CN-121984972-A - Multi-device cooperative scheduling method and device, electronic device and storage medium

CN121984972ACN 121984972 ACN121984972 ACN 121984972ACN-121984972-A

Abstract

The application discloses a multi-device collaborative scheduling method, a device, electronic equipment and a storage medium, and relates to the technical field of robot process automation, wherein the multi-device collaborative scheduling method comprises the steps of obtaining parameter indexes corresponding to each device, wherein the parameter indexes at least comprise central processing unit utilization rate, storage input and output speed, memory residual capacity and network bandwidth of the corresponding device, determining a calculation force value corresponding to the device based on the central processing unit utilization rate, the storage input and output speed, the memory residual capacity and the network bandwidth of the corresponding device, obtaining a calculation force demand threshold corresponding to a subtask to be allocated, obtaining the device corresponding to the subtask based on the calculation force demand threshold corresponding to the subtask and the calculation force value corresponding to each device, and allocating the subtask to the target device.

Inventors

LIAO WANLI
JIN ZHUO
YANG JIAN
LI ZHENDE

Assignees

珠海金智维人工智能股份有限公司

Dates

Publication Date: 20260505
Application Date: 20251231

Claims (10)

1. The multi-device cooperative scheduling method is characterized by being applied to a scheduling system, wherein the scheduling system comprises at least two devices; The method comprises the following steps: acquiring a parameter index corresponding to each device, wherein the parameter index at least comprises the central processing unit utilization rate, the storage input and output speed, the memory residual capacity and the network bandwidth of the corresponding device; Determining a calculation force value corresponding to the equipment based on the utilization rate of a central processing unit corresponding to the equipment, the storage input and output speed, the memory residual capacity and the network bandwidth; acquiring a calculation force demand threshold corresponding to a subtask to be distributed; Obtaining equipment corresponding to the subtask based on the calculated force demand threshold corresponding to the subtask and the calculated force value corresponding to each piece of equipment, and taking the equipment as target equipment; And distributing the subtasks to the target equipment.
2. The multi-device collaborative scheduling method according to claim 1, wherein the obtaining the device corresponding to the subtask as the target device based on the calculated force requirement threshold corresponding to the subtask and the calculated force value corresponding to each device includes: Screening out devices with corresponding calculation force values larger than or equal to the subtask corresponding calculation force demand threshold value in the devices to serve as candidate devices; respectively differencing the calculated force values of the candidate devices with calculated force demand thresholds corresponding to the subtasks to obtain calculated force redundancy values; And sequencing the calculated force redundancy values, and taking the candidate device corresponding to the maximum calculated force redundancy value as the target device.
3. The multi-device collaborative scheduling method according to claim 1, wherein the determining a calculated power value corresponding to the device based on a central processing unit utilization, a storage input output speed, a memory remaining capacity, and a network bandwidth corresponding to the device comprises: acquiring a preset weighting coefficient; and generating a calculation force value corresponding to the equipment through a weighting algorithm based on the weighting coefficient and the utilization rate of the central processing unit corresponding to the equipment, the storage input and output speed, the memory residual capacity and the network bandwidth.
4. The multi-device co-scheduling method of claim 1, comprising, after the assigning the sub-task to the target device: monitoring each device; And if the equipment with the variation amplitude of the corresponding calculated force value exceeding the preset threshold exists in the equipment, executing reassignment operation on the subtasks assigned by the equipment with the variation amplitude of the corresponding calculated force value exceeding the threshold.
5. The multi-device collaborative scheduling method according to claim 1, further comprising, prior to the obtaining the calculated force demand threshold corresponding to the sub-task to be allocated: acquiring a preset main task flow; and disassembling the main task flow to obtain a plurality of subtasks to be distributed.
6. The multi-device collaborative scheduling method according to claim 5, further comprising, after said assigning the subtasks to the target devices: if a first task and a second task exist in the plurality of subtasks, and the execution of the second task depends on the output data of the first task, an encryption channel is constructed between target equipment corresponding to the first task and target equipment corresponding to the second task; and synchronizing the output data of the first task to the target equipment corresponding to the second task through the encryption channel.
7. The multi-device collaborative scheduling method according to claim 1, further comprising, before the obtaining the parameter index corresponding to each device: acquiring a preset standard data format; And converting the format of the data to be output of each device into the standard data format.
8. A multi-device cooperative scheduling apparatus, characterized by being applied to a scheduling system, the scheduling system comprising at least two devices; The device comprises: The computing power evaluation module is configured to acquire parameter indexes corresponding to each device, wherein the parameter indexes at least comprise the central processing unit utilization rate, the storage input and output speed, the memory residual capacity and the network bandwidth of the corresponding device; The task splitting module is configured to acquire a calculation force demand threshold corresponding to a subtask to be distributed, acquire equipment corresponding to the subtask based on the calculation force demand threshold corresponding to the subtask and the calculation force value corresponding to each piece of equipment, and distribute the subtask to the target equipment.
9. An electronic device comprising a memory storing a computer program and a processor that when executing the computer program implements the multi-device co-scheduling method of any one of claims 1 to 7.
10. A computer readable storage medium storing a computer program, wherein the computer program when executed by a processor implements the multi-device co-scheduling method of any one of claims 1 to 7.

