Digital employee multi-channel cloud desktop expansion system and method

Abstract

The invention relates to a digital employee multi-channel cloud desktop expansion system and method, and the system comprises a universal interface API, an API adaptation layer, a dynamic scalable algorithm and a region proximity judgment module, and achieves multi-cloud access, request conversion, resource adaptation and nearby startup. The method comprises the steps of interface definition, request conversion, quantity adjustment and near resource selection. The digital employee multi-channel cloud desktop expansion method is used for controlling a related system and comprises the steps that the universal interface API module defines a uniform interface, the adaptation layer converts requests, the scalable algorithm module adjusts the number, and the region proximity principle module selects a close cloud desktop to be started. The method has a remarkable effect in the field of multi-channel cloud desktops. Uniform interface APIs reduce repeated development, reduce cost and improve efficiency; resource allocation is accurate and dynamic, and waste and waiting are avoided; task queuing is optimized, and waiting time is shortened; finely closing the strategy to reduce idleness and guarantee continuity; resources are selected near regions, delay is reduced, performance is optimized, and stability and efficiency are improved.

Description

Technical Field

[0001] This application relates to the fields of cloud computing and API interface technology, and in particular to a system and method for extending the multi-channel cloud desktop for digital employees. Background Technology

[0002] In the field of cloud desktop technology, the current market is dominated by providers such as Alibaba Cloud, Tencent Cloud, Huawei Cloud, and China Telecom Cloud, but there is a lack of unified multi-channel expansion solutions. Existing systems are mostly limited to a single cloud platform architecture, which is highly closed. This leads to enterprises having to repeatedly develop and deploy when using cloud desktops on multiple platforms, significantly increasing costs. Furthermore, cross-platform data interaction and business integration are difficult, which restricts application efficiency and flexibility.

[0003] In scenarios combining RPA tasks with cloud desktops, there are significant shortcomings in resource allocation and task management: the parameters of the dynamic scalable algorithm are set in a coarse manner, lacking RPA task execution time estimation and multi-dimensional resource management mechanisms, and failing to effectively adjust dynamically based on the number of queued orders, resulting in idle resources or excessively long task waiting times; the task queuing mechanism is imperfect, with low and imprecise adjustment frequency for queueable tasks and a rigid control mechanism for non-queuable tasks; the cloud desktop shutdown policy is executed infrequently and has a single trigger condition, easily leading to resource waste or accidental shutdown; and the lack of geographical proximity judgment when starting a new cloud desktop increases transmission latency and costs, affecting task efficiency.

[0004] These problems severely restrict the efficient utilization of cloud desktop resources and business continuity, and a comprehensive solution is urgently needed. Summary of the Invention

[0005] In view of this, this application proposes a system for extending the multi-channel cloud desktop for digital employees, characterized by comprising: a programming interface module, an interface adaptation layer module, a scalable algorithm module, and a proximity principle judgment module;

[0006] The programming interface module is used to dynamically manage access points of multiple cloud platforms, analyze the cloud desktop functions and interface specifications of each cloud platform, abstract common functions, and define a unified interface specification for basic operations.

[0007] The interface adaptation layer module contains the application programming interface calling logic and data conversion mechanism of various cloud platforms, which is used to convert upper-layer application requests into native application programming interface calls and process the results.

[0008] The scalable algorithm module is used to realize the elastic scaling of cloud desktop resources, dynamically allocate cloud desktop resources in the multi-channel cloud desktop extension system, and make the number of cloud desktops adaptively adjust according to business needs.

[0009] The proximity principle judgment module is used to select the cloud desktop resource closest to the server's location based on the predefined application programming interface launch channel and region correspondence.

[0010] In one possible implementation, the scalable algorithm module includes a template frame unit, a frame algorithm unit, and a queuing unit.

[0011] The template framework unit is used to analyze business processes and data characteristics for different top-level application scenarios, and to customize parameter templates for each top-level application, covering configuration information for data collection, processing, and storage. The parameter template is automatically loaded when executing robotic process automation tasks.

[0012] The framework algorithm unit is used for fine-grained allocation and management of cloud desktop resources in the scalable algorithm module, and multi-dimensional parameter analysis and algorithm model realize dynamic matching of cloud desktop resources and business requirements.

[0013] The queuing unit is used to handle scenarios where tasks can be queued in cloud desktop resource scheduling. Parameter settings and resource calculation logic enable dynamic matching between cloud desktop resources and queued task requirements.

[0014] In one possible implementation, the framework algorithm unit includes a task time estimation subunit and an idle cloud desktop detection subunit.

[0015] The task time estimation sub-unit uses historical task data, combined with factors such as task type, data volume, and operational complexity, to build a model using empirical formulas or machine learning algorithms to estimate the execution time of each robotic process automation task.

