Scalable Load Balancing Across Layer-2 Domains

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Conventional load balancers are expensive due to specialized hardware and have scalability limitations, as they typically support only a single virtual local area network (VLAN) and require more powerful hardware to handle increased requests, which restricts their ability to distribute traffic across multiple IP subnets.

Innovation Solution

A scalable load balancing system using inter-layer-2 domain packet delivery techniques, including IP-in-IP encapsulation and Direct Server Return (DSR) optimization, allows traffic distribution across multiple IP subnets, enabling the use of an essentially unlimited number of target devices and bypassing the load balancer for return communications, thus reducing bottlenecks and conserving resources.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If conventional load balancers use specialized hardware to handle load balancing, then load balancing functionality is achieved, but cost increases and scalability is limited

Engineering Contradiction:
Improveload balancing functionalityVSAvoidspecialized hardware
Core Design Contradiction:
ReliabilityVSDevice complexity

Solution Approach 1:

The patent uses software-based load balancing that replicates the functionality of specialized hardware load balancers using general-purpose servers. The load balancing function is copied into software form (virtual appliances, containerized applications) that can run on standard hardware, eliminating the need for expensive specialized devices while maintaining load balancing capabilities.

Inventive Principle:
Principle #26Copying

Solution Approach 2:

The patent replaces the mechanical/hardware-based load balancing system with a software-based system. Instead of using dedicated hardware appliances, the load balancing functionality is implemented through software layers that can operate on commodity servers, substituting the physical hardware approach with a software-driven architecture.

Inventive Principle:
Principle #28Mechanics substitution (Replace mechanical system)

2Productivity

If conventional load balancers handle more concurrent requests by purchasing more powerful hardware, then request handling capacity increases, but cost and hardware complexity increase

Engineering Contradiction:
Improveconcurrent request handling capacityVSAvoidhardware capacity
Core Design Contradiction:
ProductivityVSDevice complexity

Solution Approach 1:

The patent segments the load balancing function into distributed software components that can be deployed across multiple standard servers. Instead of consolidating all request handling in a single powerful hardware appliance, the load balancing capacity is divided into multiple smaller software instances that can be distributed and scaled independently across the infrastructure.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent makes standard servers universal by enabling them to perform load balancing functions through software. General-purpose servers with commodity hardware can fulfill the role previously reserved for specialized load balancers, allowing the same hardware platform to serve multiple purposes and scale elasticity without requiring dedicated high-capacity hardware.

Inventive Principle:
Principle #6Universality (Multi-functionality)

3Productivity

If Direct Server Return (DSR) optimization is used, then traffic bottlenecks are mitigated, but the technique is limited to a single VLAN

Engineering Contradiction:
Improvetraffic flow efficiencyVSAvoidVLAN scope
Core Design Contradiction:
ProductivityVSAdaptability or versatility

Solution Approach 1:

The patent extends DSR functionality from a single VLAN (two-dimensional limitation) to multiple VLANs and IP subnets (adding a third dimension of network scope). By implementing software-based load balancing with multi-VLAN support, the system maintains the traffic bypass optimization of DSR while expanding its applicability across broader network architectures through additional dimensional coverage.

Inventive Principle:
Principle #17Another dimension (Dimensionality change)

Solution Approach 2:

The patent introduces a software-based load balancing layer that acts as an intermediary between clients and servers across multiple VLANs. This software mediator enables DSR-like traffic optimization while spanning multiple network segments, bridging the gap between single-VLAN DSR limitations and multi-VLAN requirements through a flexible software intermediary that can operate across network boundaries.

Inventive Principle:
Principle #24Intermediary (Mediator)

4Device complexity

If load balancers are confined to a single VLAN, then network simplicity is maintained, but traffic distribution across multiple IP subnets is restricted

Engineering Contradiction:
Improvenetwork architecture simplicityVSAvoidmulti-subnet traffic distribution
Core Design Contradiction:
Device complexityVSAdaptability or versatility

Solution Approach 1:

The patent implements a dynamic load balancing system that can adapt to multi-VLAN and multi-subnet environments through software configuration. Unlike static hardware load balancers confined to single VLANs, the software-based solution dynamically adjusts to different network topologies, VLAN configurations, and IP subnet arrangements, maintaining operational simplicity while providing versatile multi-subnet traffic distribution capabilities.

Inventive Principle:
Principle #15Dynamics

Data Source

PatentEP2436156B1Load balancing across layer-2 domains
Publication Date: 2016.11.09 MICROSOFT TECHNOLOGY LICENSING LLC
  • EP2436156B1 patent drawingFigure 1
  • EP2436156B1 patent drawingFigure 2
  • EP2436156B1 patent drawingFigure 3

AI summary

The present application relates to network configurations and specifically to scalable load balancing network configurations. One implementation includes an external client coupled to a scalable load balancing system. The scalable load balancing system includes a load balancing layer that is configured to encapsulate individual incoming packets of a packet flow from the external client. The load balancing layer is further configured to route the incoming packets to target devices on the system. The target devices can span multiple IP subnets. The incoming packets can pass through one or more load balancers of the load balancing layer before reaching individual target devices. Individual target devices can be configured to route at least some outgoing packets of the packet flow to the external client without passing through any of the one or more load balancers.