CN-122019132-A - Resource scheduling method, device, equipment and program product

CN122019132ACN 122019132 ACN122019132 ACN 122019132ACN-122019132-A

Abstract

The application discloses a resource scheduling method, a device, equipment and a program product. The method comprises the steps of obtaining multisource analysis data of resources to be scheduled in a service system and a preset scheduling target, wherein the scheduling target comprises a global service target and at least two local service targets. And determining a plurality of candidate scheduling strategies aiming at the resources to be scheduled. And calling a large model, and based on the multi-source analysis data, evaluating a first type influence rate of each candidate scheduling strategy aiming at the global business target and a second type influence rate of each candidate scheduling strategy aiming at each local business target. And taking the global service target as a constraint and maximizing the comprehensive local value as a target, and screening target scheduling strategies from the plurality of candidate scheduling strategies, wherein the comprehensive local value corresponding to each candidate scheduling strategy is determined based on the first influence rate and the second influence rate corresponding to the candidate scheduling strategy. And executing the target scheduling strategy.

Inventors

WANG JING
WANG ZHAOYI
XU YU
LIU FANG
ZHANG YUTAO
LI XIANGLI
YANG ZHEN

Assignees

中移在线服务有限公司
中国移动通信集团有限公司

Dates

Publication Date: 20260512
Application Date: 20251224

Claims (10)

1. A method for scheduling resources, comprising: The method comprises the steps of obtaining multisource analysis data of resources to be scheduled in a service system and a preset scheduling target, wherein the scheduling target comprises a global service target and at least two local service targets; determining a plurality of candidate scheduling strategies aiming at the resources to be scheduled; invoking a large model, and based on the multi-source analysis data, evaluating a first class influence rate of each candidate scheduling strategy for the global service target and a second class influence rate of each candidate scheduling strategy for each local service target; Selecting a target scheduling strategy from the plurality of candidate scheduling strategies by taking the global service target as a constraint and maximizing the comprehensive local value as a target, wherein the comprehensive local value corresponding to each candidate scheduling strategy is determined based on a first influence rate and a second influence rate corresponding to the candidate scheduling strategy; And executing the target scheduling strategy.
2. The method of claim 1, wherein the step of determining the position of the substrate comprises, The comprehensive local value corresponding to each candidate scheduling strategy is determined based on all second class influence rates corresponding to the candidate scheduling strategy and the weight of each local business target, wherein the weight of the local business target is determined according to the business priority and the achievement urgency of the local business target.
3. The method of claim 2, wherein the step of determining the position of the substrate comprises, The achievement urgency is determined based on a degree of deviation between a current index value of the corresponding local business objective and an expected index value.
4. The method of claim 1, wherein the step of determining the position of the substrate comprises, Determining a plurality of candidate scheduling strategies for the resources to be scheduled, including: Acquiring the current available state and specification parameters of the resources to be scheduled; And determining a plurality of candidate scheduling strategies based on the current available state and specification parameters of the resources to be scheduled and combining the executable scheduling action types of the resources to be scheduled, wherein the candidate scheduling strategies are used for adjusting or reallocating the resources to be scheduled.
5. The method of claim 1, wherein the step of determining the position of the substrate comprises, Determining a plurality of candidate scheduling strategies for the resources to be scheduled, including: monitoring key indexes associated with at least one local business target in real time; and when any key index is monitored to reach a preset trigger threshold, determining a plurality of candidate scheduling strategies aiming at the resources to be scheduled.
6. The method of claim 5, wherein the step of determining the position of the probe is performed, The resources to be scheduled comprise telephone traffic resources, the multi-source analysis data comprise real-time call completing rate, queuing quantity, agent state and historical telephone traffic rule data, the global service target is that the call completing rate reaches the standard, the local service target comprises at least two of customer satisfaction, agent time load and resource consumption cost, and the key index comprises at least one of central processing unit utilization rate, memory utilization rate, interface calling success rate, service response time, queue waiting quantity and real-time call completing rate.
7. The method according to any one of claim 1 to 6, wherein, The multi-source analysis data is used to present at least one of a real-time status, a historical rule, and a demand trend for the resource to be scheduled.
8. A resource scheduling apparatus, comprising: The system comprises a data acquisition module, a scheduling module and a scheduling module, wherein the data acquisition module is used for acquiring multisource analysis data of resources to be scheduled in a service system and a preset scheduling target, and the scheduling target comprises a global service target and at least two local service targets; the strategy determining module is used for determining a plurality of candidate scheduling strategies aiming at the resources to be scheduled; The model calling module is used for calling a large model, and evaluating the influence rate of each candidate scheduling strategy on the global business target and the influence rate of each candidate scheduling strategy on each local business target based on the multi-source analysis data; The strategy optimization module is used for taking the global business target as a constraint, taking the maximized comprehensive local value as a target, and screening target scheduling strategies from the plurality of candidate scheduling strategies, wherein the comprehensive local value corresponding to each candidate scheduling strategy is determined based on the influence rate of the candidate scheduling strategy on the global business target and the influence rate of the candidate scheduling strategy on each local business target; And the policy execution module is used for executing the target scheduling policy.
9. An electronic device comprising a processor and a memory arranged to store computer executable instructions, wherein the executable instructions when executed cause the processor to perform the method of any of claims 1to 7.
10. A computer program product comprising a computer readable storage medium storing a computer program, characterized in that the computer program is operable to cause a computer to perform the method of any one of claims 1 to 7.

