Network Memory Link Architecture for High Read Rates

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Current network devices face challenges in handling high data rates and capacities without significantly increasing the cost and power consumption, as they require more access ports and higher clock speeds, which are costly and power-intensive.

Innovation Solution

A memory system for network devices that implements a distributed-linked list architecture, using a link memory, free-entry manager, and context manager to manage data elements, allowing for efficient data storage and retrieval using single-port memory with lower operating frequencies, thereby reducing costs and power consumption while maintaining performance.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Productivity

If the number of access ports of memory is increased to meet high data rate demands, then the data handling capacity is improved, but the cost and power consumption significantly increase

Engineering Contradiction:
Improvedata handling capacityVSAvoidpower consumption
Core Design Contradiction:
ProductivityVSUse of energy by stationary object

Solution Approach 1:

The patent segments the memory access operations by implementing separate read and write ports, allowing independent access paths. This segmentation enables the memory system to handle high data rates through efficient parallel operations without requiring a proportional increase in total access ports, thereby reducing power consumption while maintaining productivity.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent introduces the dimension of operational timing by implementing pipelined memory operations and asynchronous read/write cycles. This allows multiple operations to be in progress simultaneously at different stages, effectively increasing data handling capacity without linearly increasing the number of physical access ports required.

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

2Speed

If the clock speed of memory is increased to meet high data rate demands, then the data transfer rate is improved, but the cost and power consumption significantly increase

Engineering Contradiction:
Improvedata transfer rateVSAvoidpower consumption
Core Design Contradiction:
SpeedVSUse of energy by stationary object

Solution Approach 1:

The patent implements periodic memory access patterns with configurable cycle times, allowing the memory system to operate at lower clock speeds by efficiently utilizing periodic read and write operations. This reduces power consumption while maintaining effective data transfer rates through optimized timing and burst operations.

Inventive Principle:
Principle #19Periodic action

Solution Approach 2:

The patent ensures continuous useful action through overlapping read and write operations, where write operations can begin before previous reads complete. This pipelining approach maintains high data transfer rates without requiring proportionally higher clock speeds, thereby reducing power consumption while sustaining productivity.

Inventive Principle:
Principle #20Continuity of useful action

3Productivity

If the number of access ports of memory is increased to meet high capacity demands, then the memory access capability is improved, but the cost significantly increases

Engineering Contradiction:
Improvememory access capabilityVSAvoidcost
Core Design Contradiction:
ProductivityVSEase of manufacture

Solution Approach 1:

The patent implements universal memory control logic that can handle multiple types of access patterns (sequential, random, burst, pipelined) through a single set of access ports. This multi-functionality allows the memory system to achieve high access capability without requiring specialized ports for each operation type, thereby reducing manufacturing cost while maintaining productivity.

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

Solution Approach 2:

The patent introduces control logic and address translation mechanisms as intermediaries between the processor and memory ports. These intermediaries optimize memory access patterns and reduce the need for multiple dedicated access ports, enabling high memory access capability with fewer physical ports, thus reducing cost while maintaining ease of operation.

Inventive Principle:
Principle #24Intermediary (Mediator)

Data Source

PatentUS10740006B2System and method for enabling high read rates to data element lists
Publication Date: 2020.08.11 INNOVIUM INC
  • US10740006B2 patent drawing
  • US10740006B2 patent drawing
  • US10740006B2 patent drawing

AI summary

A memory system for a network device is described. The memory system includes a main memory configured to store one or more data elements. Further, the memory system includes a link memory that is configured to maintain one or more pointers to interconnect the one or more data elements stored in the main memory. The memory system also includes a free-entry manager that is configured to generate an available bank set including one or more locations in the link memory. In addition, the memory system includes a context manager that is configured to maintain metadata for a list of the one or more data elements.