SOC (System On Chip) service quality guarantee system and method based on Actor-Critic deep reinforcement learning

Abstract

The invention discloses an SOC service quality guarantee system based on Actor-Critic deep reinforcement learning. The SOC service quality guarantee system comprises a user task generator, a plurality of resource pools, a plurality of resource pool task optimal distribution modules, a service task scheduler, a service load evaluation module and a QoE evaluation module. The method comprises the following steps: inputting task lists of all users, resource pool occupation conditions, service load of each user and a current idle resource pool number, and running an Actor-Critic deep reinforcement learning algorithm by taking QoE as an effect evaluation feedback basis to obtain an allocation scheme of a to-be-executed task of the user on a resource pool at the next moment; and according to the task allocation scheme, completing task scheduling through a task scheduling interface of the SOC platform, assigning corresponding resources to execute specific tasks of specific users, searching an optimal resource allocation scheme for the tasks listed in all user task lists, and arranging the tasks of the users to complete services for the resources in the resource pool according to the optimal sequence, so that the subjective quality experience QoE of the user is maximized.

Description

technical field [0001] The invention relates to the technical field of electrical digital data processing, in particular to an SOC service quality assurance system and method based on Actor-Critic deep reinforcement learning. Background technique [0002] Security Operations Center (SOC) provides users with security services through cloud security resources, reduces the cost of security services through a large group model, and makes security services easy to obtain. The advantages of this cloud service model are reflected in the effective scheduling of cloud shared service resources. For users, cloud resources are statistically shared, so the available service quality cannot be completely determined, but in a certain fluctuate within the range. Therefore, for the SOC security operation mode, how to provide users with a pre-agreed service quality (SLA, Service Level Agreement) under the limitation of limited cloud resources has become one of the key technologies. [0003] ...

AI Technical Summary

Problems solved by technology

[0007] 2. Users are grouped by Quality of Service (QoS), and the cloud platform reserves and schedules resources according to objective indicators of service quality QoS, such as service response time and service completion time. Although these objective QoS indicators are different from user It is related to the subjective feelings of QoS, but there are still differences, and this difference will further amplify the waste of resources caused by scheduling resources based on purely objective QoS indicators

[0012] Finally, the existing algorithms directly use the collected objective indicators as the optimization target, which cannot reflect the subjective feelings of users on service quality.

Images