktop, which offers shortcuts to some or all of the integrated AI applications and / or management tools and / or file shortcuts.

[0087] In response to the editing operation on the management desktop, the shortcuts provided on the desktop are edited.

[0088] Optionally, a portion of the storage configuration information and the computing configuration information are obtained through user-configured feature capabilities, wherein the feature capabilities refer to entities that have encapsulated available resources for direct invocation of relevant capabilities;

[0089] The second building module is used to: invoke corresponding feature capabilities based on a portion of the storage configuration information and the computing configuration information to build the second AI application.

[0090] Optionally, the storage configuration information includes at least one of input source information, application metadata, log and monitoring metric aggregation information, and storage declaration information;

[0091] The computing configuration information includes at least one of the following: service startup method information, dynamic resource information, and computing capacity information.

[0092] According to a third aspect of this disclosure, an artificial intelligence application integrated management system is provided, comprising at least one computing device and at least one storage device, wherein the at least one storage device is used to store instructions for controlling the at least one computing device to execute the method according to a first aspect of this disclosure.

[0093] According to a fourth aspect of this disclosure, a computer-readable storage medium is provided having a computer program stored thereon that, when executed by a processor, implements the method described in the first aspect of this disclosure.

[0094] Through the embodiments of this disclosure, the integrated management method based on AI applications can uniformly manage the installed first AI application and the constructed second AI application in the AI ​​operating system, thereby reducing the development and management costs of AI applications.

[0095] Other features and advantages of this disclosure will become clear from the following detailed description of exemplary embodiments with reference to the accompanying drawings. Attached Figure Description

[0096] The accompanying drawings, which are incorporated in and form a part of this specification, illustrate embodiments of the present disclosure and, together with their description, serve to explain the principles of the present disclosure.

[0097] Figure 1 This is a block diagram illustrating an example of a hardware configuration of an electronic device that can be used to implement embodiments of the present disclosure.

[0098] Figure 2 This is a flowchart illustrating an integrated management method for AI applications according to embodiments of the present disclosure;

[0099] Figure 3 This is a schematic diagram of the standard development process for the second AI application according to the embodiments of this description;

[0100] Figure 4 This is a block diagram of an integrated management device for AI applications according to embodiments of the present disclosure;

[0101] Figure 5 This is a block diagram of a system according to an embodiment of the present disclosure. Detailed Implementation

[0102] Various exemplary embodiments of the present disclosure will now be described in detail with reference to the accompanying drawings. It should be noted that, unless otherwise specifically stated, the relative arrangement, numerical expressions, and values ​​of the components and steps set forth in these embodiments do not limit the scope of the present disclosure.

[0103] The following description of at least one exemplary embodiment is merely illustrative and is in no way intended to limit this disclosure or its application or use.

[0104] Techniques, methods, and equipment known to those skilled in the art may not be discussed in detail, but where appropriate, such techniques, methods, and equipment should be considered part of the specification.

[0105] In all the examples shown and discussed herein, any specific values ​​should be interpreted as merely exemplary and not as limitations. Therefore, other examples of exemplary embodiments may have different values.

[0106] It should be noted that similar labels and letters in the following figures indicate similar items; therefore, once an item is defined in one figure, it does not need to be discussed further in subsequent figures.

[0107] Hereinafter, various embodiments and examples according to the present disclosure will be described with reference to the accompanying drawings.

[0108] <Hardware Configuration>

[0109] Figure 1This is a block diagram illustrating the hardware configuration of an electronic device 1000 that can implement embodiments of the present disclosure.

[0110] Electronic device 1000 can be a portable computer, desktop computer, mobile phone, tablet computer, etc. For example... Figure 1 As shown, the electronic device 1000 may include a processor 1100, a memory 1200, an interface device 1300, a communication device 1400, a display device 1500, an input device 1600, a speaker 1700, a microphone 1800, etc. The processor 1100 may be a central processing unit (CPU), a microprocessor (MCU), etc. The memory 1200 may include, for example, ROM (Read-Only Memory), RAM (Random Access Memory), or non-volatile memory such as a hard disk. The interface device 1300 may include, for example, a USB interface, a headphone jack, etc. The communication device 1400 may be capable of wired or wireless communication, specifically including Wi-Fi communication, Bluetooth communication, 2G / 3G / 4G / 5G communication, etc. The display device 1500 may be, for example, an LCD screen, a touch screen, etc. The input device 1600 may include, for example, a touch screen, a keyboard, motion input, etc. Users can input / output voice information through the speaker 1700 and the microphone 1800.

[0111] Figure 1 The electronic devices shown are merely illustrative and in no way intended to limit this disclosure, its application, or use. In embodiments applied to this disclosure, the memory 1200 of the electronic device 1000 is used to store instructions for controlling the processor 1100 to operate and execute any of the integrated management methods for AI applications provided in embodiments of this disclosure. Those skilled in the art will understand that, although... Figure 1 The electronic device 1000 is shown with multiple devices shown; however, this disclosure may relate only to some of these devices. For example, electronic device 1000 may only relate to processor 1100 and storage device 1200. Those skilled in the art can design instructions based on the schemes disclosed herein. How the instructions control the processor to operate is well known in the art and will not be described in detail here.

[0112] <Method Implementation>

[0113] This embodiment provides an integrated management method for AI applications. This AI application management method can be implemented by an electronic device. The electronic device can be, for example,... Figure 1 The electronic device 1000 shown is an example of a device that can be pre-deployed with an integrated management system for AI applications.

[0114] To address the current issues of data incompatibility, inconsistent standards, complex operation and maintenance, and the significant management costs associated with the development and management of various AI applications, this disclosure creatively proposes a design concept for the unified management of various AI applications by integrating them according to certain standards. This proposes an integrated management system for AI applications, also known as an AI operating system (AIOS). This system allows various apps to be integrated according to certain standards and provides unified management of the integrated apps. Specifically, AI applications can be installed on AIOS, and AI applications can also be built using AIOS. AIOS also provides various system and application management tools to manage and control the hardware and software resources required for the operation of AI applications.

[0115] In one example, the electronic device in this embodiment can be a computer cluster, and the AIOS can be deployed on that computer cluster.

[0116] In one embodiment of this disclosure, AIOS can provide a runtime environment for AI applications, which specifically includes the services and interfaces necessary for the management and development of AI applications.

[0117] In this embodiment, AIOS can also be used to manage and control the hardware and software resources required for the operation of AI applications, so that all resources required by AI applications can be used to the maximum extent.

[0118] Furthermore, AIOS can also provide a user interface that conforms to the operating system's understanding, making it easier for users to manage AI applications within AIOS.

[0119] according to Figure 2 As shown, the integrated management method for AI applications in this embodiment may include the following steps S2100 to S2300:

[0120] Step S2100: Obtain the installation file of the corresponding AI application in a predetermined standard format, and install the first AI application based on the installation file in the predetermined standard format.

[0121] In one embodiment of this disclosure, the first AI application may be installed in AIOS based on an installation file of a predetermined standard format. Therefore, the first AI application in this embodiment may be an AI application installed in AIOS according to the installation file.

[0122] To integrate one or more first AI applications on AIOS, a standard format for the installation files of the first AI applications must be defined in advance. For different first AI applications to be installed on AIOS, the installation files must meet the same standard format.

