Communications Endpoint Processor Parallel Frame Transmission
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The existing communication systems between multi-processor computing systems and communication networks face a bottleneck due to the mismatch between the computational domain's functional richness and the communications domain's focus on simplicity, throughput, and low latency, leading to inefficiencies in data transmission.
Innovation Solution
A method and system that segment data packets into fragments, generate fragment descriptors, and store them in non-contiguous memory locations, allowing for simultaneous transmission and reassembly, utilizing a Communications Endpoint Processor (CEP) kernel and engine to optimize data transfer across the interface and communications medium.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If data packets are transmitted using conventional Ethernet NICs with DMA controllers, then data transmission can be performed, but only one frame can be sent at a time causing performance bottleneck
Solution Approach 1:
The patent segments data packets into multiple frames that can be transmitted simultaneously through different communication paths. The data is divided into first data and second data portions, each with associated headers, allowing parallel transmission to overcome the single-frame limitation of conventional DMA controllers and improve throughput.
2Adaptability or versatility
If the computational domain uses software-based interfaces for functional richness, then operational flexibility is achieved, but performance bottleneck occurs when interfacing with hardware-based communication domain
Solution Approach 1:
The patent introduces a communications endpoint processor (CEP) as an intermediary between the computational domain and communication network. The CEP includes a kernel for software-based functional richness and a hardware engine for performance-optimized data transfer, bridging the gap between software flexibility and hardware speed.
Solution Approach 2:
The patent replaces conventional software-based data transmission mechanisms with a hybrid system that uses hardware acceleration. The CEP engine performs hardware-level operations such as frame assembly, descriptor management, and data transfer, substituting mechanical software processing with faster hardware mechanisms while maintaining software control through the CEP kernel.
3Adaptability or versatility
If fragment descriptors are stored in non-contiguous memory portions, then data transmission flexibility is improved, but memory access complexity increases
Solution Approach 1:
The patent prepares fragment descriptors in advance, storing metadata about data fragments in a descriptor table before actual transmission. This preliminary organization of descriptor information enables flexible non-contiguous memory access during transmission while simplifying the retrieval process through pre-established descriptor structures.
Solution Approach 2:
The patent uses descriptor tables to create indirect references to actual data in memory. Instead of directly managing complex non-contiguous data structures, the system copies essential metadata (descriptors) that point to the actual data locations, simplifying memory management while maintaining transmission flexibility.
Data Source
AI summary
Embodiments of the present invention include enhanced functionalities and components within a Communication Endpoint Processor (CEP) that act as an interface between computational and communications domains. The embodiments disclosed herein deliver a complete memory mapped high performance interface that has the ability to support the simultaneous transmission of multiple frames of multiple sizes, and that has the ability to interrupt the transmission of lower priority frames in order to send higher priority frames.