Aggregate VTEP Segmentation for VXLAN Scalability

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Conventional network virtualization techniques face scalability issues due to bandwidth waste and congestion, especially in VXLAN deployments, where standard VTEPs must learn and replicate MAC addresses across large numbers of hosts, leading to performance degradation and manageability challenges.

Innovation Solution

Implementing aggregate VXLAN tunnel end points (VTEPs) that replicate frames only within physical proximity, reducing the need for remote VTEP communication and using MAC translation to minimize the number of learned MAC addresses, thereby conserving bandwidth and processing power.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If standard VTEPs replicate frames to all other VTEPs in the network, then connectivity is ensured, but bandwidth is wasted and congestion occurs

Engineering Contradiction:
ImproveconnectivityVSAvoidbandwidth waste
Core Design Contradiction:
ReliabilityVSLoss of energy

Solution Approach 1:

The patent divides the virtual overlay network into multiple broadcast domains using aggregate VTEPs as boundaries. Each aggregate VTEP manages a subset of VTEPs, creating segmented regions where frame replication is localized. This segmentation prevents unnecessary replication across the entire network while maintaining connectivity within each segment.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent implements local frame replication at aggregate VTEPs within each physical proximity, rather than global replication across all VTEPs. Each aggregate VTEP replicates frames only to VTEPs in its local physical proximity, optimizing bandwidth usage by limiting replication scope to where it is actually needed.

Inventive Principle:
Principle #3Local quality

2Measurement precision

If standard VTEPs learn MAC addresses of all hosts in the network, then forwarding accuracy is improved, but processing load increases

Engineering Contradiction:
Improveforwarding accuracyVSAvoidprocessing load
Core Design Contradiction:
Measurement precisionVSProductivity

Solution Approach 1:

The patent segments the MAC address learning scope by introducing aggregate VTEPs. Standard VTEPs only learn MAC addresses of hosts within their local broadcast domain managed by their aggregate VTEP, rather than learning all MAC addresses in the entire network. This segmentation reduces the forwarding table size and processing load on standard VTEPs while maintaining accurate forwarding within each segment.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The aggregate VTEP acts as an intermediary that learns and maintains MAC address information for its broadcast domain. Standard VTEPs interact with the aggregate VTEP rather than directly learning from all hosts, reducing their processing burden while the aggregate VTEP handles the complexity of MAC address management.

Inventive Principle:
Principle #24Intermediary (Mediator)

3Loss of energy

If aggregate VTEPs are introduced to reduce replication scope, then bandwidth is conserved, but device complexity increases

Engineering Contradiction:
Improvebandwidth conservationVSAvoidVTEP architecture complexity
Core Design Contradiction:
Loss of energyVSDevice complexity

Solution Approach 1:

The patent combines the functions of multiple standard VTEPs into a single aggregate VTEP. The aggregate VTEP performs frame replication, MAC address learning, and broadcast domain management for multiple VTEPs, reducing overall network complexity by consolidating control functions while conserving bandwidth through localized replication.

Inventive Principle:
Principle #5Merging (Combining)

Data Source

PatentUS9596099B2Scalable network virtualization with aggregate endpoints
Publication Date: 2017.03.14 CISCO TECHNOLOGY INC
  • US9596099B2 patent drawing
  • US9596099B2 patent drawing
  • US9596099B2 patent drawing

AI summary

Techniques provided herein use aggregate endpoints in a virtual overlay network. In general, aggregate endpoints operate as a single receiving entity for certain packets/frames sent between different physical proximities of the virtual overlay network.