Parallel Decode Circuits for Network-on-Chip Latency

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Network-on-chip (NoC) systems face latency issues due to sequential address decode and packet arbitration processes, which hinder efficient communication between cores and logic blocks, especially in complex multicore processors where bandwidth requirements demand improved interconnect architectures.

Innovation Solution

Implementing parallel address decode and packet arbitration in routers within the NoC system, where direction decoding occurs in parallel with packet arbitration, reducing latency by removing sequential dependencies and improving resource utilization.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Loss of time

If sequential address decode and packet arbitration are used in routers, then device complexity is reduced, but network latency increases

Engineering Contradiction:
Improvenetwork latencyVSAvoidrouter complexity
Core Design Contradiction:
Loss of timeVSDevice complexity

Solution Approach 1:

The address decode operation is performed in advance during the packet arbitration phase, so that when the arbitration is complete, the decode result is already available. This eliminates the sequential dependency where decode must wait for arbitration to finish, thereby reducing network latency without significantly increasing router complexity.

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The patent transforms the sequential execution model into a parallel execution model by introducing a temporal dimension - performing address decode and packet arbitration simultaneously in different phases of the same clock cycle. This dimensional transformation allows both operations to complete faster overall, reducing latency while maintaining manageable complexity.

Inventive Principle:
Principle #17Another dimension (Dimensionality change)

2Productivity

If parallel address decode and packet arbitration are implemented, then network latency is reduced, but device complexity increases

Engineering Contradiction:
Improvecommunication efficiencyVSAvoidrouter complexity
Core Design Contradiction:
ProductivityVSDevice complexity

Solution Approach 1:

The router operation is segmented into distinct phases within a clock cycle: the packet arbitration phase where address decode is preliminarily performed, and the forwarding phase where the decoded address is used. This segmentation allows parallel operations to be organized systematically, improving communication efficiency while controlling complexity through structured phase separation.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The address decode logic is designed to serve multiple purposes: it determines the destination router for packet forwarding, and simultaneously provides address information for routing decisions. This multi-functionality allows the same hardware structure to support both decode and arbitration functions, improving productivity without proportionally increasing device complexity.

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

Data Source

PatentEP3235194B1Parallel direction decode circuits for network-on-chip
Publication Date: 2020.11.04 INTEL CORP
  • EP3235194B1 patent drawingFigure 1
  • EP3235194B1 patent drawingFigure 2
  • EP3235194B1 patent drawingFigure 3

AI summary

A first packet and a first direction associated with the first packet are received. The first packet is forwarded to an output port of a plurality of output ports of the first router based on the first direction associated with the first packet. A second direction associated with the first packet is determined. The second direction is based at least on an address of the first packet. The first packet and the second direction are forwarded through the output port of the first router to a second router.