Arithmetic Offload Packet Handling Without Dedicated NICs

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Existing offload systems face challenges in achieving low delay and versatility due to overhead in cooperation between the OS protocol stack and ACC function/argument data, requiring dedicated NICs.

Innovation Solution

An arithmetic processing offload system integrates OS protocol processing functions with ACC function/argument data handling into single dedicated functions, eliminating overhead by using XDP/eBPF and bypassing the existing protocol stack.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If the OS protocol stack and ACC function/argument data handling are processed through separate independent units, then the system maintains modularity and standard compliance, but processing overhead increases and delay occurs

Engineering Contradiction:

Improvesystem modularityVSAvoidprocessing delay

Core Design Contradiction:

ReliabilityVSLoss of time

Solution Approach 1:

The patent merges the OS protocol stack processing functions with ACC function/argument data handling into a single integrated processing unit. This combination eliminates the overhead of inter-unit communication and data transfer between separate protocol stack and ACC handling units, thereby reducing processing delay while maintaining the functional capabilities of both components

Inventive Principle:

Principle #5Merging (Combining)

2Speed

If dedicated NICs are used for accelerator processing offloading, then processing performance and low latency are achieved, but system versatility and compatibility are reduced

Engineering Contradiction:

Improveprocessing speedVSAvoidsystem compatibility

Core Design Contradiction:

SpeedVSAdaptability or versatility

Solution Approach 1:

The patent implements a unified processing unit that can handle both standard OS protocol stack operations and ACC function/argument data processing through a single interface. This multi-functional design allows the system to achieve high-speed processing without requiring dedicated specialized NICs, thereby maintaining versatility and compatibility with standard network interfaces

Inventive Principle:

Principle #6Universality (Multi-functionality)

3Reliability

If multiple independent processing units are used for protocol stack and ACC data handling, then functional separation is maintained, but device complexity increases

Engineering Contradiction:

Improvefunctional separationVSAvoidsystem complexity

Core Design Contradiction:

ReliabilityVSDevice complexity

Solution Approach 1:

The patent combines multiple independent processing functions into a single integrated processing unit that handles both OS protocol stack operations and ACC data processing. This consolidation reduces the number of separate components and interfaces required, thereby simplifying the overall system architecture while maintaining the functional capabilities of both processing domains

Inventive Principle:

Principle #5Merging (Combining)

Data Source

PatentUS12500847B2Arithmetic processing offload system, client, server, and arithmetic processing offload method

Publication Date: 2025.12.16 NT T INC

AI summary

An operation system of a client includes: an L3/L4 protocol and ACC function and argument data packetizing unit that serializes a function name and argument input from an application side according to a format of a predetermined protocol and packetizes the function name/argument as a payload; and an L3/L4 protocol and ACC function and return value data parsing unit that deserializes packet data input from a server side according to a format of a predetermined protocol and acquires a function name/execution result.