[0123] In this embodiment, the predetermined standard format can specifically be a file format pre-defined by AIOS, used to install the corresponding first AI application using an installation file based on the predetermined standard format. For example, the predetermined standard format can be the Docker standard format.

[0124] In one embodiment of this disclosure, obtaining the installation file of the first AI application in a predetermined standard format includes the following steps S2110 to S2120:

[0125] Step S2110: Obtain the original installation file of the first AI application.

[0126] In this embodiment, the original installation file can be an application installation file provided by a third-party service, such as an installation file provided by the developer of the first AI application. This original installation file can be downloaded from the application store provided by AIOS in this embodiment, downloaded from a designated download address, obtained from other electronic devices that have pre-stored the original installation file, or obtained from the local disk of the electronic device executing this embodiment. This disclosure does not limit the source of the original installation file.

[0127] Step S2120: Convert the original installation file format to a predetermined standard format to obtain an installation file in the predetermined standard format.

[0128] In this embodiment, after converting the original installation file format into a predetermined standard format to obtain an installation file in the predetermined standard format, the corresponding first AI application can be installed based on the installation file in the predetermined standard format.

[0129] In one embodiment of this disclosure, the method may further include: providing an app store that provides at least one first AI application.

[0130] In one example, the app store may provide an installation file in a pre-defined standard format for the first AI application. Therefore, obtaining the original installation file for the first AI application could involve downloading the pre-defined standard format installation file from the app store.

[0131] In another example, the original installation file of the first AI application can be provided in the app store. Obtaining the original installation file of the first AI application could then include: downloading the original installation file of the first AI application from the app store, and converting the format of the original installation file to a predetermined standard format to obtain an installation file in the predetermined standard format.

[0132] In this embodiment, by installing a first AI application in AIOS, the installed first AI application can run on AIOS. Furthermore, AIOS can also manage the installed first AI application.

[0133] Step S2200: Obtain the construction configuration information of the second AI application, and build the second AI application based on the construction configuration information.

[0134] AIOS can not only install the aforementioned first AI applications, such as applications developed by third-party vendors or customized applications, but also provide the ability to build second AI applications. Users can build second AI applications that match their business needs by utilizing the services and tools provided by AIOS.

[0135] In one embodiment of this disclosure, the configuration information may include storage configuration information and computation configuration information. Therefore, building a second AI application based on this configuration information may include: building a second AI application based on the storage configuration information and the computation configuration information.

[0136] AIOS can provide a standard process for building AIOS. In one example, it can receive storage and computing configuration information related to the application to be built, either by user input or automatically obtained, based on a predetermined configuration process. AIOS then builds and manages the AI ​​application based on this configuration information. The predetermined configuration process can be a standard application building process provided by AIOS. Users (e.g., application developers) only need to configure (input) the information required to build the second AI application according to this standard application building process. For example, a standard application building process interface and instructions can be provided to users, who only need to input the information required to build the second AI application according to the interface and instructions. According to an exemplary embodiment of this disclosure, the configuration information required to build the second AI application can be classified into storage configuration information and computing configuration information according to the purpose of the information.

[0137] According to exemplary embodiments of this disclosure, storage configuration information may include, but is not limited to, at least one of input source information, application metadata, log and monitoring metric aggregation information, and storage declaration information. The storage configuration information according to exemplary embodiments of this disclosure is described in detail below.

[0138] According to exemplary embodiments of this disclosure, input source information may include identifiers of one or more standard Docker images. Here, Docker can be selected as a general input convention. Docker has the broadest developer base, is currently the most commonly used containerization technology, and aligns with the evolutionary design of advanced enterprise architectures. Furthermore, Docker offers the following advantages: faster startup speed than virtual machines, a consistent runtime environment, easier migration, performance closer to native applications than virtual machines, easier reuse, and a large number of high-quality official images in the Docker ecosystem, significantly reducing image creation costs. In terms of universality, Docker is compatible with major domestic and international cloud vendors, including AWS, Google Cloud, Microsoft Azure, Alibaba Cloud, Tencent Cloud, and Huawei Cloud, meaning any cloud user can contribute Docker applications. Docker is compatible with commonly used enterprise resource management and scheduling systems, such as Kubernetes (K8s) and Mesos, meaning almost any enterprise customer can contribute Docker applications. Docker is compatible with commonly used operating systems in the industry; it can run on Linux, Mac, and Windows, meaning developers of any operating system can contribute Docker applications.

[0139] According to an exemplary embodiment of this disclosure, application metadata is information required for the unified operation and management of a second AI application. Application metadata may include, but is not limited to, at least one of the following: unique application identifier, authentication information, unique application name, unique display name, application icon, documentation, creation time, update time, creator information, version number, application source type, startup resource consumption information, data source, application storage address, mirror address, and startup method. Specifically, the unique application identifier can be generated by AIOS, and the user (in this document, the user can be the application developer) is unaware of it. Authentication information can be provided by the user through a system key, and its correctness is verified by AIOS. The unique application name can be provided by the developer, and its uniqueness is verified by AIOS. The unique display name can be provided by the developer, and its uniqueness is verified by AIOS. The application icon can be provided by the developer, and its uniqueness is verified by AIOS. The documentation can be provided by the developer, and its completeness and internationalization standards are verified by AIOS; the secondary level must include an introduction, help documentation, etc. The creation time can be generated by AIOS, and the user is unaware of it. The update time can be generated by AIOS, and the user is unaware of it. Creator information can be provided by the user and verified by AIOS. Version number can also be provided by the user. Application source type can be generated by the system, which is transparent to the user and indicates whether the application is an official pre-installed application or a third-party application. Startup resource consumption information can be provided by the user, indicating the number of instances, whether automatic scaling is enabled, CPU, GPU, MEM, FPGA, etc. Data source can be an officially pre-installed data format, a standard data source provided by the underlying storage system, or another Docker image to interface with a third-party data source. Application storage address refers to the data storage space where the second AI application is running. Image address can be provided by the developer, indicating the startup path of the second AI application. Startup method can be provided by the developer, indicating whether the same second AI application can be repeatedly opened by different application users during operation. Of course, application metadata is not limited to the above information and may also include information required by AIOS to manage the second AI application.

[0140] According to exemplary embodiments of this disclosure, log and monitoring metric aggregation information can be input by the user to indicate how to obtain logs and monitoring metrics. For example, if the user obtains logs using a standard error stream, it can be done without the user's awareness. If the user obtains monitoring metrics through AIOS, it can also be done without the user's awareness. If the user needs to customize monitoring metrics, they can use the software development kit (SDK) provided by AIOS for customized development.

[0141] According to exemplary embodiments of this disclosure, storage declaration information can be entered by a user to declare the storage system used, and a corresponding time-to-live (TTL) policy can be configured. For example, the storage system may include, but is not limited to, file storage, database storage, etc.

[0142] According to exemplary embodiments of this disclosure, computing configuration information may include, but is not limited to, at least one of service startup method information, dynamic resource information, and computing capacity information. The computing configuration information according to exemplary embodiments of this disclosure is described in detail below.

[0143] According to an exemplary embodiment of this disclosure, service launch method information can be input by the user to declare the application launch method. The user does not need to be aware of the second AI application launch process; they only need to configure the service launch method information, and the system can automatically initiate application launch to the underlying resource management and high-level platform.