Description

Multi-device cooperative scheduling method and device, electronic device and storage medium Technical Field The present application relates to the field of robot process automation, and in particular, to a method and apparatus for collaborative scheduling of multiple devices, an electronic device, and a storage medium. Background Along with the penetration of digital transformation, RPA (robot process automation) is widely applied to the scenes of office automation, industrial control, data processing of the Internet of things and the like, but the current business scene is more complex, the calculation power and storage resources of a single device are limited, and the large-scale and high-concurrency RPA task demands are difficult to meet, meanwhile, the popularization of multi-terminal devices promotes the demand of cooperatively executing the RPA task across devices, and in the prior art, the situation that the calculation power of the devices is not matched is easily caused by the cooperation of multiple devices, so that the utilization rate of the final calculation power is low and the task execution rate is poor. Disclosure of Invention The present application aims to solve at least one of the technical problems existing in the prior art. Therefore, the application provides a multi-device collaborative scheduling method, a device, electronic equipment and a storage medium, which can realize reasonable distribution of subtasks based on the calculation force value corresponding to the equipment and improve the utilization rate of the calculation force of the equipment. The multi-device cooperative scheduling method of the embodiment of the first aspect of the application is applied to a scheduling system, and comprises the following steps: acquiring a parameter index corresponding to each device, wherein the parameter index at least comprises the central processing unit utilization rate, the storage input and output speed, the memory residual capacity and the network bandwidth of the corresponding device; Determining a calculation force value corresponding to the equipment based on the utilization rate of a central processing unit corresponding to the equipment, the storage input and output speed, the memory residual capacity and the network bandwidth; acquiring a calculation force demand threshold corresponding to a subtask to be distributed; Obtaining equipment corresponding to the subtask based on the calculated force demand threshold corresponding to the subtask and the calculated force value corresponding to each piece of equipment, and taking the equipment as target equipment; And distributing the subtasks to the target equipment. The multi-device collaborative scheduling method at least has the advantages that the computing force value corresponding to each device is calculated based on the acquired parameter index corresponding to each device, so that the situation that the devices are in a full-load state and the devices are in an idle state is intuitively reflected, the follow-up subtasks are distributed, the target device corresponding to the subtasks is calculated based on the computing force requirement threshold value of the subtasks and the computing force value of each device, the subtasks are distributed to the target device, the matching of the subtasks and the computing force of the devices is realized, the situation that the computing force value of the devices is insufficient to execute the subtasks or the situation that the computing force value of the devices is redundant is excessive is avoided, the utilization rate of the computing force of each device is improved, and the execution rate of the subtasks is improved. According to some embodiments of the present application, the obtaining, based on the computational power demand threshold corresponding to the subtask and the computational power value corresponding to each device, the device corresponding to the subtask as a target device includes: Screening out devices with corresponding calculation force values larger than or equal to the subtask corresponding calculation force demand threshold value in the devices to serve as candidate devices; respectively differencing the calculated force values of the candidate devices with calculated force demand thresholds corresponding to the subtasks to obtain calculated force redundancy values; And sequencing the calculated force redundancy values, and taking the candidate device corresponding to the maximum calculated force redundancy value as the target device. According to some embodiments of the present application, the determining the computing power value corresponding to the device based on the central processing unit utilization, the storage input/output speed, the memory remaining capacity and the network bandwidth corresponding to the device includes: acquiring a preset weighting coefficient; and generating a calculation force value corresponding to the equipment through a wei