Ethernet IP Packet Header Compression for HPC Bandwidth

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Current Ethernet and IP packets have large packet headers, which are detrimental to efficiency in small-payload applications, leading to significant overhead and impairment of bandwidth utilization and delay, especially in high-performance computing and voice over IP scenarios.

Innovation Solution

Optimized packet headers are introduced, which are stateless and end-to-end, eliminating or shortening unused fields in Ethernet and IPv4 headers, increasing the payload-to-packet header ratio, and are particularly beneficial for small-packet applications like high-performance computing and voice over IP.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Productivity

If standard Ethernet and IP packet headers are used, then network compatibility and device recognition are ensured, but packet header size increases leading to reduced bandwidth utilization and increased delay in small-payload applications

Engineering Contradiction:
Improvebandwidth utilizationVSAvoidpacket header size
Core Design Contradiction:
ProductivityVSDevice complexity

Solution Approach 1:

The patent extracts and removes unnecessary fields from the standard Ethernet and IP packet headers, retaining only the essential information needed for network communication. This includes eliminating redundant addressing fields and protocol-specific fields that are not required for basic packet forwarding, thereby reducing header size while maintaining network compatibility.

Inventive Principle:
Principle #2Taking out (Extraction)

Solution Approach 2:

The patent changes the parameter of packet header size by introducing a compressed header format that uses fewer bytes to represent the same network information. This involves reencoding addressing and control information in a more compact form, directly improving bandwidth utilization by reducing the overhead portion of each packet.

Inventive Principle:
Principle #35Parameter changes

2Loss of time

If standard Ethernet and IP packet headers are used, then comprehensive network information is available for routing and control, but packet header size increases leading to increased processing delay

Engineering Contradiction:
Improvepacket processing delayVSAvoidpacket header size
Core Design Contradiction:
Loss of timeVSDevice complexity

Solution Approach 1:

The patent extracts only the critical fields needed for rapid packet forwarding, removing less essential information that would increase processing time. This selective extraction reduces the amount of data that network devices must parse and process at each hop, thereby reducing end-to-end delay.

Inventive Principle:
Principle #2Taking out (Extraction)

Solution Approach 2:

The patent applies partial action by including only the necessary subset of header fields rather than the complete standard header set. This partial inclusion of header information is sufficient for most network operations while significantly reducing processing overhead and delay.

Inventive Principle:
Principle #16Partial or excessive action

3Productivity

If complete IP address and MAC address fields are included in packet headers, then precise network device identification is achieved, but packet header size increases reducing efficiency in small-packet applications

Engineering Contradiction:
Improvepacket efficiencyVSAvoidheader field size
Core Design Contradiction:
ProductivityVSQuantity of substance

Solution Approach 1:

The patent extracts and removes complete MAC address fields from packet headers, relying instead on IP address-based routing which is already present. This extraction eliminates 6 bytes of redundant addressing information per packet while maintaining precise device identification through IP addresses alone.

Inventive Principle:
Principle #2Taking out (Extraction)

Solution Approach 2:

The patent makes the IP address field serve multiple functions: it provides both routing information and device identification, replacing the need for separate MAC address fields. This multi-functionality reduces header size while maintaining the ability to precisely identify network devices.

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

Data Source

PatentUS20230412712A1Packet header optimization in ethernet internet protocol networks
Publication Date: 2023.12.21 INTEL CORP
  • US20230412712A1 patent drawing
  • US20230412712A1 patent drawing
  • US20230412712A1 patent drawing

AI summary

Described herein are optimized packet headers for Ethernet IP networks and related methods and devices. An example packet header includes a field comprising a source identifier (SID), the SID comprising a shortened representation of a complete Internet Protocol (IP) address of a source network device, a field comprising a destination identifier (DID), the DID comprising a shortened representation of a complete IP address of a destination network device, and a field having a total number of bits that is less than 8 and comprising a shortened representation of a type of encapsulation protocol for the packet. The packet header excludes fields comprising the complete IP address and a media access controller (MAC) address of the source network device, fields comprising the complete IP address and the MAC address of the destination network device, a field comprising a header checksum, and a field comprising a total size of the packet.