Interface Device Parallel Data Reception via Segmented Buffer

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing interface devices with a single reception buffer cause undesired delays and latencies because data reception by one interface circuit must be completed before another can start, limiting parallel data reception capabilities.

Innovation Solution

Implementing a plurality of interface circuits that share a single reception buffer, where each interface circuit acquires and identifies packets with a unique tag ID, allowing parallel data reception and storage in the buffer, enabling the processor to manage and process packets efficiently.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Device complexity

If a single reception buffer is used for multiple interface circuits, then device complexity is reduced, but data reception speed and parallel processing capability deteriorate

Engineering Contradiction:
Improvebuffer structureVSAvoiddata reception speed
Core Design Contradiction:
Device complexityVSSpeed

Solution Approach 1:

The reception buffer is segmented into multiple buffer regions, each assigned to a specific interface circuit. This segmentation allows multiple interface circuits to write data to different buffer regions simultaneously without conflict, while still using a single physical buffer memory, thus maintaining low device complexity while improving data reception speed and parallel processing capability.

Inventive Principle:
Principle #1Segmentation

2Ease of manufacture

If a single reception buffer is shared by multiple interface circuits, then manufacturing cost is reduced, but productivity and data processing throughput deteriorate

Engineering Contradiction:
Improvebuffer implementationVSAvoiddata processing throughput
Core Design Contradiction:
Ease of manufactureVSProductivity

Solution Approach 1:

The buffer is divided into multiple buffer regions with each region dedicated to a specific interface circuit. This allows parallel data reception from multiple interfaces simultaneously, significantly improving data processing throughput while still using a single shared buffer memory, thus maintaining ease of manufacture and low cost.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

A buffer management unit acts as an intermediary between multiple interface circuits and the buffer regions. It manages the allocation and coordination of buffer regions to different interface circuits, enabling efficient parallel data reception and processing while maintaining simple buffer implementation.

Inventive Principle:
Principle #24Intermediary (Mediator)

3Productivity

If multiple buffer regions are created in a shared buffer, then parallel data reception capability is improved, but device complexity increases

Engineering Contradiction:
Improveparallel data reception capabilityVSAvoidbuffer management structure
Core Design Contradiction:
ProductivityVSDevice complexity

Solution Approach 1:

The buffer is segmented into multiple buffer regions that can be independently accessed by different interface circuits. This segmentation enables parallel data reception while keeping the buffer management structure relatively simple through dedicated region assignment rather than complex dynamic allocation.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The single buffer memory serves multiple functions by being divided into multiple buffer regions that can handle different interface circuits simultaneously. This multi-functionality approach improves parallel data reception capability while avoiding the need for multiple separate buffer memories, thus limiting the increase in device complexity.

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

Data Source

PatentUS10445042B1Interface device and printer
Publication Date: 2019.10.15 TOSHIBA TEC KK
  • US10445042B1 patent drawing
  • US10445042B1 patent drawing

AI summary

In accordance with an embodiment, an interface device comprises a plurality of interface circuits which respectively corresponds to different transmission paths and sequentially acquires a plurality of packets formed by dividing data which is an object of any one of plural types of information processing via the corresponding transmission paths, a reception buffer and a processor. The processor writes the packet acquired by each of the plurality of the interface circuits in the reception buffer in association with an identification code for identifying the interface circuit acquiring the packet, sequentially reads a plurality of packets associated with the same identification code among the packets stored in the reception buffer from the reception buffer, and delivers the read packets to the information processing on the data which is a basis of the packet.