Boot Error Record for Server Context Migration
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Cloud computing servers often experience failures due to software or hardware issues, making it difficult to perform error mitigation actions like sending error context for analysis or migrating user context to another server, as the failure can render the input/output subsystem in a reduced functionality mode, hindering data transfer.
Innovation Solution
Implementing a method that allows error mitigation actions by detecting errors during a warm reboot in self-refresh mode, creating a boot error record with additional action identifiers, and making it available to the operating system to perform actions such as sending error context to an analysis system or migrating user context to another server.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If a failure occurs in a computing device, then error mitigation actions are needed, but the input/output subsystem enters reduced functionality mode preventing data transfer
Solution Approach 1:
The patent performs preliminary actions by detecting errors during the boot sequence before the input/output subsystem fully initializes and enters reduced functionality mode. The boot error record is created and populated with error context and mitigation actions during this early window when the system is still operational, enabling error mitigation to be prepared in advance before the harmful reduced functionality state takes effect.
2Productivity
If error context is sent for analysis or user context is migrated, then service continuity is improved, but data transfer is blocked by reduced functionality mode
Solution Approach 1:
The patent executes data transfer operations preliminarily during the boot sequence when the input/output subsystem is still functional. Error context is sent to analysis systems and user context is migrated to other servers before the failure fully manifests and blocks data transfer. This preliminary execution of mitigation actions ensures service continuity can be restored even though the subsystem later enters reduced functionality mode.
Solution Approach 2:
The boot error record serves as an intermediary data structure that captures and preserves error context, action identifiers, and mitigation information during the boot sequence. This intermediary mechanism allows error information to be stored and processed even when the input/output subsystem later becomes impaired, bridging the gap between error detection and subsequent mitigation actions.
3Reliability
If the boot sequence is used to detect errors and create error records, then error mitigation is enabled, but memory must remain in self-refresh mode limiting operations
Solution Approach 1:
The patent takes advantage of the self-refresh mode window during the boot sequence to perform preliminary error detection and record creation. Memory remains in self-refresh mode to preserve its state, but the firmware is able to detect error indications and populate the boot error record with necessary information. This preliminary error detection occurs within the constraints of self-refresh mode, enabling reliability improvements without requiring memory to exit its low-power state.
Solution Approach 2:
The memory subsystem serves itself by maintaining its own refresh operations in self-refresh mode while simultaneously allowing the firmware to detect errors and create records during the boot sequence. The memory continues its self-service refresh function without external intervention, yet the system is able to perform error detection and mitigation preparation during this period, demonstrating self-service operation with added error handling capability.
Data Source
AI summary
A computing device, such as a server in a cloud computing system, can be configured to enable error mitigation actions to be performed when the computing device experiences a failure. The computing device includes firmware that can be configured to detect an error indication during a boot sequence of the server, determine at least one desired error mitigation action based at least in part on the error indication, and create a boot error record that identifies the at least one desired error mitigation action. The computing device also includes an operating system that can be configured to obtain the boot error record during the boot sequence and cause the at least one desired error mitigation action that is identified in the boot error record to be performed.