[0144] According to exemplary embodiments of this disclosure, dynamic resource information can be entered by the user to declare the application's requested dynamic resources. For example, dynamic resources may include two parts: the first part may be resources that the user-developed dynamic service or task can control and that can be exposed to the application user for modification; the second part may be resources that the user-developed dynamic service or task can additionally request during operation to start new services or tasks.

[0145] According to exemplary embodiments of this disclosure, computing power information can be configured by the user to design application computing capabilities. For example, the user can utilize the SDK provided by AIOS to call a wide range of algorithms (e.g., AI algorithms), that is, to be able to call one or more algorithms using the SDK. Here, AI algorithms may include, but are not limited to, binary classification, multi-class classification, regression, clustering, time series analysis, image classification, image detection, optical character recognition (OCR), speech recognition, natural language processing (NLP), transfer learning, reinforcement learning, federated learning, automated machine learning, parameter tuning, etc. Furthermore, if the computing power required by the user is not within the computing power provided by AIOS, custom development can be performed through AIOS, and the user can enjoy the ability to use it in combination with the built-in computing power.

[0146] According to exemplary embodiments of this disclosure, the aforementioned storage configuration information and computing configuration information can be configured according to the standard development process provided by AIOS. For example, the standard development process according to exemplary embodiments of this disclosure can be as follows: Figure 3 As shown, Figure 3This is a schematic diagram of a standard development process according to an exemplary embodiment of the present disclosure. Specifically, the user can configure information in the following order: in step S201, receiving input source information; in step S202, receiving application metadata; in step S203, receiving service startup method information; in step S204, receiving dynamic resource information; in step S205, receiving log and monitoring indicator aggregation information; in step S206, receiving storage declaration information; and in step S207, receiving computing power information. Of course, the standard development process is not limited to the above order and may include any other feasible order. For example, steps S201 and S202 are the first and second steps, and steps S203 to S207 can be executed in any order. Alternatively, steps S201 to S207 can be executed using a user-defined process; that is, steps S201 to S207 can be executed in any possible order according to user needs.

[0147] According to an exemplary embodiment of this disclosure, a portion of the aforementioned storage configuration information and computing configuration information can be obtained by the user configuring the traits provided by the AIOS. Here, a trait can refer to an entity in the AIOS that has encapsulated available resources for direct invocation of relevant capabilities. For example, the AIOS may include a trait library, which includes at least one trait. The user can configure the traits in the trait library provided by the AIOS as needed to obtain the corresponding information, and the AIOS can invoke the corresponding trait based on the information to build a second AI application.

[0148] For example, when the input source information includes identifiers of multiple standard Docker images, the input source can be an orchestration of multiple standard Docker images. The orchestration of multiple standard Docker images can be obtained through the image orchestration features provided by AIOS, configured by the user. Here, image orchestration can include hybrid online / offline orchestration when multiple images include both online and offline images, online orchestration when multiple images are all online images, and offline orchestration when multiple images are all offline images.

[0149] For example, information about service startup methods can be obtained by configuring the service startup method features provided by AIOS.

[0150] For example, storage claim information can be obtained by configuring the storage features provided by AIOS. Of course, in addition to using storage feature capabilities to obtain storage claim information, the storage solution can also be replaced with open source, enterprise-grade storage solutions, such as Amazon's S3 storage or storage integrated with Microsoft Cloud.

[0151] For example, parameter tuning information can be obtained by configuring the storage features provided by AIOS.

[0152] According to an exemplary embodiment of this disclosure, when a portion of the storage configuration information and the computing configuration information is obtained through user-configured feature capabilities, the corresponding feature capabilities can be invoked based on the portion of information to construct a second AI application.

[0153] Step S2300: Manage the installed first AI application and the constructed second AI application.

[0154] In one embodiment of this disclosure, the AIOS may provide management tools for managing AI applications. These management tools may include, for example, a resource management module, a license verification module, a permission management module, an auditing module, a notification module, a log module, a file management module, a data platform, an application management module, and a process management module.

[0155] The resource management module can be used to monitor the computing and / or storage resources consumed by AI applications during operation, as well as the server and related hardware resources required by these applications. Users can configure and manage the resources consumed by AI applications within AIOS through this module. Monitoring and managing resources through this module ensures the stable operation of AIOS and AI applications, guaranteeing business continuity.

[0156] The license verification module can manage software licenses for the use of AIOS and / or AI applications. Users can use the license verification module to flexibly restrict the usage time, application scope, and functional modules of AIOS and / or AI applications, so as to effectively protect the intellectual property rights of the developers of AIOS and / or AI applications.

[0157] The permissions management module is used to manage users' operation permissions and capabilities on AIOS. Specifically, it can manage users of AIOS, the tools they can use within AIOS, and the management area for user resources.

[0158] The audit module provides users with auditing functions for AIOS, recording key user operations to ensure the stability and traceability of AIOS. The audit module can record each user's operations and behaviors in each AI application, providing backend functions for viewing, and is supplemented by a comprehensive monitoring panel to analyze the overall operation of AIOS.

[0159] The notification module can be used to set and manage the content and methods of notifications for AIOS and various AI applications. In one example, the notification module can be used for alerts. The alert management provides users with personalized alert capabilities for monitoring items. Users can flexibly formulate alert rules for specified monitoring items and set custom alert strategies for serious AIOS issues. When a business metric reaches a critical value or an anomaly occurs in AIOS, the notification module can send alerts via SMS, email, or other methods as expected by the user, prompting the user to be aware of and handle the anomaly in a timely manner, thereby improving the availability of AIOS and business applications.

[0160] The log module is a tool for the unified collection, management, filtering, and retrieval of various operational data for AI applications. Users can use logs for operational monitoring and analysis, call chain monitoring and tracing to meet various digital operation and maintenance scenarios such as business analysis.

[0161] In AIOS, there is typically a file system for users to store, query, and manage files throughout the AIOS ecosystem. The file management module supports user uploads and other AI application outputs that users wish to view and consume. Basic data types that can be managed include tables, images, audio, text, video, and various user-defined data formats. Users can upload pre-defined data types, such as data tables and data groups, from other data sources to AIOS. After uploading, users can preview, view meta information, and delete data within the file management module. Data stored in the file management module can be consumed by AI applications for tasks such as model training. Simultaneously, AI applications may generate a large amount of logs, reports, and various short-term intermediate data. This data can be viewed through the backend view of the file management module, allowing users to examine the data generated throughout AIOS and stored in media such as HDFS and object storage engines, and perform data cleanup tasks.

[0162] A data platform can be used to manage and store the relationships between multiple basic data sets, representing an abstract data form of governance logic. In existing technologies, building AI applications requires defining goals based on business needs, collecting data, performing feature engineering, and training a model before a possible result can be obtained. This process heavily relies on the quality of the data itself, the skill level of the modelers, and the completeness of the application environment. While this method can maximize the modeling effect, its inefficient application generation capability is also one of the barriers hindering the widespread adoption of current AI technology. Therefore, embodiments of this disclosure provide a data platform within AIOS, through which AI data related to AI applications can be obtained. The data format of AI data can be viewed as a file format, similar to traditional DOC files.

