Firmware Memory Migration Using CXL Pooled Memory During Updates
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional cloud computing systems lack a comprehensive computing logic and infrastructure to seamlessly update firmware for system memory without requiring server restarts, which is intrusive and disruptive, especially in environments where uninterrupted service is critical.
Innovation Solution
A firmware management system utilizing a firmware management engine that supports firmware management in cloud computing systems, employing hot-plugged memory (CXL pooled memory) and an orchestrator to migrate system memory data, update firmware, and manage virtual machines during the process without server restarts.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If firmware update is performed using conventional methods, then firmware can be updated, but server restart is required causing service interruption
Solution Approach 1:
The system segments memory into multiple banks (first memory bank and second memory bank) that can be independently updated. The firmware update is performed on one memory bank while the other remains operational, allowing service continuity without requiring full server restart.
Solution Approach 2:
The patent performs preliminary actions by copying data from the first memory bank to the second memory bank before updating the firmware of the first memory bank. This preparatory data migration ensures that operational data is preserved and can be restored if needed, enabling safe firmware updates without service interruption.
2Productivity
If memory training is performed to optimize performance, then memory parameters are optimized, but the process is intrusive and requires system restart
Solution Approach 1:
The system divides memory into separate banks that can undergo training independently. One memory bank can be trained and optimized while the other continues to operate, eliminating the need to interrupt service for memory optimization activities.
Solution Approach 2:
The patent uses a second memory bank as an intermediary to maintain operational continuity while the first memory bank undergoes training. The second bank serves as a buffer that allows the system to remain functional during the intrusive training process.
3Adaptability or versatility
If firmware update is performed, then new firmware features are activated, but service interruption occurs during the process
Solution Approach 1:
The system segments firmware update operations to different memory banks, allowing incremental updates where one bank is updated while the other provides continued service. This enables new firmware features to be activated without requiring complete service interruption.
Solution Approach 2:
The patent implements dynamic firmware updating where the system can switch between different memory banks containing different firmware versions. This dynamic approach allows the system to adapt and activate new features while maintaining service availability through bank switching capabilities.
Data Source
AI summary
Methods, systems, and devices for providing firmware management using a firmware management engine of a cloud computing system are described. The firmware management engine supports using a hot-plugged memory (e.g., Compute Express Link (CXL) pooled memory) to temporarily migrate system memory (e.g., Central Processing Unit (CPU) local memory or CXL attached memory) of a server node, while performing firmware management operations (e.g., a firmware update) on the server node. The hot-plugged memory can be CXL pooled memory that is a shared at a rack level, the CXL pooled memory can be used to store system memory data of system memory while updating and activating new memory initialization firmware code. A virtual machine associated with the server node stays operational temporarily using the hot-plugged memory. Firmware management is performed on a server node while virtual machines associated with the server node and the system memory of the server node remain operational.