Scalable Network Virtualization via Segregated Global and Local VLAN Tags

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing communication networks face challenges in achieving scalable and segregated network virtualization, particularly in supporting a large number of tenants, as traditional VLAN tagging methods limit the number of supported VLANs and fail to distinguish between datacenters, leading to inefficiencies and increased costs due to the need for complex and expensive switches.

Innovation Solution

The solution involves mapping edge VLANs to unique global VLANs, allowing for a larger number of supported VLANs and enabling segregation between datacenters, using a fabric switch architecture that allows arbitrary topology and automatic configuration, and employing TRILL routing to manage traffic efficiently, thereby overcoming the limitations of traditional switch stacking and VLAN tagging.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Adaptability or versatility

If traditional VLAN tagging methods are used to support multiple tenants, then network virtualization is enabled, but the number of supported VLANs is limited and segregation between datacenters is lost

Engineering Contradiction:
Improvenetwork virtualization capabilityVSAvoidnumber of supported VLANs
Core Design Contradiction:
Adaptability or versatilityVSQuantity of substance

Solution Approach 1:

The patent segments the VLAN identification into two parts: global VLAN tags for datacenter-wide segmentation and local VLAN tags for tenant-specific segmentation. This allows the network to support a much larger number of VLANs by combining the segmentation capabilities of both tag types, resolving the limitation of traditional single-tag VLAN systems.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent introduces a new dimension of segmentation by adding global VLAN tags to the existing local VLAN tag system. This creates a two-dimensional VLAN identification space (global × local) that exponentially increases the number of supported VLANs while maintaining backward compatibility with existing VLAN infrastructure.

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

2Productivity

If larger and faster switches are built to support more tenants and traffic, then bandwidth capacity increases, but device complexity and cost increase

Engineering Contradiction:
Improvebandwidth capacityVSAvoidswitch complexity
Core Design Contradiction:
ProductivityVSDevice complexity

Solution Approach 1:

The patent segments the switching function into hierarchical levels: global VLAN routing at the core and local VLAN forwarding at the edge. This allows simpler edge switches to handle most traffic locally while only core switches perform complex global routing, reducing the complexity burden on individual switches while maintaining high overall capacity.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent introduces global VLAN tags as intermediary identifiers that enable efficient traffic routing between datacenters without requiring edge switches to maintain complex routing tables for all possible destinations. The global tags act as a simplified indexing mechanism that reduces switching complexity.

Inventive Principle:
Principle #24Intermediary (Mediator)

3Adaptability or versatility

If a fabric switch architecture is used to achieve arbitrary topology, then scalability improves, but configuration complexity increases

Engineering Contradiction:
Improvetopology flexibilityVSAvoidconfiguration complexity
Core Design Contradiction:
Adaptability or versatilityVSDevice complexity

Solution Approach 1:

The patent implements automatic configuration mechanisms where switches autonomously discover and establish global VLAN mappings and routing paths without manual intervention. The system self-configures the fabric topology and VLAN associations, eliminating the need for complex manual configuration while maintaining arbitrary topology flexibility.

Inventive Principle:
Principle #25Self-service

Solution Approach 2:

The patent incorporates feedback mechanisms where switches automatically exchange information about global VLAN configurations and routing capabilities with neighboring switches. This feedback loop enables dynamic topology adaptation and automatic path optimization without requiring complex pre-planning or manual configuration.

Inventive Principle:
Principle #23Feedback

Data Source

PatentEP3261301B1Scalable and segregated network virtualization
Publication Date: 2019.01.30 AVAGO TECHNOLOGIES INTERNATIONAL SALES PTE LTD
  • EP3261301B1 patent drawingFigure 1
  • EP3261301B1 patent drawingFigure 2A~2B
  • EP3261301B1 patent drawingFigure 2C

AI summary

One embodiment of the present invention provides a switch 700 comprising a storage device 750 to store a port profile in association with a datacenter domain. The datacenter domain indicates a set of ports associated with a datacenter. The port profile includes configuration information associated with one or more media access control (MAC) addresses. The switch further comprises a port profile module 740 to identify the port profile in response to identifying that a source MAC address of a packet is associated with the datacenter domain, wherein the source MAC address is in the one or more MAC addresses. The port profile module 740 applies the port profile to an ingress port of the packet. A port profile can contain the entire configuration needed for a virtual machine to gain access to a LAN or WAN, which can include: Fibre Channel over Ethernet (FCoE) configuration, VLAN configuration, QoS related configuration, and security related configuration, such as access control lists (ACLs). In one embodiment, a port profile can be capable of operating as a self contained configuration container. In other words, if a port profile is applied to a new switch without any additional configuration, the port profile should be sufficient to set the switch's global and local (interface level) configuration and allow the switch to start carrying traffic.In an example, a switch segregates port profiles for a respective datacenter domain.