[0163] A data platform can provide conventions for how AI data is used, allowing it to be accessed by AI applications with specified data formats and integrated with their functions to solve specific problems. The data platform can include methodologies for data usage in specific business domains, such as data acquisition methods, data governance methods, and data relationships. It can also display the basic file meta-formats acquired by AIOS.

[0164] The application management module is a tool for managing all AI applications that different users have access to. Meta-information (such as application logos) can be displayed in the application management module. For application developers, the application management module provides a complete registration process for standard AI applications, including resource application, start / stop control, storage space application, permission system integration, and monitoring integration—the entire application access process.

[0165] The process management module can be used to manage and monitor the running processes and resources of AI applications. During the operation of an AI application, active processes are generated. For AI applications, process management involves managing the running program entities, which may include managing all system resources used by the running AI application. The resources used and scheduled by AI applications are typically very complex, involving scheduling between multiple computing units. The process management module can break down the operation of AI applications according to the business organization dimension, using concepts such as "process," "workflow," "operator," and "running phase." Monitoring shifts from a layered and machine-centric perspective to a service-centric perspective, understanding the dependencies between services within the monitored scope, which simplifies the complexity of managing and monitoring the operation of AI applications.

[0166] While providing convenient services for AI applications, AIOS also offers development tools for AI applications, such as Studio and HyperCycle, to meet the mainstream development methods and experiences in the market.

[0167] Studio development tools provide a visual AI application development module, which includes four main parts: a modeling IDE (Integrated Development Environment) module, an application IDE module, model management, and application management. This allows advanced developers to flexibly and independently organize and connect data, algorithms, evaluation, and debugging, and helps modeling users clearly and intuitively define the entire process of various types of model development, model management and evaluation, and application building, orchestration, and debugging.

[0168] A modeling IDE provides tools for researching and developing machine learning models. Users can process data and generate models within this tool. The modeling IDE offers two interactive modeling environments: DAG (Directed Acyclic Graph) and Notebook. DAG is a user-friendly visual modeling tool where users can visualize and manipulate data on a DAG canvas, connecting various data processing and machine learning operators to define the entire machine learning workflow. Notebook is a professional command-line modeling tool where users can quickly validate ideas and obtain feedback by editing code blocks one by one.

[0169] All models generated in the Studio development tools are centrally managed by the model management system, which also supports importing external models for unified management. Furthermore, with the model at its core, the system compiles model-related evaluation and interpretation reports to help users understand the model from multiple perspectives.

[0170] The application IDE provides tools for building, orchestrating, debugging, and packaging AI applications, offering users complete, reliable, and highly flexible enterprise-level AI application development capabilities. Within the application IDE, users can create applications in a few simple steps, such as selecting a model / DAG, configuring resources and running strategies, and packaging the application, according to their actual business needs. Once created, the application can be deployed to the application management system.

[0171] After users develop AI applications that meet their business needs, they deploy them through application management. Application management enables comprehensive control over AI applications, including operation and maintenance monitoring, debugging and experimentation, dynamic resource configuration, version management, and more, meeting all user requirements for AI applications in online production environments.

[0172] HyperCycle is a low-barrier AI application building tool (including HC scenario definition, machine learning HyperCycle, image classification HyperCycle, etc., tools for various business problems). It helps enterprises solve AI business problems at scale and achieve rapid application and implementation of AI scenarios by dividing data, business, and IT tasks and supporting automatic model strategies.

[0173] Specifically, data personnel can collect and organize the data streams and relationships to be accessed according to business requirements, defining a HyperCycle scenario specific to a particular business. Learning tasks and frequencies can then be configured using the HyperCycle development tool. The HyperCycle development tool automatically matches appropriate automatic learning strategies based on the algorithm type and data type, updating online business data according to the set update frequency to prevent data decay. Users can monitor the model's performance through performance metrics on the product front end. Operations personnel can assess the health of the business and data based on activity performance and indices in the process manager.

[0174] In one embodiment of this disclosure, AIOS can also provide users with an extremely simple operating experience. The usage logic is consistent with the user's expectations of AIOS. Furthermore, it can leverage the user's experience with other operating systems to increase the understandability and operability of AIOS, allowing users to better focus their attention on their business goals, lowering the threshold for understanding AIOS, and improving work efficiency.

[0175] Based on the above description, in one embodiment of this disclosure, managing the installed first AI application and the constructed second AI application includes at least one of the following:

[0176] Acquire and store the data generated by the target AI application during its operation;

[0177] Collect the operational data of the target AI application to obtain the operational log of the target AI application;

[0178] Monitor at least one of the currently running processes, operators, workflows, and running phases of the target AI application to obtain the resources consumed by at least one of the processes, operators, workflows, and running phases.

[0179] Monitor user actions on the target AI application and obtain audit records of the first and second AI applications;

[0180] Monitor specified metrics of the target AI application during its operation, and issue alarm notifications when the specified metrics are abnormal;

[0181] In response to the operation of running the target AI application, verify the user's permission to operate the target AI application, and run the target AI application if the user has the permission to operate the target AI application;

[0182] It accepts and stores user-uploaded data, and then provides the uploaded data to the target AI application for consumption.

[0183] In this embodiment, the target AI application may include at least one first AI application and at least one second AI application.

[0184] In an embodiment that manages the installed first AI application and the constructed second AI application, including acquiring and storing the data generated by the target AI application during its operation, the target AI application may consume data uploaded by the user to AIOS to obtain the generated data. Specifically, this may include performing tasks such as model training, prediction, classification, recommendation, and natural language processing. The data types of the user-uploaded data may include tables, images, audio, text, video, and various user-defined data formats.

[0185] In this embodiment, the data uploaded by the user to AIOS and the data generated by the target AI application during operation can be stored, viewed and managed through the file management module.

[0186] Based on this embodiment, the method may further include: in response to an operation of viewing the output data, displaying the output data for the user to view.

[0187] Specifically, users can open the file management module to view and manage the data generated by the target AI application during its operation.

[0188] In an embodiment where managing the installed first AI application and the constructed second AI application includes collecting the runtime data of the target AI application and obtaining the runtime log of the target AI application, the runtime data of the target AI application can be collected through the log module in AIOS, and the runtime log of the target AI application can be generated.

[0189] After obtaining the runtime logs of the target AI application, the method may further include: displaying the runtime logs for the user to view in response to the operation of viewing the runtime logs.

[0190] In this embodiment, the user can trigger the operation of viewing the running log by opening the log module. In response to the operation of viewing the running log, AIOS displays the running log for the user to view and analyze.

[0191] In an embodiment where managing the installed first AI application and the constructed second AI application includes monitoring at least one of the currently running processes, operators, workflows, and running phases of the target AI application to obtain the resources occupied by at least one of the currently running processes, operators, workflows, and running phases of the target AI application, the resources occupied by at least one of the currently running processes, operators, workflows, and running phases of the target AI application may include computing resources, storage resources, and / or hardware resources.

[0192] In this embodiment, a process is a running activity of a program on a certain data set in a computer. It is the basic unit for system resource allocation and scheduling and the foundation of the operating system structure. A program is a description of instructions, data, and their organization, while a process is the entity of a program. By monitoring and managing the current processes that AI applications are active in during operation, the difficulty of monitoring and managing AI applications during operation can be simplified.

