VNF Load Balancing With State Replication for Reliable Failover

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing cloud computing systems face challenges in implementing a control plane architecture for cellular virtualized network functions (VNFs) due to differences in hardware and networking abstractions, leading to difficulties in achieving high availability and reliability.

Innovation Solution

A reliable VNF (rVNF) system is introduced, featuring a load balancer interface for custom application layer load balancing, replicated application storage, and a communication interface that ensures packet delivery to the same physical VNF instance, utilizing a transactional key-value store and socket interface to maintain reliability and availability.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If virtualized cellular VNF appliances use special hardware and software with layer 2 networking abstractions to achieve high availability, then reliability is improved, but device complexity and ease of manufacture worsen

Engineering Contradiction:
Improvehigh availabilityVSAvoidhardware and software complexity
Core Design Contradiction:
ReliabilityVSDevice complexity

Solution Approach 1:

The patent creates virtual copies (virtual machines) of network function instances that can be replicated across multiple physical hosts. These VM copies maintain state information and can take over when primary instances fail, achieving high availability through replication rather than specialized hardware. The load balancer maintains a pool of VM instances that can be dynamically activated.

Inventive Principle:
Principle #26Copying

Solution Approach 2:

The load balancer acts as an intermediary component that sits between clients and the pool of VNF instances. It manages the complexity of failover and load distribution by absorbing the state management complexity internally, while presenting a simple interface to clients. The load balancer tracks which VM instance should handle which user context and routes packets accordingly.

Inventive Principle:
Principle #24Intermediary (Mediator)

2Ease of manufacture

If cloud computing systems use commodity hardware with layer 3 networking abstractions, then ease of manufacture and adaptability improve, but reliability and consistent packet delivery worsen

Engineering Contradiction:
Improvesystem deploymentVSAvoidpacket delivery consistency
Core Design Contradiction:
Ease of manufactureVSReliability

Solution Approach 1:

The load balancer implements per-user-context routing decisions, treating each user context locally and independently. It examines packet metadata to identify the user context and routes packets to the appropriate VM instance that has the relevant state information. This local quality approach ensures that packets for a given user always go to the same instance, maintaining reliability on commodity hardware.

Inventive Principle:
Principle #3Local quality

Solution Approach 2:

The system pre-establishes a pool of virtual machine instances before they are needed for service. The load balancer maintains readiness information about which VM instances are available and their current state. When packets arrive, routing decisions are made quickly based on pre-computed information rather than requiring complex real-time analysis, improving both reliability and performance.

Inventive Principle:
Principle #10Preliminary action

3Productivity

If load balancing distributes packets across multiple physical VNF instances, then productivity and load distribution improve, but reliability worsens due to potential packet loss during failover

Engineering Contradiction:
Improveload distributionVSAvoidpacket delivery
Core Design Contradiction:
ProductivityVSReliability

Solution Approach 1:

The load balancer implements feedback mechanisms by tracking the state of each VM instance and monitoring which instances are currently handling which user contexts. When a VM instance becomes unavailable, the load balancer receives feedback about this failure and dynamically updates its routing table to redirect traffic to healthy instances. This feedback loop ensures continuous service availability.

Inventive Principle:
Principle #23Feedback

Solution Approach 2:

Multiple virtual machine instances are created as copies of the same network function, each capable of handling user contexts independently. The load balancer distributes user contexts across these copies based on current system state and load conditions. If one copy fails, others can take over, ensuring productivity is maintained while reliability is protected through redundancy.

Inventive Principle:
Principle #26Copying

Data Source

PatentUS12474986B2Reliable virtualized network function system for a cloud computing system
Publication Date: 2025.11.18 MICROSOFT TECHNOLOGY LICENSING LLC
  • US12474986B2 patent drawing
  • US12474986B2 patent drawing
  • US12474986B2 patent drawing

AI summary

A reliable network function virtualization (rVNF) system includes a virtualized network function (VNF) application instance that includes a plurality of physical VNF instances. A load balancer provides an interface between a client and the VNF application instance. A load balancer interface facilitates delivery of packets related to a particular user context to the same physical VNF instance. A communication interface facilitates communication between the client and the VNF application instance. Application storage stores session data associated with the VNF application instance.