RDMA over 5G Wireless Network for Mobile Remote Memory Access
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current remote direct memory access (RDMA) technologies over traditional wireless networks face limitations in bandwidth and latency, making it difficult for mobile local computing devices to efficiently access and utilize large datasets and specialized hardware remotely, especially in applications requiring real-time data manipulation and dynamic reconfiguration.
Innovation Solution
Implementing RDMA over a 5G network, which provides high bandwidth and low latency, allowing mobile local computing devices to access remote memories for data and leverage local compute contexts, including specialized hardware and user configurations, enabling real-time data manipulation and dynamic reconfiguration of devices like multi-purpose industrial machines and expert systems.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Speed
If RDMA is implemented over traditional wireless networks, then data transfer between computing devices is enabled, but bandwidth is limited and latency is high
Solution Approach 1:
The patent changes the network parameter from traditional wireless networks to 5G wireless networks, which provides significantly higher bandwidth and lower latency. This parameter change enables RDMA operations to achieve performance levels comparable to wired connections, resolving the contradiction between transfer speed and network reliability.
2Adaptability or versatility
If large datasets are accessed from remote memory, then data availability increases, but access latency increases
Solution Approach 1:
The patent implements pre-fetching mechanisms where data is loaded into remote direct memory access (RDMA) memory buffers before it is actually needed by the application. This preliminary action allows the system to prepare data in advance over the high-bandwidth 5G connection, so that when the application needs the data, it is already available in the buffer, eliminating access latency.
3Productivity
If specialized hardware is co-located with data, then processing speed improves, but device portability and flexibility decrease
Solution Approach 1:
The patent introduces RDMA memory buffers as an intermediary between the specialized hardware and remote data sources. The hardware processes data in the local buffers at full speed, while the buffers themselves act as a mediator that can be rapidly repurposed by loading different datasets over the 5G network. This intermediary approach maintains processing speed while enabling flexibility.
4Adaptability or versatility
If devices are rapidly reconfigured for different purposes, then adaptability increases, but system complexity increases
Solution Approach 1:
The patent uses copying mechanisms where complete device configurations, including operating systems, applications, and data sets, are copied from remote storage to local RDMA memory buffers. This allows rapid reconfiguration by simply copying new configuration data over the 5G network without complex reconfiguration procedures, reducing system complexity while maintaining high adaptability.
Data Source
AI summary
A mobile local computing device is configured to access memories or storage devices associated with a remote computing device using remote direct memory access (RDMA) over a wireless fifth generation (5G) network link that provides high bandwidth and low latency relative to previous wireless network protocols. The mobile local computing device utilizes a local compute context that is unique to the local environment and which may be facilitated by devices, components, or functionalities that are local to the mobile local computing device, but which are not available with the same context to the remote computing device. The 5G network link supports high bandwidth and low latency so that the mobile local computing device can access and utilize the remote data in large datasets in a similar manner to how it would for locally stored data, while still being able to leverage the local I/O and maintain its unique local compute context.