Description

Resource scheduling method, device, equipment and program product Technical Field The present application relates to the field of resource scheduling technologies, and in particular, to a resource scheduling method, apparatus, device, and program product. Background In the fields of infrastructure and IT production, resource scheduling is a key link for guaranteeing efficient and stable operation of a system, and particularly in the scenes of cloud computing, data centers, network communication and the like, the multi-element coordination of server computing power, storage equipment, virtual machines, network bandwidth and even human resources is involved. These scenarios typically require that multiple interrelated and even conflicting objectives be met simultaneously, such as having to ensure quality of service (QoS) of critical traffic, tightly control operational costs, and maintain load balancing among nodes while promoting overall resource utilization. However, the conventional scheduling method, whether it is based on fixed rule polling, minimum connection number algorithm or queue management introducing weighted priority, has difficulty in achieving dynamic and real-time efficient trade-off and collaborative optimization among multiple targets. Disclosure of Invention The application provides a method, a device, equipment and a program product, which aim to solve the problem that multi-objective collaborative optimization is difficult in the existing resource scheduling technology. Correspondingly, the technical scheme of the application is as follows: in a first aspect, a method for scheduling resources is provided, including: The method comprises the steps of obtaining multisource analysis data of resources to be scheduled in a service system and a preset scheduling target, wherein the scheduling target comprises a global service target and at least two local service targets; determining a plurality of candidate scheduling strategies aiming at the resources to be scheduled; invoking a large model, and based on the multi-source analysis data, evaluating a first class influence rate of each candidate scheduling strategy for the global service target and a second class influence rate of each candidate scheduling strategy for each local service target; Selecting a target scheduling strategy from the plurality of candidate scheduling strategies by taking the global service target as a constraint and maximizing the comprehensive local value as a target, wherein the comprehensive local value corresponding to each candidate scheduling strategy is determined based on a first influence rate and a second influence rate corresponding to the candidate scheduling strategy; And executing the target scheduling strategy. In a second aspect, there is provided a resource scheduling apparatus, including: The system comprises a data acquisition module, a scheduling module and a scheduling module, wherein the data acquisition module is used for acquiring multisource analysis data of resources to be scheduled in a service system and a preset scheduling target, and the scheduling target comprises a global service target and at least two local service targets; the strategy determining module is used for determining a plurality of candidate scheduling strategies aiming at the resources to be scheduled; The model calling module is used for calling a large model, and evaluating the influence rate of each candidate scheduling strategy on the global business target and the influence rate of each candidate scheduling strategy on each local business target based on the multi-source analysis data; The strategy optimization module is used for taking the global business target as a constraint, taking the maximized comprehensive local value as a target, and screening target scheduling strategies from the plurality of candidate scheduling strategies, wherein the comprehensive local value corresponding to each candidate scheduling strategy is determined based on the influence rate of the candidate scheduling strategy on the global business target and the influence rate of the candidate scheduling strategy on each local business target; And the policy execution module is used for executing the target scheduling policy. In a third aspect, an embodiment of the application provides an electronic device comprising a processor and a memory configured to store computer-executable instructions that, when executed, cause the processor to perform the method of the first aspect. In a fourth aspect, there is provided a computer program product comprising a computer readable storage medium storing a computer program operable to cause a computer to perform the method of the first aspect. The embodiment of the application provides a resource scheduling scheme, which can quantitatively evaluate the influence of a scheduling strategy on a multi-level business target through a large model and realize comprehensive optimization of local value under glob