Main Memory Bus Interfacing for Co-Processor I/O Bottlenecks
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current computer systems face communication bottlenecks due to the lower bandwidths supported by traditional I/O buses, which restrict the performance of co-processors and I/O devices that operate at faster speeds, necessitating a method to interface these devices through the main memory system for enhanced I/O capabilities.
Innovation Solution
The system connects co-processors and I/O devices directly to the main memory bus, allowing them to communicate at higher bandwidths, eliminating bottlenecks by using a TeraDIMM architecture that integrates non-volatile memory into the memory channel, enabling parallelism and scalability similar to CPU virtual memory systems.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If co-processors and I/O devices interface via traditional I/O bus, then device compatibility and ease of connection are maintained, but communication bandwidth and system performance are limited
Solution Approach 1:
The patent applies universality by enabling the main memory bus to serve dual functions: its traditional role in connecting CPU to memory, and a new role in interfacing co-processors and I/O devices. The memory controller is enhanced to handle both memory operations and CPIO device communications through the same bus infrastructure, eliminating the need for separate I/O bus connections and thereby improving I/O performance while reducing overall system complexity.
2Speed
If co-processors and I/O devices operate at higher speeds, then processing capability is improved, but communication bottlenecks occur due to lower I/O bus bandwidth
Solution Approach 1:
The patent introduces the main memory bus as an intermediary between high-speed CPIO devices and the CPU. Instead of direct I/O bus connections that create bottlenecks, data from fast co-processors and I/O devices is routed through the high-bandwidth memory bus, which acts as a mediator capable of handling higher data rates. This resolves the bottleneck problem by providing a communication path that matches the speed capabilities of modern CPIO devices.
3Adaptability or versatility
If separate I/O bus is used for CPIO devices, then device independence is maintained, but system bandwidth and parallelism are reduced
Solution Approach 1:
The patent merges the I/O communication path with the memory bus infrastructure. Co-processors and I/O devices are integrated into the memory subsystem architecture, allowing them to share the high-bandwidth memory bus with memory operations. This combining of previously separate pathways (memory bus and I/O bus) into a unified high-speed interface increases available data bandwidth while maintaining device functionality through the memory controller's ability to manage multiple device types.
Data Source
AI summary
A system for interfacing with a co-processor or input/output device is disclosed. According to one embodiment, the system includes a computer processing unit, a memory module, a memory bus that connects the computer processing unit and the memory module, and a co-processing unit or input/output device, wherein the memory bus also connects the co-processing unit or input/output device to the computer processing unit.


