DRuM Standby Slice Services Prevent Cascading Failures
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
During the upgrade of a storage node in a cluster, the existing data protection systems face challenges in maintaining continuous data protection, as the node is taken offline, leaving only one copy of the data available, making the system vulnerable to data loss in case of failure, and the data migration process is costly in terms of time and bandwidth.
Innovation Solution
The degraded redundant metadata (DRuM) technique ensures continuous data protection by maintaining multiple copies of data across storage nodes during upgrades, using standby slice services to redirect and rebuild data, ensuring no data loss by activating additional standby services in case of failures, and using Index Incarnation Numbers for chronological ordering.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If a storage node is taken offline for upgrade, then the node can be updated to improve system capabilities, but data protection is compromised as only one copy of data remains available
Solution Approach 1:
The system pre-activates standby slice services before the primary storage node is taken offline for upgrade. This preliminary action ensures that redundant data copies are already in place and operational before the upgrade begins, maintaining data protection throughout the upgrade process without requiring data migration during the offline period.
Solution Approach 2:
Standby slice services act as intermediary components that take over data protection functions when the primary storage node is upgraded. These standby services receive and store redundant data copies, serving as a bridge that maintains data protection continuity during the transition period when the primary node is unavailable.
2Reliability
If data is migrated from the storage node being upgraded to another node, then data protection can be maintained, but time and bandwidth are substantially consumed
Solution Approach 1:
Instead of migrating data during the upgrade process, the system performs preliminary actions by activating standby slice services that already have the capability to store and manage data copies. This eliminates the need for time-consuming data migration during the upgrade window.
Solution Approach 2:
The system uses copying mechanisms where standby slice services create and maintain copies of data on standby nodes. This copying approach is more efficient than migration because it operates in parallel and doesn't require the source node to remain online, thus reducing upgrade time and bandwidth consumption.
3Reliability
If data is migrated during node upgrade, then data protection is maintained, but the upgrade process becomes more complex
Solution Approach 1:
The invention extracts the data protection function from the primary storage node during upgrade by activating independent standby slice services. This separation allows the primary node to be upgraded without complex coordination of data migration, as the standby services independently handle data protection.
Solution Approach 2:
Standby slice services serve as intermediaries that simplify the upgrade process by automatically taking over data protection responsibilities. This intermediary mechanism eliminates the need for complex manual data migration procedures and reduces upgrade process complexity.
4Reliability
If standby slice services are activated to redirect subsequent data, then cascading failures are handled, but system complexity increases
Solution Approach 1:
The system implements dynamic service activation where standby slice services are automatically promoted to active status when failures occur. This dynamic behavior allows the system to adapt to cascading failures without requiring complex manual intervention or pre-configuration of multiple standby hierarchies.
Solution Approach 2:
The system uses feedback mechanisms to monitor the status of storage nodes and automatically activate standby slice services when failures are detected. This feedback-driven automation handles cascading failures efficiently without increasing operational complexity, as the system self-regulates based on real-time status information.
Data Source
AI summary
A technique is configured to maintain multiple copies of data served by storage nodes of a cluster during upgrade of a storage node to ensure continuous protection of the data served by the nodes. The data is logically organized as one or more volumes on storage devices of the cluster and includes metadata that describe the data of each volume. A data protection system may be configured to maintain at least two copies of the data in the cluster during upgrade to a storage node that is assigned to host one of the copies of the data but that is taken offline during the upgrade. As a result, an original slice service of the node may be rendered unavailable during the upgrade. In response, the technique redirects replicated data targeted to the original slice service to a standby pool of slice services in accordance with a degraded redundant metadata service of the cluster. In the event the standby slice service itself subsequently becomes unavailable, another standby slice service from the standby pool is activated to receive the subsequent data. In this manner, cascading failure of secondary slice slices is handled.