Network Interface Controller Data Scattering via Static Steering Rules

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Current communication systems face inefficiencies in data processing and ordering in memory, particularly in 5G networks, where data from multiple user equipment devices is interleaved and requires complex CPU and GPU processing to reorder and scatter data correctly, leading to processor-intensive solutions that are not scalable.

Innovation Solution

A data communication system that uses a network interface controller to scatter data into memory buffers using a static set of steering rules, allowing the same scatter pattern to be applied across multiple time slots, reducing the need for frequent CPU intervention and enabling efficient data ordering for processing units like GPUs and CPUs.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Manufacturing precision

If complex CPU and GPU processing is used to reorder and scatter data correctly, then data ordering accuracy is improved, but processor load and system complexity increase

Engineering Contradiction:
Improvedata ordering accuracyVSAvoidprocessor load
Core Design Contradiction:
Manufacturing precisionVSDevice complexity

Solution Approach 1:

The patent applies preliminary action by pre-configuring static steering rules in the network interface controller before data arrives. These rules define the scatter pattern in advance, allowing the NIC to automatically scatter data packets to the correct memory buffers without requiring complex real-time processing by CPU or GPU, thus maintaining data ordering accuracy while reducing processor load

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The network interface controller performs self-service by autonomously scattering data packets to memory buffers using the pre-configured static steering rules. The NIC independently determines the destination buffer for each packet based on packet timing information and the static rules, eliminating the need for external CPU/GPU intervention in the scattering process

Inventive Principle:
Principle #25Self-service

2Manufacturing precision

If frequent CPU intervention is used to scatter data, then data scattering accuracy is improved, but processing latency increases

Engineering Contradiction:
Improvedata scattering accuracyVSAvoidprocessing latency
Core Design Contradiction:
Manufacturing precisionVSLoss of time

Solution Approach 1:

The network interface controller autonomously scatters data packets to memory buffers using pre-configured static steering rules without requiring frequent CPU intervention. The NIC independently uses packet timing information to determine scatter destinations, maintaining accuracy while significantly reducing processing latency by eliminating CPU involvement in the time-critical scattering operation

Inventive Principle:
Principle #25Self-service

3Adaptability or versatility

If dynamic scatter patterns are used for each time slot, then adaptability is improved, but device complexity and processing overhead increase

Engineering Contradiction:
Improvescatter pattern adaptabilityVSAvoidprocessing overhead
Core Design Contradiction:
Adaptability or versatilityVSDevice complexity

Solution Approach 1:

The patent applies universality by using a single static set of steering rules that can be repeatedly applied across multiple buffer cycles and time slots. The same static rules handle diverse data scattering needs by combining them with packet timing information, providing adaptability without requiring separate dynamic rule sets for each time slot, thus reducing processing overhead while maintaining versatility

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

Data Source

PatentUS12068968B2Efficient scattering to buffers
Publication Date: 2024.08.20 MELLANOX TECHNOLOGIES LTD(IL)
  • US12068968B2 patent drawing
  • US12068968B2 patent drawing
  • US12068968B2 patent drawing

AI summary

In one embodiments, data communication system include a communication apparatus, which is configured to receive data from different user equipment devices a schedule of time periods, and packetize the data from respective ones of the user equipment devices for respective ones of the time periods into packets, a memory including a plurality of buffers, and a network interface controller configured to receive the packets from the communication apparatus, and scatter respective portions of the data belonging to respective groups of successive ones of the time periods to the buffers, responsively to a static set of steering rules, and timing information of respective ones of the packets, and wherein each respective portion of the data is scattered to the buffers a same scatter pattern.