Hybrid Memory Autonomous Save and Restore Operations
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Legacy memory systems require a host throughout save and restore operations, leading to prolonged exchange times that can exceed the time available before a catastrophic event, often resulting in failure to save volatile memory to non-volatile memory, thus failing to protect against such events.
Innovation Solution
A hybrid memory system with a non-volatile memory controller that can autonomously disconnect from the host, perform read/write operations between volatile and non-volatile memory, and execute save and restore operations independently, using a command buffer and programmable sequences to handle events like power-downs without host intervention.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If legacy protocols require host involvement throughout save and restore operations, then protocol compliance and host control are maintained, but operation time becomes too long to complete before catastrophic events
Solution Approach 1:
The memory system performs save and restore operations autonomously without requiring host involvement during the actual data transfer. The system detects catastrophic events, initiates save operations to non-volatile memory, and completes restore operations independently, allowing these critical functions to execute quickly without host protocol overhead while maintaining data protection reliability
Solution Approach 2:
The system prepares save and restore capabilities in advance by maintaining non-volatile memory as a ready backup target and pre-configuring the autonomous operation framework. This preliminary preparation enables the system to immediately execute save operations when catastrophic events are detected, eliminating the time-consuming host protocol setup phase
2Productivity
If the memory system autonomously disconnects from the host to perform save operations, then operation speed increases, but host control and coordination are reduced
Solution Approach 1:
The system segments control into two modes: host-controlled mode for normal operations and autonomous mode for catastrophic event responses. This segmentation allows the system to maintain ease of host control during routine operations while enabling high-speed autonomous save and restore operations when needed, resolving the contradiction between speed and control
3Reliability
If multiple protocol exchanges are required to initiate and complete save operations, then protocol compliance is ensured, but the time available before catastrophic events is exceeded
Solution Approach 1:
The system performs preliminary configuration and capability setup during host-controlled operations, so that when catastrophic events occur, the autonomous save operation can proceed with minimal protocol exchanges. The non-volatile memory target and transfer parameters are pre-established, allowing rapid execution that completes within the limited time window before catastrophic events
Data Source
AI summary
An apparatus forms a memory system that is physically populated into a host. In a powered-on state, the apparatus logically connects to the host through a host memory controller configured to receive host-initiated commands. The memory system includes a command buffer coupled to the host memory controller to receive the host-initiated commands. The memory system comprises both volatile memory (e.g., RAM) and non-volatile memory (e.g., FLASH). A non-volatile memory controller (NVC) is coupled to the volatile memory, and is also coupled to the non-volatile memory. A command sequence processor that is co-resident with the NVC responds to a trigger signal by logically disconnecting from the host, then dispatching command sequences that perform successive read/write operations between the volatile memory and the non-volatile memory. The successive read/write operations are performed even when the host is in a powered-down state.