[0193] In one example, computing resources may include a Hadoop cluster, storage resources may include MySQL, and hardware resources may include at least one of CPU, GPU, and memory.

[0194] By monitoring the computing and / or storage resources consumed by AI applications during operation, users can easily manage the current activities of AI applications during operation, thereby ensuring the stable operation of AIOS and AI applications.

[0195] After obtaining the resources used by at least one of the processes, operators, workflows, and runtime phases, the method further includes: in response to an operation to view the used resources, displaying the used resources for the user to view. Thus, when the resources used by at least one of the processes, operators, workflows, and runtime phases are excessive, the user can stop running the target AI application according to their needs, or adjust the target AI application to improve its usability.

[0196] In this embodiment, by monitoring at least one of the currently running processes, operators, workflows, and running phases of the target AI application, the dependencies between the various services involved in the operation of the target AI application can be understood, which can simplify the process management module's difficulty in managing and monitoring the operation of the AI ​​application.

[0197] In an embodiment where managing the installed first AI application and the constructed second AI application includes monitoring user actions on the target AI application and obtaining audit records of the first and second AI applications, the user actions on the target AI application can be obtained through the audit module in AIOS, and the audit records can be obtained.

[0198] After obtaining the audit logs of the target AI application, the method may further include: displaying the audit logs for the user to view in response to an operation to view the audit logs.

[0199] Users can trigger the action of viewing audit logs by opening the audit module. AIOS responds to this action by displaying the audit logs for users to view and analyze. This ensures the traceability of the target AI application.

[0200] In an embodiment where managing the installed first AI application and the constructed second AI application includes monitoring specified metrics of the target AI application during operation and issuing alarm notifications when specified metrics are abnormal, the effective range of the specified metric values ​​for the target AI application during operation can be set through the notification module in AIOS. This allows monitoring of the specified metrics during operation, and if the actual value of the specified metric exceeds the preset effective range, the specified metric is determined to be abnormal, and an alarm notification is issued according to the notification method set by the user in the notification module. Users can view historical alarm notifications by opening the notification module.

[0201] In managing the installed first AI application and the constructed second AI application, including responding to an operation to run the target AI application and verifying the user's permission to operate the target AI application, the implementation of running the target AI application if the user has permission can be achieved through a permission verification module. This module verifies whether the current time of receiving the operation to run the target AI application in the AIOS is within the licensed usage period of the target AI application, thus determining whether the user has permission to operate the target AI application. If the user has permission to operate the target AI application, the target AI application runs in the AIOS; if the user does not have permission to operate the target AI application, running the target AI application in the AIOS can be prohibited. This provides effective intellectual property protection for the developers of the target AI application.

[0202] In an embodiment where managing the installed first AI application and the constructed second AI application includes accepting and storing user-uploaded data and providing the uploaded data to the target AI application for consumption, a file management module can be used to store, view, and manage the data uploaded by the user to AIOS. The data types of user-uploaded data can include tables, images, voice, text, video, and various user-defined data formats.

[0203] The target AI application can consume data uploaded by users, which can include performing tasks such as model training, prediction, classification, recommendation, and natural language processing.

[0204] In one embodiment of this disclosure, providing AI data to a target AI application for consumption further includes:

[0205] Extract AI data related to the target AI application from the uploaded data, and provide the relevant AI data to the target AI application for consumption.

[0206] In this embodiment, users can pre-set the data format of AI data related to the target AI application through the data platform, so that the data platform can filter the data stored in the file management module according to the data format and provide only the AI ​​data related to the target AI application.

[0207] Based on this embodiment, the method may further include: displaying AI data in response to an operation of viewing AI data.

[0208] Specifically, users can open the data platform and trigger the action of viewing the data platform. In response to this action, AIOS displays AI data related to AI applications for users to view and select.

[0209] In another embodiment of this disclosure, providing AI data to a target AI application for consumption further includes:

[0210] Extract AI data related to the target AI application from the uploaded data, filter out data that matches the application scenario corresponding to the target AI application from the AI ​​data, and provide it to the target AI for consumption.

[0211] Specifically, users can specify the application scenarios corresponding to the target AI application in advance according to their actual needs.

[0212] The data platform in AIOS can filter AI data to identify those relevant to the specific application scenario. This automatically provides the target AI application with AI data tailored to its application scenario, eliminating the need for manual filtering by the user, reducing user interaction, and improving the user experience.

[0213] In one embodiment of this disclosure, the method may further include:

[0214] Determine the user's operating permissions for the integrated AI applications, and provide the corresponding AI applications based on the operating permissions.

[0215] The AI ​​applications mentioned in the embodiments of this disclosure may include at least one first AI application and / or at least one second AI application.

[0216] Specifically, the AI ​​applications provided by AIOS can differ for users with different operating permissions. In this embodiment, users can only use the AI ​​applications provided according to their corresponding operating permissions.

[0217] For example, considering a first user and a second user with different operating permissions, the AI ​​applications provided to the first user could include AI application A and AI application B, while the AI ​​applications provided to the second user could include AI application B and AI application C. Therefore, the first user can use AI application A and AI application B, but cannot use AI application C; the second user can use AI application B and AI application C, but cannot use AI application A.

[0218] In addition, the method may also include: providing corresponding computing resources and / or storage resources in AIOS according to the operation permissions.

[0219] In one embodiment of this disclosure, the method may further include:

[0220] Provides a management desktop, which offers shortcuts to some or all of the integrated AI applications and / or management tools and / or file shortcuts; and allows editing of the shortcuts provided on the desktop in response to editing operations on the desktop.

[0221] The editing operations in this embodiment can include adding, moving, deleting, and organizing. Users can add shortcuts to AI applications and / or management tools and / or files on the desktop according to their needs, or delete shortcuts to AI applications and / or management tools and / or files displayed on the desktop, or adjust the display position of these shortcuts on the desktop, or change the folder on the desktop used to organize these shortcuts, thus enabling users to manage their desktop more efficiently and conveniently.

[0222] In this embodiment, users can add any accessible shortcuts to the desktop according to their own preferences and work habits, helping them to open the corresponding management tools, AI applications or files more conveniently.

[0223] The integrated management method in this embodiment can be used to integrate and manage AI applications deployed in AIOS, thereby reducing the development and management costs of AI applications.

[0224] <Example of Integrated Management Device>

[0225] In this embodiment, an integrated management device 3000 for AI applications is provided, such as... Figure 4As shown, the integrated management device 3000 may include a first installation module 3100, a second construction module 3200, and an application management module 3300. The first installation module 3100 is used to obtain an installation file of a first AI application in a predetermined standard format, and install the first AI application based on the installation file. The second construction module 3200 is used to obtain construction configuration information for a second AI application, and build the second AI application based on the construction configuration information. The application management module 3300 is used to manage the installed first AI application and the built second AI application.

[0226] In one embodiment of this disclosure, the first installation module 3100 may specifically be used for:

[0227] Obtain the original installation file of the target AI application and convert the original installation file format to a predetermined standard format.

[0228] In one embodiment of this disclosure, the predetermined standard format includes the Docker standard format.

[0229] In one embodiment of this disclosure, constructing configuration information includes storing configuration information and calculating configuration information;

[0230] The second building block 3200 can be used for:

[0231] A second AI application is built based on storage configuration information and computing configuration information.

