A Q-learning-based energy-efficient task-driven dynamic on-chip interconnect communication method
By adopting the Q-learning reinforcement learning algorithm in the on-chip network, dynamically dividing the sub-network area and selecting the appropriate topology structure and routing algorithm, the problems of low resource utilization and performance-energy imbalance are solved, and highly energy-efficient task execution is achieved.
CN119254686BActive Publication Date: 2025-09-12BEIJING UNIV OF TECH
2 Cites 0 Cited by
Patent Information
- Application Number
- CN202411409537.3
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2024-10-10
- Publication Date
- 2025-09-12
- Estimated Expiration
- 2044-10-10
AI Technical Summary
Technical Problem
Existing on-chip networks have low resource utilization and lack of targeted task topology selection, resulting in an imbalance between performance and energy consumption.
Method used
A Q-learning-based reinforcement learning algorithm is used to dynamically divide the network area on the sub-chip, select the appropriate topology and routing algorithm, and flexibly adjust the number of virtual channels to achieve a balance between performance and power consumption.
Benefits of technology
It improves the resource utilization of on-chip networks, reduces resource waste, and optimizes task execution efficiency and energy efficiency.
✦ Generated by Eureka AI based on patent content.
Abstract
A high-energy-efficiency task-driven dynamic on-chip interconnect communication method based on Q-learning belongs to the field of on-chip interconnection of computer systems. The present invention adopts an adaptive topology structure to map tasks, and divides different areas for task processing according to the characteristics of the tasks. This method can allocate specific on-chip resources and topology structure types to each task, thereby saving on-chip resource consumption. Through the adaptive topology structure, the system can flexibly adjust the connection relationship and routing method in the on-chip network according to the needs and characteristics of the task. At the same time, by dividing different areas, the processing processes of different tasks can be isolated, minimizing the occurrence of interference and conflicts. Using the Q-learning reinforcement learning algorithm to dynamically change the transmission direction and the number of virtual channels of the routing algorithm with the goal of low latency and low power consumption during task operation can effectively achieve a balance between performance and power consumption in the on-chip network.
Need to check novelty before this filing date? Find Prior Art