Accelerator Card Firmware Update Without Host Reboot
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Traditional methods for updating the firmware of accelerator cards require rebooting the host system, which is undesirable due to lost computing time and system downtime, especially in data centers with uptime requirements, and affects all expansion cards even if only a subset needs updating.
Innovation Solution
The implementation of a warm boot mechanism that allows the firmware of selected accelerator cards to be updated without rebooting the host system by using a warm boot controller to load a new platform image from non-volatile memory into the integrated circuit while maintaining communication with the host system.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If traditional firmware update methods are used, then the accelerator card firmware can be updated, but the host system must be rebooted causing system downtime and lost computing time
Solution Approach 1:
The system separates the firmware update process into independent segments: the host system continues operating while the accelerator card undergoes firmware updates independently through a warm boot sequence initiated by the host, eliminating the need for full system reboot and allowing continuous operation during updates
Solution Approach 2:
The host system performs preliminary actions by setting update flags and preparing new firmware images before initiating the warm boot sequence, allowing the accelerator card to load and execute new firmware while the host system remains operational, thus preventing downtime
2Reliability
If the host system is rebooted to update firmware, then all expansion cards can be updated, but other expansion cards that do not need updating are also affected and must be rebooted
Solution Approach 1:
The update mechanism is segmented to target individual accelerator cards specifically through unique identification and flag setting, allowing only the intended card to undergo warm boot and firmware update while other expansion cards continue operating normally without interruption
Solution Approach 2:
The accelerator card performs self-service updates by detecting the update flag, autonomously initiating the warm boot sequence, loading new firmware from its own non-volatile memory, and completing the update process independently without requiring host system reboot or affecting other cards
Data Source
AI summary
Changing accelerator card images without rebooting a host system includes receiving, within an integrated circuit (IC) of an accelerator card, an address of a platform image stored in a non-volatile memory of the accelerator card. The address is received over a communication link between the host system and the accelerator card while the communication link is connected. Changing accelerator card images includes detecting, within a register of the IC, that a warm boot enable flag is set and that the communication link with the host system is disconnected. In response to detecting that the warm boot enable flag is set and that the communication link is disconnected, loading of the platform image from the address of the non-volatile memory into the integrated circuit is initiated.