[0016] The online idle cloud desktop quantity monitoring subunit establishes a distributed real-time monitoring system that interacts with the cloud platform management application programming interface to periodically obtain the online idle cloud desktop quantity.

[0017] In one possible implementation, the framework algorithm unit includes an order and new desktop subunit, a desktop idle management subunit, and an order utilization subunit;

[0018] The order urgency and new cloud desktop startup sub-unit sets an urgency level for orders and sets a monitoring module to count the number of queued orders and waiting time. When the waiting time of a queued order exceeds the preset queuing time and the urgency level is high, the new cloud desktop is triggered.

[0019] The cloud desktop idle management subunit sets a parameter to shut down the cloud desktop after the idle time exceeds the preset idle time based on enterprise cost-effectiveness and business fluctuations. The cloud desktop records the idle time during runtime, calls the application programming interface to shut it down, and combines order forecast and task scheduling information to avoid accidental shutdown.

[0020] The order queue quantity is updated in real time using sub-units, and the dynamic calculation model determines whether to launch a new cloud desktop or close an idle cloud desktop.

[0021] In one possible implementation, the queuing unit includes a parameter setting subunit and a resource calculation and startup logic subunit.

[0022] The parameter setting sub-unit presets key parameters related to task processing and cloud desktop resource management, providing a baseline value for cloud desktop resource allocation in scenarios where tasks can be queued.

[0023] The resource and startup subunit monitors the task status in real time and calculates the required amount of cloud desktop resources based on preset parameters, thereby achieving dynamic matching between cloud desktop resources and the needs of queued tasks.

[0024] This invention also includes a method for extending the multi-channel cloud desktop for digital employees, characterized in that the method for controlling the above-mentioned system includes the following steps:

[0025] The programming interface module analyzes the functions and interface specifications of cloud desktops on various cloud platforms, abstracts common functions, and then defines a unified interface specification for basic operations.

[0026] The interface adaptation layer module converts upper-layer application requests into native application programming interface calls and processes the results;

[0027] The scalable algorithm module adaptively adjusts the number of cloud desktops according to business needs; and

[0028] The proximity principle is used to determine the correspondence between the module's startup channel and the region, and the cloud desktop resource closest to the server's region is selected for startup.

[0029] In one possible implementation, the scalable algorithm module processing steps include:

[0030] The template framework unit customizes the parameter template of the configuration information for each top-level application, and automatically loads the parameter template when executing robotic process automation tasks;

[0031] The framework algorithm unit allocates and manages cloud desktop resources, enabling dynamic matching of cloud desktop resources with business needs;

[0032] In scenarios where tasks can be processed in a queue, cloud desktop resources are dynamically matched with the needs of queued tasks.

[0033] In one possible implementation, the frame algorithm unit processing steps include:

[0034] The task time estimation sub-unit builds a model to estimate the execution time of each robotic process automation task;

[0035] The online idle cloud desktop quantity monitoring subunit interacts with the cloud platform management application programming interface to periodically obtain the online idle cloud desktop quantity.

[0036] In one possible implementation, the framework algorithm unit processing steps further include:

[0037] The order urgency level and new cloud desktop startup sub-unit set urgency level indicators for orders and count the number of queued orders and waiting time;

[0038] When the waiting time for a queued order exceeds the preset queuing time and the urgency level is high, a new cloud desktop will be launched.

[0039] The cloud desktop idle management subunit closes the idle cloud desktop after the idle time exceeds the preset idle time.

[0040] The cloud desktop records idle time during runtime, calls the application programming interface to shut it down, and combines order forecasting and task scheduling information to avoid accidental shutdown.

[0041] In one possible implementation, the queuing unit processing steps include:

[0042] The parameter setting sub-unit presets key parameters related to task processing and cloud desktop resource management;

[0043] The resource and startup subunits combine real-time monitoring of task status and preset parameters to calculate the required amount of cloud desktop resources.

[0044] The resource and startup subunit dynamically matches cloud desktop resources with queued task requirements based on the number of desktop resources available.

[0045] The beneficial effects of this invention are:

[0046] This invention has yielded significant results in the field of multi-channel cloud desktops. By using a unified API, it reduces redundant development, lowers costs and complexity, promotes cross-platform data interaction and business integration, and improves efficiency.

[0047] Resource allocation is precise and dynamic, with template parameters set for different applications and calculations based on multi-dimensional data to avoid waste and waiting. Task queuing is optimized to cope with traffic fluctuations and shorten waiting time.

[0048] The cloud desktop shutdown policy is more granular, employing high-frequency detection and multi-condition judgment to reduce idle time while ensuring business continuity. Resources are selected based on geographical proximity, reducing transmission latency, optimizing network performance, and enhancing system stability and overall efficiency.

