RDMA Network Interface Data Encapsulation for Low-Latency Memory Access
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing remote direct memory access (RDMA) technologies suffer from high memory access latency due to the processor core's involvement in data encapsulation and maintenance of communication protocols, leading to increased overheads and resource consumption.
Innovation Solution
A processor system with an RDMA network interface card that encapsulates data independently, reducing the need for processor core involvement in data encapsulation and protocol maintenance, thereby optimizing memory access.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If the processor core assembles data based on a specified communication protocol, then data transmission can be achieved, but memory access latency increases and processor overhead increases
Solution Approach 1:
The patent extracts the data assembly and protocol maintenance functions from the processor core and transfers them to the RDMA network interface card. The NIC independently assembles data packets and maintains communication protocol states, allowing the processor core to simply issue memory access instructions without involvement in data preparation, thereby reducing memory access latency and processor overhead while maintaining reliable data transmission
Solution Approach 2:
The RDMA network interface card serves as an intermediary between the processor core and the network/memory system. It handles the complex tasks of data assembly, protocol maintenance, and memory management, enabling the processor core to communicate with remote memory through simple instructions without directly managing data transmission details, thus reducing latency and overhead
2Reliability
If the processor core assembles data and maintains communication protocols, then data transmission is achieved, but computing power is consumed and processor overhead increases
Solution Approach 1:
The patent extracts the computationally intensive functions of data assembly and protocol maintenance from the processor core and relocates them to the RDMA network interface card. The NIC performs these operations independently using its own processing resources, freeing the processor core from unnecessary computational overhead and reducing overall system energy consumption while maintaining reliable data transmission
Solution Approach 2:
The RDMA network interface card is designed to be self-sufficient, independently assembling data packets, maintaining communication protocol states, and managing memory operations without requiring continuous processor intervention. This self-service capability reduces processor computing power consumption while ensuring reliable data transmission through dedicated hardware functionality
Data Source
AI summary
Example data transmission methods, processor systems, and memory access systems are provided. One example processor system is applied to a source end device. The processor system includes a processor core and a first remote direct memory access (RDMA) network interface card. The processor core is configured to deliver a memory write instruction. The memory write instruction includes to-be-written data and a destination address of the to-be-written data. The destination address is a memory address of a destination end device. The first RDMA network interface card is configured to encapsulate the to-be-written data based on the destination address of the to-be-written data and configuration information of the destination end device and send the encapsulated to-be-written data to a second RDMA network interface card of the destination end device.