Free Index Flop Array Segmentation for Selective Clock Gating

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Traditional free index flop array designs consume unnecessary power due to a single clock gating signal activating the entire array for every access, leading to inefficient power utilization.

Innovation Solution

Divide the free index flop array into subgroups, each controlled by separate clock gating signals, allowing only necessary subgroups to be activated during memory access, thereby optimizing power consumption.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Device complexity

If a single clock gating signal is used for the entire free index flop array, then the control circuit is simple, but power consumption is high because the entire array is activated for every access

Engineering Contradiction:
Improvecontrol circuit complexityVSAvoidpower consumption
Core Design Contradiction:
Device complexityVSUse of energy by moving object

Solution Approach 1:

The free index flop array is divided into multiple subgroups, with each subgroup controlled by its own dedicated clock gating signal. This segmentation allows selective activation of only the necessary subgroups during memory access operations, preventing unnecessary power consumption while maintaining reasonable control circuit complexity through modular organization.

Inventive Principle:
Principle #1Segmentation

2Use of energy by moving object

If the free index flop array is divided into subgroups with separate clock gating signals, then power consumption is reduced, but the control circuit becomes more complex

Engineering Contradiction:
Improvepower consumptionVSAvoidcontrol circuit complexity
Core Design Contradiction:
Use of energy by moving objectVSDevice complexity

Solution Approach 1:

The array is segmented into subgroups that can be independently controlled, enabling power savings by activating only necessary portions. The complexity increase is managed through systematic organization of multiple simple clock gating units rather than one complex control mechanism.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

Instead of activating the entire flop array for every access, only the specific subgroups that contain relevant data are activated. This partial action approach reduces power consumption by avoiding unnecessary activation of unrelated subgroups, while the control overhead remains proportional to the number of subgroups rather than the total array size.

Inventive Principle:
Principle #16Partial or excessive action

3Area of stationary object

If traditional free index logic is used with a single clock gating signal, then the area utilization is efficient, but power utilization is inefficient due to unnecessary activation

Engineering Contradiction:
Improvearea utilizationVSAvoidpower waste
Core Design Contradiction:
Area of stationary objectVSLoss of energy

Solution Approach 1:

The flop array is divided into subgroups with dedicated clock gating control. This segmentation enables selective activation of only the subgroups that need to be accessed, reducing power waste while maintaining efficient area utilization through compact subgroup organization and shared control logic where applicable.

Inventive Principle:
Principle #1Segmentation

Data Source

PatentUS20250316300A1Clock gating of free index flop arrays
Publication Date: 2025.10.09 NVIDIA CORP
  • US20250316300A1 patent drawing
  • US20250316300A1 patent drawing
  • US20250316300A1 patent drawing

AI summary

A system is described to include a first storage element having a plurality of data storage locations, a second storage element having an array of bits indicating an availability for a corresponding data storage location in the plurality of data storage locations, and a control circuit. The control circuit may include at least a first switching element coupled to a first set of bits in the array of bits, the at least a first switching element selectively enables the first set of bits to be read or written in response to being activated by a first clock gating signal as well as at least a second switching element coupled to a second set of bits in the array of bits, the at least a second switching element selectively enables the second set of bits to be read or written in response to being activated by a second clock gating signal.