[0049] Other features and aspects of this application will become clear from the following detailed description of exemplary embodiments with reference to the accompanying drawings. Attached Figure Description

[0050] The accompanying drawings, which are included in and form part of this specification, illustrate exemplary embodiments, features, and aspects of this application together with the specification and serve to explain the principles of this application.

[0051] According to the document, the figures involved and their corresponding titles are as follows:

[0052] Figure 1 This diagram illustrates the main structural structure of the apparatus according to an embodiment of this application.

[0053] Figure 2 A flowchart illustrating a method for extending a multi-channel cloud desktop for digital employees, according to an embodiment of this application, is shown.

[0054] Figure 3 A flowchart illustrating the processing steps of the scalable algorithm module in an embodiment of this application is shown.

[0055] Figure 4A A flowchart illustrating the processing steps of a framework algorithm unit according to an embodiment of this application is shown.

[0056] Figure 4B A flowchart illustrating the processing steps of a framework algorithm unit according to an embodiment of this application is shown.

[0057] Figure 5 A flowchart illustrating the processing steps of the queuing unit according to an embodiment of this application is shown. Detailed Implementation

[0058] Various exemplary embodiments, features, and aspects of this application will now be described in detail with reference to the accompanying drawings. The same reference numerals in the drawings denote elements that have the same or similar functions. Although various aspects of the embodiments are shown in the drawings, they are not necessarily drawn to scale unless specifically indicated otherwise.

[0059] It should be understood that the terms "center," "longitudinal," "lateral," "length," "width," "upper," "lower," "front," "rear," "left," "right," "vertical," "horizontal," "top," "bottom," "inner," "outer," "clockwise," "counterclockwise," "axial," "radial," and "circumferential" indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings. They are used only for the convenience of describing the present invention or simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, they should not be construed as limitations on the present invention.

[0060] Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of that feature. In the description of this invention, "a plurality of" means two or more, unless otherwise explicitly specified.

[0061] The term “exemplary” as used herein means “serving as an example, embodiment, or illustration.” Any embodiment illustrated herein as “exemplary” is not necessarily to be construed as superior to or better than other embodiments.

[0062] Furthermore, to better illustrate this application, numerous specific details are provided in the following detailed embodiments. Those skilled in the art should understand that this application can be implemented without certain specific details. In some instances, methods, means, components, and circuits well-known to those skilled in the art have not been described in detail in order to highlight the main points of this application.

[0063] The solution proposed in this application is applied to flexible employment scenarios in enterprises, is compatible with multi-cloud platform environments, and solves the problems of scattered management and inefficient resource allocation of multi-platform cloud desktops.

[0064] The overall function of the technical solution in this application is to build a multi-channel cloud desktop extension system. A programming interface module unifies multi-cloud access standards, reducing cross-platform development costs; an application programming interface (API) adaptation layer implements request conversion and result processing, connecting upper-layer applications and the cloud platform; a scalable algorithm module adaptively adjusts the number of cloud desktops based on parameters such as task type and urgency, balancing resource utilization and business needs; and a proximity judgment module selects the nearest resource for startup, reducing latency. Furthermore, the system in this application supports automated execution of robotic process automation tasks, optimizes task queuing and resource scheduling, and improves the efficiency and stability of remote work and business process automation for enterprises.

[0065] Digital Employee Multi-Channel Cloud Desktop Extension System

[0066] specifically refer to Figure 1 , Figure 1 A structural diagram of a digital employee multi-channel cloud desktop extension system according to an embodiment of this application is shown.

[0067] A digital employee multi-channel cloud desktop extension system 100 is characterized by comprising: a programming interface module 101, an interface adaptation layer module 102, a scalable algorithm module 103, and a proximity principle judgment module 104.

[0068] The programming interface module 101 is used to dynamically manage access points of multiple cloud platforms, analyze the cloud desktop functions and interface specifications of each cloud platform, abstract common functions, and define a unified interface specification for basic operations.

[0069] The interface adaptation layer module 102 includes the application programming interface calling logic and data conversion mechanism of various cloud platforms, which is used to convert upper-layer application requests into native application programming interface calls and process the results.

[0070] The scalable algorithm module 103 is used to realize the elastic scaling of cloud desktop resources, dynamically allocate cloud desktop resources in the multi-channel cloud desktop extension system 100, and make the number of cloud desktops adaptively adjust according to business needs.

[0071] The proximity principle judgment module 104 is used to select the cloud desktop resource closest to the server's location based on the predefined application programming interface launch channel and region correspondence.

