Dual Cast Mirroring for Volatile Memory Failover
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing data storage systems with mirroring techniques require time and processing resources to read and write data to mirrored volatile memory, as they typically involve reading from one memory and writing to another, which can lead to increased latency and resource utilization during failover scenarios.
Innovation Solution
Implementing dual casting logic in storage systems to create and send simultaneous copies of data to both volatile memories before writing, thereby avoiding the need to read from either memory and reducing latency and resource usage.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If traditional mirroring reads from one volatile memory and writes to another, then data consistency is maintained, but processing time and resource utilization increase
Solution Approach 1:
The system pre-establishes dual cast addresses in both volatile memories before receiving write operations. When data needs to be mirrored, the system directly writes to both pre-configured addresses simultaneously, eliminating the need to read from one memory and write to another sequentially. This preliminary address preparation enables faster mirroring operations while maintaining data consistency.
Solution Approach 2:
Instead of reading from one volatile memory and writing to another, the system uses dual casting to create simultaneous copies of data in both volatile memories. The dual cast logic generates two separate write operations that target different volatile memories at the same time, achieving mirroring without the sequential read-write bottleneck of traditional approaches.
2Reliability
If traditional mirroring reads from one volatile memory and writes to another, then data consistency is maintained, but processing resources increase
Solution Approach 1:
The system pre-establishes dual cast addresses in both volatile memories before receiving write operations. When data needs to be mirrored, the system directly writes to both pre-configured addresses simultaneously, eliminating the need to read from one memory and write to another sequentially. This preliminary address preparation enables faster mirroring operations while maintaining data consistency.
Solution Approach 2:
Instead of reading from one volatile memory and writing to another, the system uses dual casting to create simultaneous copies of data in both volatile memories. The dual cast logic generates two separate write operations that target different volatile memories at the same time, achieving mirroring without the sequential read-write bottleneck of traditional approaches.
3Speed
If dual casting writes to both volatile memories simultaneously, then failover latency is reduced, but device complexity increases
Solution Approach 1:
The dual cast logic is integrated into the existing compute node architecture, allowing the same hardware components to handle both single writes and dual cast writes. The logic reuses existing memory controllers and address translation mechanisms, so the complexity increase is minimized while still achieving simultaneous writes to both volatile memories for reduced failover latency.
Data Source
AI summary
Mirrored volatile memory in a storage system is configured with a dual cast region of addresses. Buffers in the dual cast region are allocated for data associated with a received Write IO. A host IO device associates the dual cast addresses with the data. A switch or CPU complex recognizes the dual cast addresses associated with the data and, in response, creates and sends a first copy of the data to a first volatile memory mirror and creates and sends a second copy of the data to a second volatile memory mirror. The second copy may be sent via PCIe NTB between switches or CPU complexes.