Hybrid vRAN Scheduling for Real-Time Compute Sharing
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing telecommunications networks face inefficiencies and high costs due to overprovisioning of computing resources for virtualized Radio Access Networks (vRANs), leading to significant resource wastage during non-peak periods, and existing scheduling systems struggle to share resources between vRANs from multiple vendors without requiring vendor cooperation or knowledge of specific implementations.
Innovation Solution
A hybrid scheduling system that designates real-time cores (R-cores) and shared cores (S-cores) within virtual machines, infers traffic-based key performance indicators without vendor cooperation, and generates scheduling instructions to optimize resource allocation across physical cores, ensuring real-time constraints are met.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If compute resources are overprovisioned to meet peak capacity demands, then reliability of real-time processing is improved, but resource utilization efficiency deteriorates
Solution Approach 1:
The system dynamically adjusts compute resource allocation between vRAN workloads and other workloads based on real-time demand. The hybrid scheduler continuously monitors workload characteristics and modifies resource distribution, transitioning from static overprovisioning to adaptive resource sharing that maintains reliability while improving utilization efficiency.
Solution Approach 2:
Instead of allocating all compute resources to vRAN workloads (excessive action), the system allocates only the necessary portion during non-peak periods to other workloads. The scheduler provides partial resource allocation to vRAN functions when full capacity is not needed, reducing waste while maintaining service requirements.
2Productivity
If compute resources are shared between vRAN and other workloads, then resource utilization efficiency is improved, but complexity of scheduling system deteriorates
Solution Approach 1:
The system segments compute resources into dedicated portions for vRAN workloads and shareable portions for other workloads. The hybrid scheduler separates real-time critical tasks from best-effort tasks, applying different scheduling policies to each segment. This segmentation simplifies the scheduling complexity by creating clear boundaries and priority levels.
Solution Approach 2:
The hybrid scheduler acts as an intermediary layer between the physical compute resources and the virtualized workloads. It translates resource allocation decisions into scheduling instructions that the underlying operating system can execute, managing the complexity of multi-vendor, multi-workload environments through a standardized interface.
3Measurement precision
If vendor-specific integration is implemented for resource scheduling, then scheduling precision is improved, but adaptability to multi-vendor environments deteriorates
Solution Approach 1:
The hybrid scheduler implements a universal scheduling framework that can handle workloads from multiple vendors without requiring vendor-specific integration. It uses standardized interfaces and generic scheduling policies that work across different vRAN implementations, maintaining adaptability while achieving effective resource allocation through observed workload characteristics.
Solution Approach 2:
The system changes scheduling parameters dynamically based on observed workload behavior rather than relying on fixed vendor-specific parameters. By monitoring actual workload characteristics and adjusting scheduling decisions in real-time, the system achieves precision comparable to vendor-specific solutions while maintaining broad adaptability to different vendors and workload types.
Data Source
AI summary
The present disclosure relates to systems, methods, and computer-readable media for implementing a hybrid scheduler for vRAN compute sharing. The systems described herein involve a hybrid scheduling system that considers KPIs and telemetry data associated with usage of vCPUs on a vRAN VM, container, or other service construct and determines estimated runtimes for tasks of workloads running on the vCPUs. The hybrid scheduling system may generate scheduling instructions to be used by an operating system on the server device to schedule allocation of computing resources to any number of vCPUs hosted by the server device. The hybrid scheduling system provides features that enables optimization of not only physical layer processing tasks, but a holistic approach that involves optimizing scheduling of tasks associated with multiple processing layers of VMs, and particular vRAN workload VMs.