[0232] In one embodiment of this disclosure, the integrated management device 3000 may further include:

[0233] The app store provides a module for providing an app store, which offers at least one first AI application;

[0234] The first installation module 3100 can also be used for:

[0235] Download the pre-defined standard format installation file for the first AI application from the app store; or

[0236] Download the original installation file of the First AI app from the app store and convert the original installation file to a predefined standard format.

[0237] In one embodiment of this disclosure, the application management module 3300 is configured to perform at least one of the following:

[0238] Acquire and store the data generated by the first and second AI applications during their operation.

[0239] Collect the operational data of the first AI application and the second AI application to obtain the operational logs of the first AI application and the second AI application;

[0240] Monitor at least one of the currently running processes, operators, workflows, and running phases of the first and second AI applications to obtain the resources consumed by at least one of the processes, operators, workflows, and running phases.

[0241] Monitor user actions on the first AI application and the second AI application, and obtain audit records of the first AI application and the second AI application;

[0242] Monitor specified metrics of the first and second AI applications during operation, and issue alarm notifications when specified metrics are abnormal.

[0243] In response to the operation of running the first AI application and the second AI application, the user's permission to operate the first AI application and the second AI application is verified, and the first AI application and the second AI application are run if the user has the permission to operate the first AI application and the second AI application.

[0244] It accepts and stores user-uploaded data, and then provides the uploaded data to the first AI application and the second AI application for consumption.

[0245] In one embodiment of this disclosure, the integrated management device 3000 may further include a display module for:

[0246] After storing the output data, the system displays the output data in response to operations that require viewing it.

[0247] or,

[0248] After obtaining the audit records of the first AI application and the second AI application, the audit records are displayed in response to the operation of viewing the audit records;

[0249] or,

[0250] After obtaining the running logs of the first AI application and the second AI application, the system responds to the operation of viewing the running logs and displays the running logs.

[0251] or,

[0252] After obtaining the resources used by at least one of the processes, operators, workflows, and runtime phases, display the used resources in response to the operation of viewing the used resources.

[0253] In one embodiment of this disclosure, the application management module 3300 can be used to:

[0254] Extract AI data related to the first AI application and the second AI application from the uploaded data, and provide the AI ​​data to the first AI application and the second AI application for consumption;

[0255] or,

[0256] AI data related to the first AI application and the second AI application is obtained from the uploaded data. Data matching the application scenarios of the first AI application and the second AI application is filtered out from the AI ​​data and provided to the first AI application and the second AI application for consumption.

[0257] In one embodiment of this disclosure, the integrated management device 3000 may further include:

[0258] The permission management module is used to determine the user's operation permissions for the integrated AI applications, and provide the corresponding AI applications based on the operation permissions.

[0259] In one embodiment of this disclosure, the integrated management device 3000 may further include:

[0260] Desktop management module, used for:

[0261] Provides an administrative desktop, which offers shortcuts to some or all of the integrated AI applications and / or management tools and / or file shortcuts;

[0262] In response to editing operations on the management desktop, edit the shortcuts provided on the desktop.

[0263] In one embodiment of this disclosure, a portion of the storage configuration information and computing configuration information is obtained through user-configured feature capabilities, wherein feature capabilities refer to entities that have encapsulated available resources for direct invocation of related capabilities;

[0264] The second building module 3200 can also be used to: invoke corresponding feature capabilities based on a portion of the stored configuration information and the computed configuration information in order to build a second AI application.

[0265] In one embodiment of this disclosure, the storage configuration information may include at least one of input source information, application metadata, log and monitoring metric aggregation information, and storage declaration information;

[0266] The computing configuration information includes at least one of the following: service startup method information, dynamic resource information, and computing capacity information.

[0267] In one embodiment of this disclosure, constructing a second AI application based on storage configuration information and computing configuration information includes at least one of the following steps:

[0268] Based on the input source information, connect to the input source;

[0269] The second AI application is managed based on application metadata.

[0270] Based on the service startup method information, an application startup command is sent to the underlying resource management and scheduling platform;

[0271] Dynamic resource management is performed based on dynamic resource information.

[0272] Based on log and monitoring metric aggregation information, it provides log and monitoring information for the second AI application;

[0273] Based on storage declaration information, the storage of the second AI application is managed;

[0274] Based on computing power information, configure the computing power of the second AI application.

[0275] In one embodiment of this disclosure, the input source information includes identifiers of one or more standard Docker images;

[0276] When the input source information includes the identifiers of multiple standard Docker images, the input source is an orchestration of multiple standard Docker images, where the orchestration based on multiple standard Docker images is obtained through the user's ability to configure image orchestration features.

[0277] In one embodiment of this disclosure, the application metadata includes at least one of the following:

[0278] Unique application identifier, authentication information, unique application name, unique display name, application icon, documentation, creation time, update time, creator information, version number, application source type, startup resource consumption information, data source, application storage address, image address, and startup method.

[0279] In one embodiment of this disclosure, service startup method information is obtained through the user's ability to configure service startup method features.

[0280] In one embodiment of this disclosure, storage claim information is obtained through user-configured storage feature capabilities.

[0281] In one embodiment of this disclosure, computing power information is obtained by a user calling one or more algorithms using a software development kit, and / or by custom development through an operator platform;

[0282] Among them, the parameter tuning information in the computing power information is obtained through the user's configuration of parameter tuning features.

[0283] Those skilled in the art will understand that the integrated management device 3000 for AI applications can be implemented in various ways. For example, the integrated management device 3000 for AI applications can be implemented by configuring a processor with instructions. For example, instructions can be stored in a ROM, and when the device is started, the instructions can be read from the ROM into a programmable device to implement the integrated management device 3000 for AI applications. For example, the integrated management device 3000 for AI applications can be embedded in a dedicated device (e.g., an ASIC). The integrated management device 3000 for AI applications can be divided into independent units, or they can be combined together. The integrated management device 3000 for AI applications can be implemented in one of the various implementation methods described above, or in a combination of two or more of the various implementation methods described above.

[0284] In this embodiment, the AI ​​application integrated management device 3000 can have various implementation forms. For example, the AI ​​application integrated management device 3000 can be any functional module running in a software product or application that provides management services for AI applications, or peripheral embeddings, plug-ins, patches, etc. of these software products or applications, or it can be the software products or applications themselves.

[0285] <System Implementation>

[0286] In this embodiment, as Figure 5 As shown, an AI application integrated management system 4000, which may also be referred to as an AI operating system, is also provided, comprising at least one computing device 4100 and at least one storage device 4200. The at least one storage device 4200 is used to store executable instructions; these instructions are used to control the at least one computing device 4100 to execute an AI application integrated management method according to any embodiment of this disclosure.

[0287] In this embodiment, the system 4000 can be a mobile phone, tablet computer, handheld computer, desktop computer, laptop computer, workstation, game console, or a distributed system composed of multiple devices.

[0288] Computer-readable storage media

[0289] In this embodiment, a computer-readable storage medium is also provided, on which a computer program is stored. When the computer program is executed by a processor, it implements an integrated management method for AI applications as described in any embodiment of this disclosure.

[0290] This disclosure may be an apparatus, method, and / or computer program product. A computer program product may include a computer-readable storage medium having computer-readable program instructions loaded thereon for causing a processor to implement various aspects of this disclosure.

