Distributed Unit Layer 2 Scheduler for Low-Activity Core Power Management
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing power management mechanisms for computing networks, particularly in 5G-NR Distributed Units (DUs), are ineffective during short inactivity periods due to latency impacts and downtime, and lack application-level coordination for optimal power savings.
Innovation Solution
A Layer 2 scheduler is enhanced to manage power consumption of Layer 1 and Layer 2 cores in DUs, utilizing statistical and machine learning models to determine power management profiles and frequency assignments based on core processor categories and load patterns, enabling coordinated power control across different core groups.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Use of energy by moving object
If existing power management mechanisms are applied during extended low activity periods, then power consumption is reduced, but they cannot be applied during short inactivity periods due to latency impacts and downtime
Solution Approach 1:
The patent implements dynamic power management by enabling cores to transition between active and sleep states based on real-time processing load conditions. The layer 2 scheduler dynamically selects cores for power management and controls their power states, allowing the system to adapt power consumption levels to actual workload requirements rather than using static power management configurations.
Solution Approach 2:
The patent applies preliminary action by having the layer 2 scheduler proactively select cores for power management before they are needed, and by provisioning power management profiles in advance. The scheduler identifies suitable cores and prepares power management configurations based on predicted or anticipated load patterns, allowing smooth transitions to sleep states without causing latency when traffic arrives.
2Use of energy by moving object
If power management is implemented at the layer 2 scheduler level with core selection based on processing load, then power consumption is reduced during intermittent low activity, but system complexity increases
Solution Approach 1:
The layer 2 scheduler is enhanced to perform multiple functions: traditional scheduling tasks plus power management functions. By integrating core selection for power management and power state control into the existing layer 2 scheduler, the patent avoids adding separate dedicated power management hardware or software modules, thereby reducing overall system complexity while achieving effective power savings.
3Use of energy by moving object
If statistical and machine learning models are used to determine power management profiles, then power consumption control is optimized, but computational overhead and processing time increase
Solution Approach 1:
The patent applies partial action by using statistical and machine learning models selectively rather than continuously. The models are used to determine power management profiles based on observed load patterns and historical data, but not for every single scheduling decision. This selective application optimizes power consumption control while avoiding excessive computational overhead that would occur with continuous model evaluation.
Data Source
AI summary
Facilitating a scheduler for distributed unit power management in advanced communication networks is provided herein. A method includes, based on a processing load of a group of core processors being determined to be below a threshold traffic amount for a first transmission time slot, selecting, by network equipment comprising a processor, a core processor of the group of core processors for power management, resulting in an identified core processor. Selection of the core processor can be based on a configuration of the identified core processor. The method also includes, based on a category assigned to the identified core processor, controlling, by the network equipment, a power consumption of the identified core processor during a second transmission time slot. The network equipment comprises a distributed unit, and controlling the power consumption is performed in a layer 2 scheduler of the distributed unit.