Virtual Machine Cluster Redundancy for Telecommunications
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The existing N×(1+1) configuration in telecommunications networks is inefficient in resource utilization, as backup virtual machines use and reserve more resources than needed during normal processing, leading to increased hardware costs and complexity, especially in virtualized environments where high availability and emergency call handling are critical.
Innovation Solution
Implementing an N+k configuration where active virtual machines share backup resources, using a communication session classifier to route emergency and high-priority calls to dedicated 1:1 redundant pairs while routing non-priority calls to shared simplex instances, thereby reducing hardware requirements and maintaining high availability.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If N×(1+1) configuration with dedicated backup virtual machines is used, then high availability and reliability are ensured, but hardware costs and resource utilization efficiency deteriorate
Solution Approach 1:
Multiple active virtual machines share a common backup virtual machine instead of each having a dedicated backup. The backup VM is merged into a shared resource that serves multiple active VMs, reducing total hardware consumption while maintaining failover capability.
Solution Approach 2:
The backup virtual machine is designed to be universal, capable of taking over for any of the active virtual machines in the group. This multi-functional backup can assume the role of any failed active VM, eliminating the need for dedicated one-to-one backup pairs.
2Reliability
If N×(1+1) configuration with dedicated backup virtual machines is used, then failover capability is ensured, but device complexity increases
Solution Approach 1:
The backup resources and failover management are merged into a shared infrastructure. Instead of managing separate backup pairs, the system uses a unified backup pool that automatically handles failover for multiple active VMs, simplifying configuration and management.
Solution Approach 2:
A virtualization management layer acts as an intermediary between active virtual machines and shared backup resources. This mediator handles the complexity of resource allocation, state synchronization, and failover coordination, shielding users from underlying complexity.
3Reliability
If all virtual machines use dedicated backup resources, then maximum redundancy is achieved, but resource utilization efficiency during normal processing deteriorates
Solution Approach 1:
The backup resource allocation is made dynamic rather than static. During normal operation, backup VMs consume minimal resources in standby mode. When failover is needed, the backup dynamically activates and assumes the full resource requirements of the failed active VM, optimizing resource utilization across different operational states.
Solution Approach 2:
The resource allocation parameters of backup virtual machines are changed based on operational state. In standby mode, backups use reduced resources; upon activation, they dynamically adjust to full resource requirements, allowing efficient resource usage while maintaining redundancy capability.
Data Source
AI summary
A virtualized computing environment of a telecommunications network comprises a cluster of virtual machines with a one-to-one ratio of active and backup virtual machines. One or more additional clusters of virtual machines have a N-to-K ratio of active and backup virtual machines where N>K. The backup virtual machines are configured to provide failover capacity for processing communications sessions in an event of a failure of one of the active virtual machines. A cluster redundancy capability indicates the ratio of the active and backup virtual machines for that cluster. A predetermined type associated with a requested communication session is determined. A cluster having a cluster redundancy capability corresponding to the predetermined type is selected. Data for the requested communication session is sent to an active virtual machine in the selected cluster.