[0072] Specifically, the proximity principle for rescalable cloud desktops determines the IP address by pinging the main service platform's website domain name, mapping it to a province or region using an IP address database or GIS technology, and selecting the cloud desktop resource closest to the server's region based on the predefined API launch channel and region correspondence, thereby reducing data transmission latency and improving network performance and resource utilization efficiency.

[0073] In one possible implementation, the scalable algorithm module 103 includes a template frame unit 110, a frame algorithm unit 111, and a queuing unit 112.

[0074] The template framework unit 110 is used to analyze business processes and data characteristics for different top-level application scenarios, and to customize parameter templates covering configuration information for data collection, processing and storage for each top-level application. The parameter template is automatically loaded when executing robotic process automation tasks.

[0075] The framework algorithm unit 111 is used for the fine-grained allocation and management of cloud desktop resources in the scalable algorithm module 103, and the multi-dimensional parameter analysis and algorithm model realize the dynamic matching of cloud desktop resources and business requirements.

[0076] The queuing unit 112 is used to handle scenarios where tasks can be queued in cloud desktop resource scheduling. Parameter settings and resource calculation logic enable dynamic matching between cloud desktop resources and queued task requirements.

[0077] In one possible implementation, the framework algorithm unit 111 includes a task time estimation subunit 120 and an idle cloud desktop detection subunit 121;

[0078] The task time estimation subunit 120 estimates the execution time of each robotic process automation task by building a model using empirical formulas or machine learning algorithms, based on historical task data and considering factors such as task type, data volume, and operational complexity. Specifically, the task time estimation subunit 120 uses historical task data as a foundation, comprehensively analyzes key factors such as task type, data volume, and operational complexity, and constructs a quantitative evaluation model using empirical formulas or machine learning algorithms to accurately estimate the execution time of each robotic process automation task. This function provides a time-based reference for subsequent resource allocation, avoiding resource shortages or idleness caused by errors in task time estimation.

[0079] The online idle cloud desktop quantity monitoring subunit 121 establishes a distributed real-time monitoring system that interacts with the cloud platform management application programming interface to periodically obtain the number of online idle cloud desktops. Specifically, the online idle cloud desktop quantity monitoring subunit 121 builds a distributed real-time monitoring system that interacts in real time with the application programming interfaces of each cloud platform management system to obtain the number of currently online and idle cloud desktops at a preset frequency. This subunit ensures that the system can grasp the resource inventory in real time, providing immediate data support for dynamically adjusting the number of cloud desktops and avoiding resource scheduling delays.

[0080] In one possible implementation, the framework algorithm unit 111 includes an order and new desktop subunit 122, a desktop idle management subunit 123, and an order utilization subunit 124.

[0081] In one specific embodiment, the order urgency level and the new cloud desktop startup subunit 122 set an urgency level identifier for each order, based on factors such as order source, business type, and user priority. Simultaneously, the monitoring module continuously monitors the number of queued orders and the waiting time for each order. When the waiting time for an order exceeds a preset threshold and its urgency level is high, the system automatically triggers the new cloud desktop startup process, thereby prioritizing the timeliness of urgent business and preventing delays in critical tasks.

[0082] In one specific embodiment, the cloud desktop idle management subunit 123 presets a cloud desktop idle shutdown threshold based on the enterprise's cost-effectiveness goals and business fluctuation patterns. During cloud desktop operation, its idle time is recorded in real time. When the accumulated idle time reaches the threshold, the cloud platform API is automatically invoked to execute a shutdown operation. Simultaneously, this subunit combines order forecast data and task scheduling plans to avoid mistakenly shutting down cloud desktops when new tasks are about to be assigned, thereby reducing resource idleness while ensuring business continuity.

[0083] In one specific embodiment, the order utilization subunit 124 synchronizes and updates the number of queued orders in real time, incorporating this data along with parameters such as the execution time of the robotic process automation task, the number of online idle cloud desktops, the order urgency threshold, and the cloud desktop idle shutdown threshold into a dynamic calculation model. For example, when the number of queued orders surges, the model determines that more cloud desktops are needed to speed up processing; when the number of queued orders decreases sharply and there are a large number of idle cloud desktops, it triggers a shutdown command for some idle resources. In this way, precise matching of resources and demand is achieved, improving overall resource utilization efficiency.

[0084] The order urgency level and new cloud desktop startup subunit 122 assigns urgency level indicators to orders. A monitoring module tracks the number of queued orders and their waiting times. When a queued order's waiting time exceeds a preset queuing time and its urgency level is high, a new cloud desktop is triggered. After the cloud desktop starts, task priority allocation and resource binding are performed. Specifically, the system marks the high-urgency order task as the highest priority, assigns it to the newly started cloud desktop, and completes the transmission and loading of task data to the cloud desktop via an interface. Simultaneously, the queued order list is updated, and the task is removed from the waiting queue to ensure immediate execution. Furthermore, the startup time of the new cloud desktop, bound task information, and its region are recorded synchronously to provide data support for subsequent resource monitoring and task progress tracking.

