Dynamic Priority Arbitration for Data Transmission Balance

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing many-core chip designs face challenges in effectively determining data source priority for transmission and dynamically adjusting priority orders, leading to imbalances in data transmission delay and the number of data sources reaching the destination.

Innovation Solution

A data processing method that determines candidate data pieces and assigns priority sequence numbers based on nonlinear relationships, ensuring that the same data source has varying priorities across processing cycles, and uses encoding modes and bit-to-bit swaps to generate target codes for arbitrating data sources, thereby dynamically adjusting priority orders.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Ease of operation

If fixed priority arbitration is used for data channel allocation, then arbitration simplicity is improved, but data transmission balance deteriorates

Engineering Contradiction:
Improvearbitration simplicityVSAvoiddata transmission balance
Core Design Contradiction:
Ease of operationVSReliability

Solution Approach 1:

The patent implements dynamic priority adjustment where the priority of each data source is not fixed but changes over time based on a round-robin scheduling mechanism. The arbitration unit dynamically updates priority sequences in each cycle, ensuring that data sources that have not transmitted recently receive higher priority, thus balancing data transmission across multiple sources while maintaining manageable arbitration complexity through systematic priority rotation.

Inventive Principle:
Principle #15Dynamics

2Reliability

If round-robin scheduling is used for priority adjustment, then data transmission balance is improved, but arbitration complexity increases

Engineering Contradiction:
Improvedata transmission balanceVSAvoidarbitration complexity
Core Design Contradiction:
ReliabilityVSDevice complexity

Solution Approach 1:

The patent segments the arbitration process into multiple levels (first-level arbitration units and second-level arbitration units), where each level handles specific aspects of priority determination. The first-level units perform initial arbitration based on round-robin priorities, and the second-level unit performs final arbitration. This segmentation distributes the arbitration complexity across multiple simple units rather than requiring one complex arbitration unit, thus achieving balanced data transmission while keeping individual arbitration units relatively simple.

Inventive Principle:
Principle #1Segmentation

3Productivity

If multi-level arbitration is implemented, then data channel arbitration effectiveness is improved, but system complexity increases

Engineering Contradiction:
Improvedata channel arbitration effectivenessVSAvoidsystem complexity
Core Design Contradiction:
ProductivityVSDevice complexity

Solution Approach 1:

The patent merges the priority determination function and the arbitration function into an integrated arbitration unit. The arbitration unit simultaneously maintains priority sequences for multiple data sources and performs arbitration based on these priorities in a unified process. This merging reduces system complexity by eliminating separate priority management and arbitration components, while still achieving effective multi-level arbitration through the integrated unit's ability to handle both functions cooperatively.

Inventive Principle:
Principle #5Merging (Combining)

Data Source

PatentUS11863461B2Data processing method, data processing apparatus, electronic device, storage medium, and program product
Publication Date: 2024.01.02 LYNXI TECH CO LTD
  • US11863461B2 patent drawing
  • US11863461B2 patent drawing
  • US11863461B2 patent drawing

AI summary

Provided in the present disclosure are a data processing method and apparatus, and an electronic device, the method includes: determining a plurality of candidate data pieces, where the candidate data pieces are provided from corresponding data sources; and determining a target data piece based on priorities of the data sources corresponding to the plurality of candidate data pieces in a current cycle, wherein a same data source has different priorities in different processing cycles, and priority sequence numbers of a same data source in different processing cycles satisfy a nonlinear relationship.