[0291] Computer-readable storage media can be tangible devices capable of holding and storing instructions for use by an instruction execution device. Computer-readable storage media can be, for example—but not limited to—electrical storage devices, magnetic storage devices, optical storage devices, electromagnetic storage devices, semiconductor storage devices, or any suitable combination thereof. More specific examples (a non-exhaustive list) of computer-readable storage media include: portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), static random access memory (SRAM), portable compact disc read-only memory (CD-ROM), digital multifunction disc (DVD), memory sticks, floppy disks, mechanical encoding devices, such as punch cards or recessed protrusions storing instructions thereon, and any suitable combination thereof. The computer-readable storage media used herein are not to be construed as transient signals themselves, such as radio waves or other freely propagating electromagnetic waves, electromagnetic waves propagating through waveguides or other transmission media (e.g., light pulses through fiber optic cables), or electrical signals transmitted through wires.

[0292] The computer-readable program instructions described herein can be downloaded from computer-readable storage media to various computing / processing devices, or downloaded via a network, such as the Internet, local area network, wide area network, and / or wireless network, to an external computer or external storage device. The network may include copper transmission cables, fiber optic transmission, wireless transmission, routers, firewalls, switches, gateway computers, and / or edge servers. A network adapter card or network interface in each computing / processing device receives the computer-readable program instructions from the network and forwards them to the computer-readable storage media in the respective computing / processing device.

[0293] Computer program instructions used to perform the operations of this disclosure may be assembly instructions, instruction set architecture (ISA) instructions, machine instructions, machine-dependent instructions, microcode, firmware instructions, status setting data, or source code or object code written in any combination of one or more programming languages, including object-oriented programming languages ​​such as Smalltalk, C++, etc., and conventional procedural programming languages ​​such as the "C" language or similar programming languages. The computer-readable program instructions may execute entirely on the user's computer, partially on the user's computer, as a standalone software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving a remote computer, the remote computer may be connected to the user's computer via any type of network—including a local area network (LAN) or a wide area network (WAN)—or may be connected to an external computer (e.g., via the Internet using an Internet service provider). In some embodiments, electronic circuitry, such as programmable logic circuitry, field-programmable gate arrays (FPGAs), or programmable logic arrays (PLAs), is personalized by utilizing the status information of the computer-readable program instructions to implement various aspects of this disclosure.

[0294] Various aspects of this disclosure are described herein with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this disclosure. It should be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer-readable program instructions.

[0295] These computer-readable program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing apparatus to produce a machine such that, when executed by the processor of the computer or other programmable data processing apparatus, they create means for implementing the functions / actions specified in one or more blocks of the flowchart and / or block diagram. These computer-readable program instructions can also be stored in a computer-readable storage medium that causes a computer, programmable data processing apparatus, and / or other device to operate in a particular manner; thus, the computer-readable medium storing the instructions comprises an article of manufacture that includes instructions for implementing aspects of the functions / actions specified in one or more blocks of the flowchart and / or block diagram.

[0296] Computer-readable program instructions may also be loaded onto a computer, other programmable data processing apparatus, or other device to cause a series of operational steps to be performed on the computer, other programmable data processing apparatus, or other device to produce a computer-implemented process, thereby causing the instructions executed on the computer, other programmable data processing apparatus, or other device to perform the functions / actions specified in one or more boxes of a flowchart and / or block diagram.

[0297] The flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to various embodiments of the present disclosure. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of an instruction containing one or more executable instructions for implementing a specified logical function. In some alternative implementations, the functions marked in the blocks may occur in a different order than those marked in the drawings. For example, two consecutive blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in the block diagrams and / or flowcharts, and combinations of blocks in the block diagrams and / or flowcharts, can be implemented using a dedicated hardware-based system that performs the specified function or action, or using a combination of dedicated hardware and computer instructions. It will be known to those skilled in the art that implementation in hardware, implementation in software, and implementation in a combination of software and hardware are equivalent.

[0298] The various embodiments of this disclosure have been described above. These descriptions are exemplary and not exhaustive, and are not limited to the disclosed embodiments. Many modifications and variations will be apparent to those skilled in the art without departing from the scope and spirit of the described embodiments. The terminology used herein is chosen to best explain the principles, practical application, or technical improvements to the embodiments in the market, or to enable others skilled in the art to understand the embodiments disclosed herein. The scope of this disclosure is defined by the appended claims.

Claims

1. An integrated management method for artificial intelligence applications, comprising: Obtain an installation file in a predetermined standard format for the first artificial intelligence application, and install the first artificial intelligence application based on the installation file in the predetermined standard format; Obtain the construction configuration information of the second artificial intelligence application, and construct the second artificial intelligence application based on the construction configuration information; Manage the installed first artificial intelligence application and the constructed second artificial intelligence application; The acquisition of the installation file in the predetermined standard format of the first artificial intelligence application includes: Obtain the original installation file of the first artificial intelligence application, and convert the format of the original installation file into the predetermined standard format; the original installation file is an application installation file provided by a third-party service. The process of obtaining the construction configuration information of the second artificial intelligence application includes: The second artificial intelligence application's build configuration information is received based on a predetermined configuration process, wherein the predetermined configuration process is a standard application build process.

2. The method according to claim 1, wherein the predetermined standard format includes the Docker standard format.

3. The method according to claim 1, The construction configuration information includes storage configuration information and computation configuration information; The construction of the second artificial intelligence application based on the configuration information includes: Based on the storage configuration information and the computing configuration information, the second artificial intelligence application is constructed.

4. The method according to claim 1, further comprising: Provide an app store that offers at least one first artificial intelligence application; The acquisition of the installation file in the predetermined standard format of the first artificial intelligence application includes: Download the installation file of the first artificial intelligence application in a predetermined standard format from the app store; or Download the original installation file of the first artificial intelligence application from the app store, and convert the format of the original installation file to the predetermined standard format.

5. The method according to claim 1, wherein managing the installed first artificial intelligence application and the constructed second artificial intelligence application includes at least one of the following: Acquire the data generated by the first artificial intelligence application and the second artificial intelligence application during operation, and store the generated data; Collect the operation data of the first artificial intelligence application and the second artificial intelligence application to obtain the operation logs of the first artificial intelligence application and the second artificial intelligence application; Monitor at least one of the currently running processes, operators, workflows, and running phases of the first and second artificial intelligence applications to obtain the resources occupied by at least one of the processes, operators, workflows, and running phases; Monitor user actions on the first AI application and the second AI application, and obtain audit records of the first AI application and the second AI application; The system monitors specified metrics of the first and second AI applications during operation and issues an alarm notification when the specified metrics are abnormal. In response to the operation of running the first artificial intelligence application and the second artificial intelligence application, the user's permission to operate the first artificial intelligence application and the second artificial intelligence application is verified, and if the user has the permission to operate the first artificial intelligence application and the second artificial intelligence application, the first artificial intelligence application and the second artificial intelligence application are run; The system accepts and stores user-uploaded data, and then provides the uploaded data to the first and second AI applications for consumption.