[0085] This operation ensures the timely processing of high-urgency tasks, avoids chaotic task allocation through resource binding, and works in conjunction with other parameters in the overall dynamic scalable algorithm to maintain the accuracy of resource allocation.

[0086] The cloud desktop idle management subunit 123 sets parameters for shutting down cloud desktops after an idle period exceeding a preset time, based on enterprise cost-effectiveness and business fluctuations. The cloud desktop records idle time during runtime, calls the application programming interface to shut it down, and incorporates order forecasting and task scheduling information to avoid accidental shutdowns. High-frequency detection accelerates resource release and reduces idle time; multi-condition judgment avoids accidental shutdowns and optimizes management. It selects nearby resources to reduce transmission latency and improve the speed of sensitive services; it optimizes network performance and increases system stability.

[0087] The order uses sub-unit 124 to update the number of queued orders in real time, and the dynamic calculation model determines whether to start a new cloud desktop or close an idle cloud desktop.

[0088] In one possible implementation, the queuing unit 112 includes a parameter setting subunit 130 and a resource calculation and startup logic subunit 131.

[0089] The parameter setting subunit 130 presets key parameters related to task processing and cloud desktop resource management, providing a baseline value for cloud desktop resource allocation in scenarios where tasks can be queued.

[0090] Specifically, the parameters preset by the parameter setting subunit 130 include core parameters related to task execution, such as the average execution time of robotic process automation tasks; threshold parameters related to queuing mechanisms, such as the maximum queuing time; and cloud desktop resource retention parameters, such as the preset cloud desktop retention time. These parameters together constitute a baseline numerical system for resource allocation, providing a fixed reference standard for subsequent dynamic calculations and ensuring the consistency of resource allocation logic in different scenarios.

[0091] The resource and startup subunit 131 monitors the task status in real time and calculates the required amount of cloud desktop resources based on preset parameters, so as to achieve dynamic matching between cloud desktop resources and the needs of queued tasks.

[0092] Specifically, the workflow of the resource and startup subunit 131 includes receiving and synchronizing the current task status data in real time; calling the preset baseline parameters in the parameter setting subunit 130; and outputting resource adjustment instructions based on the calculation results: when the result is greater than 0, a signal to start a new cloud desktop is generated to ensure that the added resources meet the requirements of the queued tasks; when the result is less than or equal to 0, the startup of new resources is not triggered to avoid resource redundancy.

[0093] The entire process is linked to the task scheduling system, ensuring that the calculation results are synchronized with the actual resource allocation actions, and achieving a real-time dynamic balance between cloud desktop resource supply and queued task demand.

[0094] Obviously, those skilled in the art should understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. The program can be stored in a computer-readable storage medium, and when executed, it can include the processes of the embodiments of the control methods described above. The modules or steps of the present invention described above can be implemented using general-purpose computing devices. They can be centralized on a single computing device or distributed across a network of multiple computing devices. Optionally, they can be implemented using computer-executable program code, thereby storing them in a storage device for execution by a computing device, or fabricating them separately as individual integrated circuit modules, or fabricating multiple modules or steps into a single integrated circuit module. Thus, the present invention is not limited to any specific hardware and software combination.

[0095] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. The program can be stored in a computer-readable storage medium, and when executed, it can include the processes of the embodiments of the control methods described above. The storage medium can be a magnetic disk, optical disk, read-only memory, random access memory, flash memory, hard disk, or solid-state drive, etc.; the storage medium can also include combinations of the above types of memory.

[0096] Methods for extending digital employee cloud desktops across multiple channels

[0097] specifically refer to Figure 2 , Figure 2 A flowchart illustrating a method for extending a digital employee multi-channel cloud desktop according to an embodiment of this application is shown.

[0098] This invention also includes a method for expanding multi-channel cloud desktops for digital employees. The method, used to control the aforementioned system, includes the following steps: Step 201: A programming interface module analyzes the cloud desktop functions and interface specifications of various cloud platforms, abstracts common functions, and defines a unified interface specification for basic operations. Step 202: An interface adaptation layer module converts upper-layer application requests into native application programming interface calls and processes the results. Step 203: A scalable algorithm module adaptively adjusts the number of cloud desktops according to business needs. Step 204: A proximity principle is used to determine the correspondence between the module's launch channel and region, selecting the cloud desktop resource closest to the server's region for launch. The unified interface reduces redundant development, lowers costs and complexity, promotes cross-platform data interaction, and improves business integration and processing efficiency.

