Live VM Migration for Uncorrectable Memory Errors
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Cloud computing systems face significant downtime and data integrity issues due to uncorrectable memory errors, which can lead to the shutdown of entire host machines and affect thousands of virtual machines, as existing technologies lack effective methods for containment and recovery of such errors.
Innovation Solution
The method involves detecting uncorrectable memory errors, identifying and isolating poisoned memory pages, pausing operations of affected virtual machines, transferring unpoisoned memory pages to a target host, and resuming operations, while excluding poisoned pages from checksum calculations to prevent error propagation and ensure data integrity during live migration.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If uncorrectable memory errors are detected on a host machine, then the host machine shuts down to prevent error propagation, but this causes abrupt termination of all hosted virtual machines and results in lengthy downtime for reestablishment
Solution Approach 1:
The patent segments the memory space into individual memory pages and identifies only the specific poisoned pages associated with uncorrectable errors, rather than shutting down the entire host. This allows unaffected memory pages and virtual machines to continue operating, thereby maintaining service availability while ensuring data integrity by isolating only the corrupted segments.
Solution Approach 2:
The patent extracts and isolates the poisoned memory pages from the host system's address space, removing them from active use while allowing the rest of the system to continue functioning. This extraction enables the host to remain operational with reduced downtime while preventing error propagation through targeted isolation of only the affected memory segments.
2Reliability
If all memory pages are migrated to a target host to ensure data integrity, then data integrity is preserved, but the migration time and system downtime increase significantly
Solution Approach 1:
The patent applies local quality by treating different memory pages differently based on their integrity status. Unpoisoned pages are migrated to the target host to maintain data integrity, while poisoned pages are identified and excluded from migration. This selective approach ensures that only necessary data is transferred, significantly reducing migration time while preserving integrity of unaffected data.
Solution Approach 2:
The patent performs partial migration by transferring only the unpoisoned memory pages to the target host rather than migrating all memory pages. This partial action is sufficient to maintain service continuity and data integrity for unaffected virtual machines, thereby reducing migration time while avoiding the excessive overhead of transferring corrupted data that would require additional handling.
3Reliability
If virtual machines are paused during memory page transfer to ensure data consistency, then data consistency is maintained, but the operational availability of virtual machines is reduced
Solution Approach 1:
The patent performs preliminary identification and isolation of poisoned memory pages before initiating the migration process. By pre-marking the poisoned pages and configuring the system to exclude them from migration, the patent enables unpoisoned pages to be transferred without pausing virtual machine operations, thereby maintaining operational availability while ensuring data consistency through advance error isolation.
Solution Approach 2:
The patent introduces a memory management intermediary layer that handles the identification, isolation, and selective migration of memory pages. This intermediary mechanism allows the system to transfer unpoisoned pages to the target host while maintaining virtual machine operational continuity, as the intermediary manages the complexity of selective page transfer without requiring full system pausing.
Data Source
AI summary
The disclosed technology provides techniques, systems, and apparatus for containing and recovering from uncorrectable memory errors in distributed computing environment through migration of virtual machines and associated memory to a target host machine. An aspect of the disclosed technology includes a hypervisor or virtual machine manager that receives signaling of an uncorrectable memory error detected by a host machine. The virtual machine manager then uses information received via the signaling to identify virtual memory addresses or memory pages associated with the corrupted memory element so as to allow for containment and recovery from the error, and for live migration of the virtual machine.