Cache-Coherent NVM Backup for Rack-Level Power Loss Recovery
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing data backup solutions for cache-coherent interconnect nodes in data centers rely on node-level battery backups, leading to data integrity risks, architectural fragmentation, and increased total cost of ownership due to node-level failures.
Innovation Solution
Implementing a cache-coherent interconnect node-based non-volatile memory system with rack-level power loss triggers for data backup and recovery, using a centralized battery backup and memory pooling architecture that is independent of CPU architectures.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If node-level battery backup is used for data backup, then data backup capability is provided, but data integrity risks increase and reliability decreases due to node-level failures
Solution Approach 1:
The patent merges multiple node-level backup systems into a single centralized backup storage system managed by the cache-coherent interconnect node. This consolidation eliminates the reliability issues of node-level failures by providing a unified backup mechanism that is independent of individual compute node failures, thereby improving data integrity while reducing data loss risk.
Solution Approach 2:
The cache-coherent interconnect node serves multiple functions: it acts as a backup storage system, a power management hub, and a coordination point for multiple compute nodes. This multi-functionality eliminates the need for separate node-level backup systems, improving reliability by centralizing backup capabilities that can serve the entire rack regardless of which compute node fails.
2Reliability
If node-level battery backup is implemented, then data backup is achieved, but device complexity and total cost of ownership increase
Solution Approach 1:
The patent combines multiple scattered node-level backup implementations into a single centralized backup storage system. This merging reduces architectural fragmentation and device complexity by eliminating the need for separate backup systems at each compute node, while maintaining data backup capability across the entire rack.
Solution Approach 2:
The cache-coherent interconnect node is designed to perform multiple functions including backup storage, power management, and coordination for multiple compute nodes. This universal design simplifies the overall system architecture by replacing multiple specialized node-level systems with a single multi-functional platform, thereby reducing complexity and total cost of ownership.
3Reliability
If centralized rack-level backup is implemented, then reliability and data persistence are improved, but device complexity increases due to centralized power supply and storage management
Solution Approach 1:
The cache-coherent interconnect node automatically manages the centralized backup process without requiring manual intervention from compute nodes. The node self-manages power distribution to backup storage and coordinates data backup operations across multiple compute nodes, reducing the operational complexity of centralized management while maintaining high reliability and data persistence.
4Reliability
If rack-level power loss triggers backup, then data persistence across power events is ensured, but loss of time increases due to backup and recovery operations
Solution Approach 1:
The system performs preliminary actions by maintaining a ready state at the centralized backup storage and pre-configuring power management protocols. When rack-level power loss occurs, the backup process can be initiated immediately without delays, and the battery-powered backup storage is already prepared to receive data, reducing the overall time loss during backup and recovery operations while ensuring data persistence.
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
This approach ensures data persistence across power loss events, reduces total cost of ownership, and provides a scalable, unified solution that simplifies installation and maintenance across different compute node architectures.
Implementation Method 1
a backup power supply including power circuitry and a battery-based power source
Data Source
AI summary
Systems and methods are provided for implementing data backup and recovery using cache-coherent interconnect node-based non-volatile memory. A cache-coherent interconnect node partitions a memory pool into a plurality of memory regions as well as a backup storage into a plurality of memory portions, and pre-allocates a memory region and a corresponding memory portion to each compute node. When a rack-level power loss occurs, and a battery-based power source is activated, a cache-coherent interconnect controller saves data from each memory region into the corresponding memory portion, and subsequently saves an entry for each memory portion in an index portion of the backup storage. Subsequently, the controller causes a power circuitry to shut down the backup power source. After rack-level power restoration and memory region initialization, the controller restores, for each memory region, the data saved in a corresponding memory portion into that memory region, based on information in a corresponding entry.