O-RAN DU Parallel Instance Switchover for Continuous Operation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing cellular networks face challenges in maintaining reliable and continuous content transmission, particularly with the increased use of smartphones and IoT devices, as they struggle to efficiently manage parallel software instances in Open Radio Access Networks (O-RAN) with disaggregated hardware and software, leading to potential network failures and downtime.
Innovation Solution
Implementing parallel software instances by initiating and running concurrent instances of the Distributed Unit (DU) on distinct portions of core processors, allowing for seamless swapping between active instances, thereby minimizing downtime and enhancing network resilience.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If parallel software instances are implemented on distinct portions of core processors, then network resilience and continuous operation are improved, but system complexity and resource management difficulty increase
Solution Approach 1:
The system divides the core processor resources into distinct portions, with each software instance running on a separate portion. This segmentation allows independent operation of each instance, ensuring that if one instance fails, the other continues to operate, thereby improving network resilience while managing complexity through structured resource division.
Solution Approach 2:
The system utilizes virtualization parameters to create isolated execution environments for each software instance. By changing the operational parameters of the core processors through virtualization, the system can run multiple instances simultaneously on distinct portions of resources, maintaining reliability while the virtualization layer abstracts the underlying complexity.
2Loss of time
If software instances are swapped between distinct portions of core processors, then downtime is reduced, but switching complexity and potential failure points increase
Solution Approach 1:
The system prepares backup software instances in advance on distinct portions of core processors, keeping them ready to immediately take over if the active instance fails. This preliminary preparation eliminates downtime by ensuring a pre-configured standby instance is available, while the switching mechanism manages complexity through automated failover procedures.
Solution Approach 2:
The system creates duplicate copies of software instances running on distinct portions of core processors. These copies serve as immediate replacements, eliminating downtime by allowing instant switching. The copying approach manages switching complexity by maintaining identical operational states that can be seamlessly transferred between instances.
3Adaptability or versatility
If multiple vendor versions and technologies are supported in parallel, then adaptability and flexibility are improved, but compatibility management and testing complexity increase
Solution Approach 1:
The system is designed to universally support multiple vendor versions and technologies by running them as parallel software instances on distinct portions of core processors. Each instance can operate independently with its specific vendor requirements, providing adaptability and flexibility. The universality is achieved through a standardized virtualization interface that manages the diversity of vendor implementations.
Solution Approach 2:
The system introduces a virtualization intermediary layer that sits between the diverse vendor software instances and the underlying core processors. This intermediary manages compatibility by translating between different vendor interfaces and the standardized hardware interface, reducing testing complexity by providing a consistent abstraction layer that isolates vendor-specific variations.
Data Source
AI summary
Systems and methods for implementing parallel software instances in Open Radio Access Network (O-RAN) with disaggregated hardware and software in a cellular network are disclosed. One such method includes: initiating a first instance of a DU of a RAN on a first portion of a plurality of core processors in the cellular network; initiating a second instance of the DU of the RAN on a second portion of the plurality of core processors in the cellular network, wherein the first portion of the plurality of core processors is distinct from the second portion of the plurality of core processors; running the first instance of the DU of the RAN concurrently with the second instance of the DU of the RAN in the cellular network; and swapping the active instance from the first instance of the DU of the RAN with the second instance of the DU of the RAN.