vRAN Resource Pooling for Dynamic DU Compute Allocation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional vRAN deployments face inefficiencies in computing resource utilization due to static binding and underutilization of resources, leading to increased costs and limited scalability, with proprietary hardware causing redundancies and network downtime.
Innovation Solution
A disaggregated architecture that dynamically allocates and reconfigures computing resources using an L2 controller, enabling flexible resource pooling and real-time optimization, eliminating the need for proprietary hardware and reducing infrastructure costs.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If computing resources are statically bound in conventional vRAN architectures, then resource allocation is simple and stable, but resource utilization is low and scaling is restricted
Solution Approach 1:
The patent implements dynamic resource allocation by transitioning from static binding to a pool-based architecture where computing resources can be dynamically assigned and reassigned based on demand. The L2 controller monitors resource usage and dynamically allocates computing resources between L1 and L2 functions, enabling flexible scaling and improved utilization without permanent dedications.
Solution Approach 2:
The patent creates a universal resource pool where computing resources can serve multiple functions and multiple layers (L1 and L2). Instead of dedicated resources for specific functions, the same pool of computing resources can be allocated to different functions as needed, making the system more versatile and efficient in resource utilization.
2Adaptability or versatility
If computing resources are statically bound, then system stability is maintained, but on-demand scaling is restricted
Solution Approach 1:
The system maintains stability through dynamic monitoring and controlled allocation. The L2 controller continuously monitors resource usage and system state, making incremental adjustments to resource allocation rather than abrupt changes, thus maintaining system stability while enabling flexible scaling responses to demand changes.
Solution Approach 2:
The patent implements feedback mechanisms where the L2 controller monitors resource usage, performance metrics, and system state, then uses this information to dynamically adjust resource allocation. This closed-loop control ensures system stability is maintained while enabling on-demand scaling based on actual conditions.
3Productivity
If proprietary hardware is used for L1 accelerators, then performance is optimized, but visibility between computing resources is limited and costs increase
Solution Approach 1:
The L2 controller serves as an intermediary that provides visibility and coordination between computing resources. It monitors and manages the pool of computing resources, enabling information sharing and coordination between L1 and L2 functions without requiring direct hardware coupling, thus improving visibility while maintaining performance.
Solution Approach 2:
The patent uses universal computing resources in the pool that can function as both L1 accelerators and L2 processing resources. This eliminates the need for proprietary dedicated hardware and enables full visibility and coordination through the unified resource pool managed by the L2 controller.
4Productivity
If computing resources are statically allocated, then infrastructure costs are predictable, but resource underutilization increases
Solution Approach 1:
The patent merges L1 and L2 computing resources into a single unified pool, eliminating the need for separate dedicated hardware for each function. This consolidation improves resource utilization efficiency by allowing dynamic sharing of the same physical resources between different functions while the L2 controller manages the combined pool.
Data Source
AI summary
Configurations of a system and a method for optimizing an allocation of computing resources via a disaggregated architecture, are described. In one aspect, the disaggregated architecture may include a Layer 2 (L2) controller that may be configured to optimize an allocation of computing resources in a virtualized radio access network (vRAN). The disaggregated architecture in a distributed unit may disaggregate an execution of the operations of the distributed unit by the computing resources deployed therein. Further, the disaggregated architecture may provision statistical multiplexing and provision a mechanism for allocating the computing resources based on real-time conditions in the network. The disaggregated architecture may provision a mechanism that may enable dynamic swapping, allocation, scaling up, management, and maintenance of the computing resources deployed in the distributed unit (DU).