Coherent Memory Mapping in Configurable Logic Block Networks
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing hardware computing platforms for extensive computations, such as those in machine learning and artificial intelligence, are limited by resource constraints like bus bandwidth and network latency, necessitating improved data transfer and memory management.
Innovation Solution
Implementing a coherent shared memory system using configurable logic blocks with memory mapping units and advanced extensible interfaces to facilitate data transfer across various devices, reducing latency and optimizing performance.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Speed
If data is transferred across devices or networks using traditional bus or interconnect methods, then data transfer can be achieved, but latency increases and bandwidth is limited
Solution Approach 1:
The patent merges the memory spaces of multiple devices into a unified coherent memory system, allowing processing elements to access data from any device's memory without traditional data transfer protocols. This consolidation eliminates the need for separate data transfer operations across devices, directly reducing latency and increasing effective data transfer speed.
Solution Approach 2:
The patent introduces a coherent memory interface as an intermediary layer that manages data access across multiple devices. This interface handles memory address translation and data routing, allowing processing elements to access remote memory as if it were local, thereby reducing the latency associated with traditional bus-based data transfer.
2Productivity
If more memory resources are allocated to processing elements, then computational capacity increases, but device complexity increases
Solution Approach 1:
The coherent memory system provides a universal memory interface that can access memory resources from any device in the network. Instead of each processing element requiring dedicated local memory, the system allows any processing element to access any memory resource through the unified coherent memory interface, increasing computational capacity while avoiding the complexity of managing multiple separate memory systems.
Data Source
AI summary
Apparatuses and methods for managing a coherent memory are described. These may include one or more algorithmic logic units (ALUs) and an input/output (IO) interface. The I/O interface may receive one or more commands and retrieve data from or write data to a memory device. Each command may contain a memory address portion associated with a memory device. The apparatus may also include a memory mapping unit and a device controller. The memory mapping unit may map the memory address to a memory portion of the memory device, and the device controller may communicate with the memory device to retrieve data from or write data to the memory device. The apparatus may be implemented as a processing element in a configurable logic block network, which may additionally include a control logic unit that receives programming instructions from an application and generate the one or more commands based on the instructions.