Parallel Processing System Memory Broadcasting Architecture

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Current parallel processing systems face challenges in effectively managing data access among multiple processors, leading to traffic contention and reduced processing efficiency due to shared memory limitations.

Innovation Solution

A parallel processing system comprising a bus, multiple parallel processing processors, and shared memories connected via separate channels and a memory connection line, with a main processor controlling data broadcasting between shared memories to minimize contention and optimize data access.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Ease of operation

If multiple processors share a common memory system, then data access is simplified, but traffic contention increases and processing efficiency decreases

Engineering Contradiction:
Improvedata access simplicityVSAvoidprocessing efficiency
Core Design Contradiction:
Ease of operationVSProductivity

Solution Approach 1:

The patent divides the memory system into multiple independent shared memories (first shared memory, second shared memory, etc.), each connected to processors through separate channels. This segmentation eliminates traffic contention by providing dedicated memory resources for different processor groups, thereby maintaining processing efficiency while simplifying data access through localized memory allocation.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent introduces a memory connection line that directly interconnects shared memories, creating a new dimensional pathway for data transfer independent of the processor bus. This additional dimension allows memories to exchange data directly without contending for processor memory access, resolving the contradiction between access simplicity and processing efficiency.

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

2Speed

If data is broadcast between shared memories, then data access speed increases, but system complexity increases

Engineering Contradiction:
Improvedata access speedVSAvoidsystem complexity
Core Design Contradiction:
SpeedVSDevice complexity

Solution Approach 1:

The patent pre-establishes memory connection lines between shared memories during system configuration, so that data broadcasting pathways are ready in advance. This preliminary action enables fast data access through pre-configured direct connections while avoiding the complexity of dynamic routing logic, as the broadcast path is determined statically during system setup.

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The patent introduces a memory controller as an intermediary that manages data broadcasting between shared memories. The controller simplifies the broadcasting operation by automatically handling data routing and synchronization, thereby achieving high-speed data access without requiring complex coordination logic between processors and memories.

Inventive Principle:
Principle #24Intermediary (Mediator)

Data Source

PatentUS11237992B2Parallel processing system and operation method thereof
Publication Date: 2022.02.01 SAMSUNG ELECTRONICS CO LTD
  • US11237992B2 patent drawing
  • US11237992B2 patent drawing
  • US11237992B2 patent drawing

AI summary

Provided are a parallel processing system and an operation method thereof. The parallel processing system includes: a bus; a plurality of parallel processing processors; a plurality of shared memories connected to the bus via separate individual channels and connected to each other via a memory connection line; and a main processor configured to set a broadcasting state for the plurality of shared memories and control data stored in one shared memory among the plurality of shared memories to be broadcast to another shared memory via the memory connection line according to the broadcasting state.