[0099] specifically refer to Figure 3 , Figure 3 A flowchart illustrating the processing steps of the scalable algorithm module according to an embodiment of this application is shown.

[0100] In one possible implementation, the scalable algorithm module processing steps include: Step 301, where the template framework unit customizes parameter templates with configuration information for each top-level application, and automatically loads these parameter templates when executing robotic process automation tasks; Step 302, where the framework algorithm unit allocates and manages cloud desktop resources, dynamically matching cloud desktop resources with business needs; and Step 303, where the queuing unit handles scenarios where tasks can be queued, dynamically matching cloud desktop resources with queued task requirements. The combination of targeted templates and multiple parameters accurately calculates resource requirements; dynamically adjusting resources to adapt to task changes improves resource adaptability.

[0101] specifically refer to Figure 4A , Figure 4A A flowchart illustrating the processing steps of a framework algorithm unit according to an embodiment of this application is shown.

[0102] In one possible implementation, the framework algorithm unit processing steps include step 401, where the task time estimation subunit constructs a model to estimate the execution time of each robotic process automation task. Step 402, where the online idle cloud desktop quantity monitoring subunit interacts with the cloud platform management application programming interface to periodically obtain the number of online idle cloud desktops. In non-queuing situations, thresholds and rhythms are optimized to cope with fluctuations and improve timeliness and utilization. In queuing situations, resources are dynamically calculated to accelerate the execution rhythm, shorten waiting time, and improve satisfaction.

[0103] specifically refer to Figure 4B , Figure 4B A flowchart illustrating the processing steps of a framework algorithm unit according to an embodiment of this application is shown.

[0104] Specifically, the framework algorithm unit processing steps also include step 411, where the order urgency and new cloud desktop startup subunit sets an urgency level identifier for each order and counts the number of queued orders and their waiting time. Step 412, when the waiting time of a queued order exceeds a preset queuing time and the urgency level is high, a new cloud desktop startup is triggered. Step 413, the cloud desktop idle management subunit shuts down idle cloud desktops after their idle time exceeds a preset idle time. And step 414, during cloud desktop runtime, the idle time is recorded, the application programming interface is called to shut it down, and order prediction and task scheduling information are combined to avoid accidental shutdown. High-frequency detection accelerates resource release and reduces idle time; multi-condition judgment avoids accidental shutdown and optimizes resource management.

[0105] specifically refer to Figure 5 , Figure 5 A flowchart illustrating the processing steps of the queuing unit according to an embodiment of this application is shown.

[0106] In one possible implementation, the queuing unit processing steps include step 501, where the parameter setting subunit presets key parameters related to task processing and cloud desktop resource management. Step 502, where the resource and startup subunit calculates the required number of cloud desktop resources based on real-time monitoring of task status and preset parameters. And step 503, where the resource and startup subunit dynamically matches cloud desktop resources with the queuing task requirements based on the number of desktop resources. This approach selects resources from the nearest available location to reduce transmission latency and improve the speed of sensitive services; it also optimizes network performance and enhances system stability and efficiency.

[0107] In one specific embodiment, when tasks cannot be queued, optimization is first achieved through fixed idle quantity management and execution rhythm and resource adjustment configuration fields.

[0108] Specifically, based on enterprise workload and cost-benefit analysis, a minimum and maximum number of idle desktops are both set to 5. Upon task execution or completion, devices are activated or deactivated based on the relationship between the number of idle desktops and the threshold. A cloud desktop retention time of 60 minutes is set to handle sudden task surges or business fluctuations. Furthermore, the number of idle cloud desktops is checked against task creation and completion as a rhythmic point to ensure it meets task requirements, and resources are adjusted accordingly.

[0109] In one specific embodiment, the variables within the framework parameters and algorithm modules include: RPA task execution time estimation (X), which uses historical task data, combined with factors such as task type, data volume, and operational complexity, to build a model using empirical formulas or machine learning algorithms to estimate the execution time X minutes for each RPA task, for rational resource allocation; the number of online idle cloud desktops (Y), where a distributed real-time monitoring system is established to periodically obtain the number of online idle cloud desktops by interacting with the cloud platform management API, providing a basis for resource allocation; order urgency and new cloud desktop launch (Z), where urgency is assigned to orders, and a monitoring module counts the number of queued orders and waiting time. When a queued order exceeds Z minutes and has a high urgency level, a new cloud desktop is launched; and cloud desktop idle shutdown time (D), where a parameter is set to close the cloud desktop after D minutes of idle time based on enterprise cost-effectiveness and business fluctuations. The cloud desktop records idle time during runtime, and calls the API to close it after D minutes, combining order prediction and task scheduling information to avoid accidental closure. The number of queued orders (O) is combined with parameters such as X, Y, Z, and D to determine whether to launch a new cloud desktop or shut down an idle cloud desktop through a dynamic calculation model.

