VXLAN Header L2 Frame Type Detection

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Current methods for determining the L2 frame type in VXLAN packets require either wider and more costly TCAM or increase latency and power consumption through additional processing steps, as typical routers struggle to strip VXLAN headers without leaving the packet intact.

Innovation Solution

A system comprising a VXLAN stripper, L2 frame type reader, and VXLAN re-encapsulator that strips encapsulation headers from VXLAN packets, reads the L2 frame type from the VXLAN header without de-encapsulating the packet, and re-encapsulates the packet in the same format, allowing for efficient L2 frame type discovery and transmission.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If TCAM is used to search entire contents of VXLAN packets, then L2 frame type detection is achieved in single clock cycle, but TCAM width and cost increase

Engineering Contradiction:

ImproveL2 frame type detection accuracyVSAvoidTCAM width and cost

Core Design Contradiction:

Measurement precisionVSDevice complexity

Solution Approach 1:

The patent extracts only the necessary L2 frame type information from the VXLAN packet by reading specific bytes from the VXLAN header (bytes 4-6) rather than searching the entire packet content. This extraction approach obtains the required detection information without requiring wide TCAM, thereby resolving the contradiction between detection accuracy and device complexity.

Inventive Principle:

Principle #2Taking out (Extraction)

2Measurement precision

If current router engines perform multi-step de-encapsulation to remove VXLAN headers, then L2 frame type can be discovered, but latency and power consumption increase

Engineering Contradiction:

ImproveL2 frame type detection accuracyVSAvoidprocessing latency

Core Design Contradiction:

Measurement precisionVSLoss of time

Solution Approach 1:

The patent performs preliminary action by pre-positioning the L2 frame type information in the VXLAN header during encapsulation (bytes 4-6 of the VXLAN header contain the L2 frame type). This allows the receiving router to directly read the frame type without performing multi-step de-encapsulation, significantly reducing processing latency while maintaining detection accuracy.

Inventive Principle:

Principle #10Preliminary action

3Ease of operation

If typical routers strip headers above VXLAN packet, then de-encapsulation is simplified, but VXLAN packet integrity is lost

Engineering Contradiction:

Improvede-encapsulation simplicityVSAvoidVXLAN packet integrity

Core Design Contradiction:

Ease of operationVSReliability

Solution Approach 1:

The patent uses the VXLAN header itself as an intermediary that carries the L2 frame type information (in bytes 4-6). This allows routers to simplify de-encapsulation by reading the frame type directly from the VXLAN header without completely stripping the VXLAN packet, thereby maintaining packet integrity while improving operational simplicity.

Inventive Principle:

Principle #24Intermediary (Mediator)

Data Source

PatentUS11115506B2Inner VXLAN tunnel packet detection

Publication Date: 2021.09.07 LENOVO GLOBAL TECHNOLOGIES SWITZERLAND INTERNATIONAL GMBH

AI summary

An apparatus for writing an L2 frame type in a VXLAN header includes a VXLAN stripper configured to strip encapsulation headers in layers above a VXLAN packet encapsulating a L2 frame of a data packet in response to receiving the data packet over a VXLAN. The apparatus includes a L2 frame type reader configured to read a L2 frame type stored in the VXLAN header in response to stripping the encapsulation headers, and a VXLAN re-encapsulator configured to re-encapsulate the VXLAN packet in a same format as the layers and associated encapsulation headers stripped from the received data packet in response to the L2 frame type indicating the L2 frame comprises data for transmission.