Media Gateway Controller Resource Allocation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for managing Media Gateways (MGs) in network computer systems, such as those using the H.248 protocol, are reactive and prone to unbalanced load distribution, leading to congestion due to the round robin approach and lack of proactive resource management, which can result in inefficient use of resources and potential congestion.
Innovation Solution
A method where a controlling computer node, such as a Media Gateway Controller (MGC), proactively manages resource allocation by receiving information on resource impacts and available resources from subordinate MGs, allowing it to assign tasks based on calculated resource usage impacts, thereby distributing calls to balance resource usage and prevent congestion.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If a round robin approach is used to distribute calls to Media Gateways, then fairness is improved by spreading calls evenly, but load balance deteriorates because calls with different resource requirements are distributed uniformly without considering actual resource availability
Solution Approach 1:
The system changes the distribution parameters from uniform round-robin to resource-aware selection. The MGC considers multiple parameters including current resource availability, predicted resource usage, and gateway capacity when assigning calls, transforming the simple sequential distribution into a multi-parameter optimization problem that balances fairness with efficiency
Solution Approach 2:
The system performs preliminary resource assessment and prediction before call assignment. By evaluating current resource states and predicting future resource requirements of each gateway, the MGC makes informed distribution decisions in advance, preventing congestion before it occurs rather than reacting to it
2Reliability
If congestion detection mechanisms are used to redirect traffic when congestion occurs, then congestion handling is improved, but response time deteriorates because the mechanism is reactive and applied only after congestion is present
Solution Approach 1:
The system applies preliminary anti-action by predicting potential congestion scenarios and taking preventive measures before congestion actually occurs. The MGC analyzes resource usage trends and predicts future states, then proactively redistributes calls to prevent congestion, rather than waiting for congestion to trigger a response
Solution Approach 2:
The system implements continuous feedback loops where the MGC monitors resource usage at Media Gateways, compares actual usage against predicted usage, and adjusts call distribution decisions in real-time. This feedback mechanism enables dynamic adaptation to changing conditions while maintaining proactive congestion prevention
3Extent of automation
If the MGC reserves and allocates media capability on MGs with holistic view of internal limitations, then resource control is improved, but system complexity increases and the impact of using one unit of media capability on overall MG performance is not considered
Solution Approach 1:
The system introduces an intermediary layer between the MGC and MGs that handles complex resource assessment. This intermediary component evaluates the impact of media capability allocation on overall MG performance, translating complex internal limitations into simplified allocation decisions that the MGC can make without excessive complexity
Solution Approach 2:
The system segments the resource management function into separate components: the MGC handles high-level call distribution, while individual MGs report their resource states and capacity limitations. This segmentation allows the MGC to maintain holistic control without needing to process all detailed internal limitations directly, reducing overall system complexity
Data Source
AI summary
A method performed by a controlling computer node in a network with a plurality of subordinate computing nodes, the method including: receiving, from a first subordinate computer node, information indicating computing resource impacts for a plurality of different processing functions; receiving, from a communication endpoint, a processing task; determining an impact upon available resources of the first subordinate computer node if the task were to be assigned to the first subordinate computer node, wherein the determining is performed using the information indicating computing resource impacts; and assigning the task to either the first subordinate computer node or another one of the subordinate computer nodes based at least in part on the determining the impact.