Media Server Load Balancing via Dynamic Scaling and Session Quiescing
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing load balancing systems struggle to dynamically adjust computing resources to match changing workloads, leading to insufficient resources during high demand and excess capacity during low demand, resulting in increased response times or costs.
Innovation Solution
A system for load balancing across media server instances that dynamically resizes computing resources by adding or removing instances based on workload conditions, maintaining long-running sessions during scaling-in operations by transitioning instances to a quiesced state before termination.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If computing resources are statically allocated to media server instances, then system stability is maintained, but resource utilization efficiency deteriorates during varying workload conditions
Solution Approach 1:
The patent implements dynamic resource allocation by enabling media server instances to be added or removed from the pool based on real-time workload conditions. The system transitions from static to dynamic configuration, where the number and capacity of computing resources automatically adjust to match demand, thereby improving resource utilization efficiency while maintaining system stability through controlled scaling operations.
Solution Approach 2:
The system changes the parameter of resource allocation from fixed to variable by implementing scaling policies that adjust the number of media server instances based on workload metrics. This parameter change allows the system to optimize resource utilization during different demand periods while maintaining operational stability through gradual and controlled adjustments.
2Productivity
If media server instances are terminated during scaling-in operations, then excess capacity is reduced, but long-running sessions are disrupted
Solution Approach 1:
The patent applies preliminary action by identifying and migrating long-running sessions to other available media server instances before terminating the original instance. This proactive approach ensures that sessions are relocated in advance, allowing the system to remove excess capacity and optimize resource utilization without disrupting ongoing sessions, thereby maintaining both resource optimization and session continuity.
Solution Approach 2:
The system uses an intermediary mechanism (session migration process) to transfer active sessions from instances being terminated to remaining instances. This intermediary action serves as a bridge between the need to remove excess capacity and the requirement to maintain session continuity, enabling resource optimization while protecting session integrity.
3Reliability
If computing resources are increased during high demand, then service quality is maintained, but costs increase due to excess capacity during low demand
Solution Approach 1:
The system dynamically changes the parameter of resource allocation by scaling the number of media server instances up during high demand periods to maintain service quality and scaling down during low demand periods to reduce costs. This parameter adjustment enables the system to match resource capacity with actual workload requirements, optimizing both service quality and cost efficiency.
Solution Approach 2:
The patent implements feedback mechanisms that monitor workload conditions and automatically trigger scaling operations. The system continuously receives feedback about resource utilization and demand patterns, using this information to adjust the number of active instances accordingly, thereby maintaining service quality during high demand while reducing costs during low demand through automated, data-driven decisions.
Data Source
AI summary
A system is presented for load balancing across media server instances. In an embodiment, media is broken out into a multi-tenanted service allowing the media to be scaled independently of the number of organizations supported on a cloud-based collaboration platform. Scaling may occur in a scaling-out or a scaling-in operation. States for a media service may comprise in-service, quiescing, quiesced, failed, etc. The states may be used to monitor sessions associated with an instance and determine which media instances to terminate during a scaling-in operation. In an embodiment, new instances may be added to a collection of media instances in response to an increased workload in a scaling-out operation.