Variable Sized Memory Tables Power Reduction

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Network devices with large forwarding tables experience excessive power consumption due to unused memory, leading to increased operating expenses and heat generation, especially when implemented in smaller network infrastructures where not all memory spaces are utilized.

Innovation Solution

Implementing a dynamic power management system that organizes memory into power domains, allowing for the reduction of power to unused memory sections by using switches to control the powered state of memory entries, and optimizing data distribution across these domains to reduce overall power draw.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Adaptability or versatility

If a fixed memory size is implemented on a singular integrated circuit to support large numbers of network devices, then the device can accommodate various network scales, but a large portion of memory remains unused causing excessive power draw

Engineering Contradiction:
Improveadaptability to various network device countsVSAvoidpower consumption
Core Design Contradiction:
Adaptability or versatilityVSUse of energy by moving object

Solution Approach 1:

The memory is divided into multiple power domains, each independently controllable. This segmentation allows the system to activate only the memory sections corresponding to the actual number of connected network devices, rather than keeping the entire fixed memory powered on. The patent implements this by organizing memory into discrete power domains that can be selectively enabled or disabled based on usage requirements.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The system dynamically adjusts power consumption by changing the powered state of different memory sections based on the actual number of connected network devices. The controller monitors device count and dynamically enables or disables corresponding power domains, transforming the static power consumption model into a dynamic one that adapts to real-time usage conditions.

Inventive Principle:
Principle #15Dynamics

2Reliability

If a large fixed memory size is allocated for forwarding tables, then the network device can handle maximum network scale, but power consumption increases due to unused memory sections

Engineering Contradiction:
Improvecapability to handle network trafficVSAvoidenergy waste
Core Design Contradiction:
ReliabilityVSLoss of energy

Solution Approach 1:

Different sections of memory are assigned different power states based on their usage. The patent implements local quality by enabling power domains corresponding to active network devices while keeping other domains in a low-power or disabled state. This creates a non-uniform power distribution across the memory architecture, optimizing energy efficiency while maintaining necessary functionality.

Inventive Principle:
Principle #3Local quality

Solution Approach 2:

The system changes the operational parameters of memory sections by adjusting their power states based on usage. The controller modifies parameters such as voltage and power supply to memory domains dynamically, transitioning them between active, standby, and powered-off states according to the actual number of connected devices, thereby reducing energy waste.

Inventive Principle:
Principle #35Parameter changes

3Speed

If all memory sections are kept powered to ensure immediate data access, then data retrieval speed is maintained, but power consumption increases

Engineering Contradiction:
Improvedata retrieval speedVSAvoidpower draw
Core Design Contradiction:
SpeedVSUse of energy by moving object

Solution Approach 1:

The system employs periodic monitoring and dynamic power state transitions for memory domains. Rather than keeping all memory permanently powered, the controller periodically assesses usage patterns and adjusts power states accordingly, enabling memory sections only when needed and disabling them when not in use, thus balancing speed requirements with power conservation.

Inventive Principle:
Principle #19Periodic action

Data Source

PatentUS11126249B1Power reduction methods for variable sized tables
Publication Date: 2021.09.21 AMAZON TECH INC
  • US11126249B1 patent drawing
  • US11126249B1 patent drawing
  • US11126249B1 patent drawing

AI summary

Disclosed are devices, systems, and methods for the use of memory including a data table configured to store a plurality of elements, wherein the plurality of elements are arranged into a plurality of buckets and each of the plurality of buckets comprising a plurality of entries. A first power domain can be associated with an entry of each bucket or with a first bucket. A second power domain can be associated with a second entry of each bucket or a second bucket. Processing logic can be configured to search for a particular value stored in an element of the plurality of elements by selecting buckets of the plurality of buckets and selecting at least one entry of each of the buckets. A programmable register can be used to select a powered state of the second power domain based on a configuration of the programmable register.