6. The method according to claim 5, After storing the generated data, the method further includes: In response to an operation to view the output data, the output data is displayed; or, After obtaining the audit records of the first AI application and the second AI application, the method further includes: displaying the audit records in response to an operation of viewing the audit records; or, After obtaining the operation logs of the first artificial intelligence application and the second artificial intelligence application, the method further includes: displaying the operation logs in response to the operation of viewing the operation logs; or, After acquiring the resources occupied by at least one of the process, the operator, the workflow, and the running phase, the method further includes: displaying the occupied resources in response to an operation of viewing the occupied resources.

7. The method according to claim 5, wherein providing the uploaded data to the first AI application and the second AI application for consumption comprises: Artificial intelligence data related to the first artificial intelligence application and the second artificial intelligence application is obtained from the uploaded data, and the artificial intelligence data is provided to the first artificial intelligence application and the second artificial intelligence application for consumption; or, Artificial intelligence data related to the first and second artificial intelligence applications is obtained from the uploaded data. Data matching the application scenarios corresponding to the first and second artificial intelligence applications is filtered out from the artificial intelligence data and provided to the first and second artificial intelligence applications for consumption.

8. The method according to claim 1, further comprising: Determine the user's operating permissions for the integrated artificial intelligence application, and provide the corresponding artificial intelligence application based on the operating permissions.

9. The method according to claim 1, further comprising: Provide an administrative desktop, on which shortcuts to some or all integrated artificial intelligence applications and / or management tools and / or file shortcuts are provided; In response to an editing operation on the desktop, the shortcuts provided on the desktop are edited.

10. The method according to claim 3, wherein a portion of the storage configuration information and the computing configuration information is obtained through user configuration feature capabilities, wherein, The aforementioned feature capabilities refer to entities whose available resources have been encapsulated for direct invocation of related capabilities; The step of constructing the second artificial intelligence application based on the storage configuration information and the computing configuration information includes: calling corresponding feature capabilities based on a portion of the information in the storage configuration information and the computing configuration information to construct the second artificial intelligence application.

11. The method according to claim 3, wherein the storage configuration information includes at least one of input source information, application metadata, log and monitoring metric aggregation information, and storage declaration information; The computing configuration information includes at least one of the following: service startup method information, dynamic resource information, and computing capacity information.

12. An integrated management device for an artificial intelligence application, comprising: The first installation module is used to obtain an installation file of a predetermined standard format for the first artificial intelligence application, and to install the first artificial intelligence application based on the installation file of the predetermined standard format. The second construction module is used to obtain the construction configuration information of the second artificial intelligence application and construct the second artificial intelligence application based on the construction configuration information; The application management module is used to manage the installed first artificial intelligence application and the constructed second artificial intelligence application; The first installation module is further configured to obtain the original installation file of the first artificial intelligence application and convert the format of the original installation file into the predetermined standard format; the original installation file is an application installation file provided by a third-party service. The second building module is used to obtain the building configuration information of the second artificial intelligence application, including: The second construction module is used to receive the construction configuration information of the second artificial intelligence application based on a predetermined configuration process, wherein the predetermined configuration process is a standard application construction process.

13. The integrated management device according to claim 12, wherein the predetermined standard format includes the Docker standard format.

14. The integrated management device according to claim 12, The construction configuration information includes storage configuration information and computation configuration information; The second building module is used to: build the second artificial intelligence application based on the storage configuration information and the computing configuration information.

15. The integrated management device according to claim 12, further comprising: An app store providing module is used to provide an app store, which provides at least one first artificial intelligence application; The first installation module is configured to include: Download the installation file of the first artificial intelligence application in a predetermined standard format from the app store; or Download the original installation file of the first artificial intelligence application from the app store, and convert the format of the original installation file to the predetermined standard format.

16. The integrated management device according to claim 12, wherein the application management module is configured to perform at least one of the following: Acquire the data generated by the first artificial intelligence application and the second artificial intelligence application during operation, and store the generated data; Collect the operation data of the first artificial intelligence application and the second artificial intelligence application to obtain the operation logs of the first artificial intelligence application and the second artificial intelligence application; Monitor at least one of the currently running processes, operators, workflows, and running phases of the first and second artificial intelligence applications to obtain the resources occupied by at least one of the processes, operators, workflows, and running phases; Monitor user actions on the first AI application and the second AI application, and obtain audit records of the first AI application and the second AI application; The system monitors specified metrics of the first and second AI applications during operation and issues an alarm notification when the specified metrics are abnormal. In response to the operation of running the first artificial intelligence application and the second artificial intelligence application, the user's permission to operate the first artificial intelligence application and the second artificial intelligence application is verified, and if the user has the permission to operate the first artificial intelligence application and the second artificial intelligence application, the first artificial intelligence application and the second artificial intelligence application are run; The system accepts and stores user-uploaded data, and then provides the uploaded data to the first and second AI applications for consumption.

17. The integrated management device according to claim 16, further comprising a display module, used for: After storing the generated data, the generated data is displayed in response to the operation of viewing the generated data; or, After obtaining the audit records of the first AI application and the second AI application, in response to the operation of viewing the audit records, the audit records are displayed; or, After obtaining the operation logs of the first AI application and the second AI application, the operation logs are displayed in response to the operation of viewing the operation logs; or, After obtaining the resources occupied by at least one of the process, the operator, the workflow, and the running phase, the occupied resources are displayed in response to the operation of viewing the occupied resources.

18. The integrated management device according to claim 16, wherein the application management module is used for: Artificial intelligence data related to the first artificial intelligence application and the second artificial intelligence application is obtained from the uploaded data, and the artificial intelligence data is provided to the first artificial intelligence application and the second artificial intelligence application for consumption; or, Artificial intelligence data related to the first and second artificial intelligence applications is obtained from the uploaded data. Data matching the application scenarios corresponding to the first and second artificial intelligence applications is filtered out from the artificial intelligence data and provided to the first and second artificial intelligence applications for consumption.

19. The integrated management device according to claim 12, further comprising: The permission management module is used to determine the user's operation permissions for the integrated artificial intelligence applications, and to provide the corresponding artificial intelligence applications based on the operation permissions.

20. The integrated management device according to claim 12, further comprising a desktop management module, used for: Provide an administrative desktop, on which shortcuts to some or all integrated artificial intelligence applications and / or management tools and / or file shortcuts are provided; In response to the editing operation on the management desktop, the shortcuts provided on the desktop are edited.

21. The integrated management device according to claim 14, wherein a portion of the storage configuration information and the computing configuration information is obtained through user configuration feature capabilities, wherein, The aforementioned feature capabilities refer to entities whose available resources have been encapsulated for direct invocation of related capabilities; The second building module is used to: invoke corresponding feature capabilities based on a portion of the storage configuration information and the computing configuration information to build the second artificial intelligence application.

22. The integrated management device according to claim 14, wherein the storage configuration information includes at least one of input source information, application metadata, log and monitoring metric aggregation information, and storage declaration information; The computing configuration information includes at least one of the following: service startup method information, dynamic resource information, and computing capacity information.

23. An integrated management system for artificial intelligence applications, comprising at least one computing device and at least one storage device, wherein, The at least one storage device is used to store instructions for controlling the at least one computing device to perform the method according to any one of claims 1 to 11.

24. A computer-readable storage medium having a computer program stored thereon, the computer program implementing the method as described in any one of claims 1 to 11 when executed by a processor.

Images