[0110] In one specific embodiment, when tasks can be queued, configuration fields are first optimized. Specifically, when the average product runtime (X) is set to 5 minutes and the maximum queue waiting time (Z) is set to 2 minutes, historical task data is statistically analyzed to reflect task execution speed and acceptable waiting time. The cloud desktop retention time is set to 60 minutes. During the task creation and completion phases, based on the current number of queued orders O, the required number of cloud desktops to be launched is calculated using the formula [ROUNDUP((O*X) / Z)]-Y. If the number is greater than O, resources are launched.

[0111] In one specific embodiment, the improved cloud desktop shutdown policy includes increasing execution frequency and resource adjustment. Specifically, this includes a high-frequency execution mechanism, increasing the shutdown policy execution frequency to once every 20 seconds, and determining the number of cloud desktops to be launched based on a formula; if the number is less than or equal to 0, redundant idle cloud desktops are shut down. It also includes a shutdown condition based on idle time; that is, in addition to considering the number of cloud desktops to be launched, if each cloud desktop resource is idle for more than D minutes, the shutdown API interface is activated to shut down the cloud desktop.

[0112] In one specific embodiment, alternatives to the situation where tasks cannot be queued include a dynamic idle quantity threshold scheme, which does not fix the minimum and maximum idle quantities but dynamically adjusts them based on recent task fluctuations, and can dynamically adjust the cloud desktop retention time according to business peaks and troughs. It also includes device management based on resource utilization, which determines device creation and shutdown based on the overall cloud desktop resource utilization, and determines the utilization rate by calculating the ratio of resources occupied by the cloud desktop executing the task to the total resources.

[0113] In one specific embodiment, alternatives to the task queuing scenario include resource allocation based on a queuing theory model. This involves using a queuing theory model to analyze the relationship between task queuing and cloud desktop resources, calculating the optimal number of cloud desktops based on task arrival rates and the service rate of each cloud desktop, and dynamically adjusting resources according to changes in the queuing system status. It also includes a queuing strategy that combines priority and time slices. In addition to considering order urgency, a time slice is allocated to each task, and the queuing order is arranged according to priority and time slice. Resource allocation is determined at execution cadence points based on the priority of queued tasks, time slices, and the number of idle cloud desktops.

[0114] In one specific embodiment, alternatives to the cloud desktop shutdown strategy include an energy consumption model-based shutdown strategy. This involves establishing a cloud desktop energy consumption model, analyzing energy consumption in different states, and initiating a shutdown API interface when the energy consumption of idle cloud desktops falls below a threshold, thus balancing resource utilization and energy costs. Another approach includes a predictive shutdown strategy, which uses machine learning algorithms to predict future task distribution. Based on historical data and business trends, it predicts cloud desktop task allocation; if no tasks are predicted, the desktop is shut down, proactively managing resources.

[0115] In one specific embodiment, the alternative to the proximity principle for determining scalable cloud desktops includes region selection based on network performance monitoring. This involves deploying network performance monitoring points to collect network performance metrics between the cloud desktop and the main service platform, in addition to obtaining regional information through ping, and selecting the best-performing resource when launching a new cloud desktop. It also includes region selection combined with load balancing, which, in addition to regional factors, incorporates the current task load of cloud desktop resources in each region into the selection criteria, choosing cloud desktop resources from regions with lighter loads.

[0116] The various embodiments of this application have been described above. These descriptions are exemplary and not exhaustive, nor are they 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 improvement of the technology in the market, or to enable others skilled in the art to understand the embodiments disclosed herein.

Claims

1. A system for digital employee multi-channel cloud desktop extension, characterized in that, include: The module includes a programming interface module, an interface adaptation layer module, a scalable algorithm module, and a proximity principle judgment module. The programming interface module is used to dynamically manage access points of multiple cloud platforms, analyze the cloud desktop functions and interface specifications of each cloud platform, abstract common functions, and define a unified interface specification for basic operations. The interface adaptation layer module includes application programming interface calling logic and data conversion mechanism for each cloud platform, which is used to convert upper-layer application requests into native application programming interface calls and process the results. The scalable algorithm module is used to realize the elastic scaling of cloud desktop resources, dynamically allocate cloud desktop resources in the multi-channel cloud desktop extension system, and make the number of cloud desktops adaptively adjust according to business needs. The proximity principle judgment module is used to select the cloud desktop resource closest to the server's location based on the predefined application programming interface launch channel and region correspondence.

