Connection Device for Cross-Cluster Memory Address Translation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The challenge of adapting ultra-low latency communication protocols, such as CXL or UB, to the RDMA protocol in data centers to facilitate efficient memory access across computing clusters is unresolved.
Innovation Solution
A data processing system and method that includes a connection device managing memory address information across computing clusters, enabling computing devices to access memory spaces in other clusters using ultra-low latency communication protocols, and employing RDMA for inter-cluster networking.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Speed
If ultra-low latency communication protocol is used within a computing cluster, then memory access speed is improved, but compatibility with RDMA protocol across computing clusters deteriorates
Solution Approach 1:
The connection device serves as an intermediary between computing clusters, managing memory address information and enabling address translation between ultra-low latency communication protocol and RDMA protocol. This mediator allows computing devices to access memory across clusters without direct protocol compatibility issues
Solution Approach 2:
The system segments memory address management by introducing a connection device that independently manages memory address information for each computing cluster. This segmentation allows each cluster to use its own ultra-low latency protocol internally while maintaining RDMA compatibility at the inter-cluster level
2Adaptability or versatility
If computing devices access memory spaces across multiple computing clusters, then system flexibility is improved, but memory address management complexity increases
Solution Approach 1:
The connection device acts as a mediator that centralizes memory address management. It receives memory address information from computing devices, manages the address mappings, and facilitates access across clusters, thereby reducing the complexity burden on individual devices
Solution Approach 2:
The connection device performs multiple functions including managing memory address information, translating addresses between different protocols, and coordinating access across multiple computing clusters. This multi-functionality consolidates complexity into a single component
Data Source
AI summary
An example data processing system includes a first computing cluster, a second computing cluster, and a connection device. The first computing cluster includes a first computing device, and the second computing cluster includes a second computing device, and the connection device is connected to the first computing device. The connection device is configured to manage memory address information of the first computing cluster provided by the first computing device and memory address information of the second computing cluster provided by the second computing device. The first computing device is configured to receive an access request, where the access request is used to access memory space of the second computing device, search the memory address information of the second computing cluster managed by the connection device for an address of the memory space, and access the memory space of the second computing device based on the address.