Session Border Controller Resource Allocation in Cloud Clusters
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In cloud-based distributed environments, efficiently managing resources such as call limits, bandwidth, and transcoders across multiple nodes in a telecom cluster is challenging due to the distributed nature of cloud systems, leading to issues like call rejections and increased messaging overhead, especially under non-uniform traffic conditions.
Innovation Solution
A resource management system where an individual session border controller reports utilization to a master resource manager, which determines operation modes (normal, critical, or overshoot) and adjusts resource allocation accordingly, allowing for efficient resource distribution without significant delay or increased messaging, even under varying traffic loads.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If a central entity is deployed to keep track of resource usage count across the cloud, then resource management accuracy is improved, but call processing delay increases significantly
Solution Approach 1:
The patent divides the resource management function into segments distributed across multiple SBCs. Each SBC maintains local resource state and performs local call admission control, eliminating the need for a central entity. This segmentation reduces call processing delay while maintaining resource management accuracy through local knowledge of resource usage.
Solution Approach 2:
Each SBC is enabled to perform self-managed resource tracking and call admission control decisions. The SBCs autonomously monitor their own resource usage and make local decisions about call acceptance, eliminating dependency on a central entity for real-time decisions, thus reducing processing delay while maintaining accurate resource management.
2Measurement precision
If entities synch their usage to one another in a frequent manner, then resource management accuracy is improved, but intra cloud messaging overhead increases significantly
Solution Approach 1:
Each SBC maintains continuous local tracking of resource usage state without requiring frequent synchronization messages. The SBCs proactively manage their own resource state information, eliminating the need for frequent inter-SBC messaging to update usage information, thus reducing messaging overhead while maintaining accurate resource management.
Solution Approach 2:
Each SBC independently tracks and manages its own resource usage state without requiring frequent synchronization with other SBCs. This self-service approach allows each SBC to maintain accurate local resource state information while eliminating the need for frequent intra-cloud messaging, thus reducing messaging overhead.
3Ease of operation
If statistical resource distribution is used across the cloud, then resource allocation simplicity is improved, but call rejection rate increases under non-uniform traffic
Solution Approach 1:
The patent implements local resource allocation decisions at each SBC based on its specific traffic conditions and resource availability. Each SBC makes call admission decisions locally considering its own traffic patterns and resource state, rather than relying on statistical distribution. This local quality approach maintains allocation simplicity while reducing call rejections under non-uniform traffic by adapting to local conditions.
Solution Approach 2:
The system dynamically adjusts resource allocation decisions at each SBC based on real-time local traffic conditions and resource availability. This dynamic local decision-making allows the system to adapt to non-uniform traffic patterns, reducing call rejections while maintaining the simplicity of distributed resource allocation without requiring complex centralized control.
Data Source
AI summary
An individual session border controller in a cluster of SBCs, e.g., in a Cloud, sends a resource utilization reporting message to the current master resource manager for the cluster. In response to the received utilization reporting message from the individual SBC, the master resource manager determines a mode of operation, e.g. one of: normal, critical or overshoot, for the cluster and a resource allocation for the SBC, the resource allocation method used being different for at least two different modes. The master resource manager communicates the determined cluster mode and the determined resource allocation to the individual SBC, e.g., in a response message. The SBC controls its resource utilization reporting rate as a function of the mode. In at least one mode, e.g., normal mode, the SBC is allowed to use more resources, e.g., X % more, than the amount of resources allocated to the SBC by the master resource manager.