2. The system of claim 1, wherein, The scalable algorithm module includes a template frame unit, a frame algorithm unit, and a queuing unit. The template framework unit is used to analyze business processes and data characteristics for different top-level application scenarios, and to customize parameter templates covering configuration information for data collection, processing and storage for each top-level application. The parameter template is automatically loaded when executing robotic process automation tasks. The framework algorithm unit is used for the fine-grained allocation and management of cloud desktop resources in the scalable algorithm module, and multi-dimensional parameter analysis and algorithm model realize the dynamic matching of cloud desktop resources and business requirements. The queuing unit is used to handle scenarios where tasks can be queued in cloud desktop resource scheduling. Parameter settings and resource calculation logic enable dynamic matching between cloud desktop resources and queued task requirements.

3. The system of claim 2, wherein, The framework algorithm unit includes a task time estimation subunit and an idle cloud desktop detection subunit; The task time estimation sub-unit uses historical task data, combined with factors such as task type, data volume, and operational complexity, to build a model using empirical formulas or machine learning algorithms to estimate the execution time of each robotic process automation task. The online idle cloud desktop quantity monitoring subunit establishes a distributed real-time monitoring system, which interacts with the cloud platform management application programming interface to periodically obtain the online idle cloud desktop quantity.

4. The system of claim 3, wherein, The framework algorithm unit includes an order and new desktop subunit, a desktop idle management subunit, and an order utilization subunit; The order urgency and new cloud desktop startup subunit sets an urgency level identifier for each order, and sets a monitoring module to count the number of queued orders and waiting time. When the waiting time of a queued order exceeds the preset queuing time and the urgency level is high, the new cloud desktop is triggered. The cloud desktop idle management subunit sets a parameter to close the cloud desktop after the idle time exceeds the preset idle time based on enterprise cost-effectiveness and business fluctuations. The cloud desktop records the idle time during runtime, calls the application programming interface to close it, and combines order prediction and task scheduling information to avoid accidental closure. The order uses sub-units to update the number of queued orders in real time, and the dynamic calculation model determines whether to start a new cloud desktop or close an idle cloud desktop.

5. The system of claim 4, wherein, The queuing unit includes a parameter setting subunit and a resource and startup subunit; The parameter setting subunit pre-sets key parameters related to task processing and cloud desktop resource management, and provides a benchmark value basis for cloud desktop resource allocation in a task queuing scenario; The resource and starting subunit monitors the task state in real time and calculates the required cloud desktop resource quantity in combination with the pre-set parameters, to realize dynamic matching of cloud desktop resources and queuing task demand.

6. A method of digital employee multi-channel cloud desktop extension, characterized in that, The method is implemented using the system of any one of claims 1-5, and the method comprises the following steps: The programming interface module analyzes the cloud desktop functions and interface specifications of each cloud platform, abstracts common functions, and defines a basic operation unified interface specification; The interface adaptation layer module converts the upper-layer application request into a native application programming interface call and processes the result; The scalable algorithm module adaptively adjusts the number of cloud desktops according to business requirements; and The proximity principle judgment module starts the channel and region corresponding relationship, and selects the cloud desktop resource closest to the region where the server is located.

7. The method of claim 6, wherein, The scalable algorithm module processing step comprises: The template framework unit customizes a parameter template of configuration information for each top-level application, and automatically loads the parameter template when executing the robot process automation task; The framework algorithm unit performs cloud desktop resource allocation and management, so that the cloud desktop resources and business requirements are dynamically matched; The queuing waiting unit processes the scenario that the task can be queued and waited, so that the cloud desktop resources and queuing task demand are dynamically matched.

8. The method of claim 7, wherein, The framework algorithm unit processing step comprises: The task time estimation subunit constructs a model to estimate the execution time of each robot process automation task; The online idle cloud desktop quantity monitoring subunit interacts with the cloud platform management application programming interface to obtain the online idle cloud desktop quantity at regular intervals.

9. The method of claim 8, wherein, The framework algorithm unit processing step further comprises: The order urgency and new cloud desktop starting subunit sets an urgency identifier for the order and counts the number of queued orders and waiting time; When the waiting time of the queued order exceeds the pre-set queuing time and the urgency is high, a new cloud desktop is started; The cloud desktop idle management subunit closes the idle cloud desktop when the idle time of the cloud desktop exceeds the pre-set idle time; The cloud desktop idle management subunit records the idle time during the cloud desktop runtime, calls the application programming interface to close, and avoids false closing in combination with order prediction and task scheduling information.

10. The method of claim 9, wherein, The queuing waiting unit processing step comprises: The parameter setting subunit pre-sets key parameters related to task processing and cloud desktop resource management; The resource and starting subunit calculates the required cloud desktop resource quantity in combination with the real-time monitoring of the task state and the pre-set parameters; The resource and starting subunit dynamically matches the cloud desktop resources and queuing task demand according to the desktop resource quantity.

Images