Multi-Tier Storage Data Migration for Power Reduction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The increasing demand for memory usage in computer systems due to larger applications and data storage leads to higher power consumption and heat generation in storage sub-systems, reducing their lifespan and increasing costs.
Innovation Solution
A method involving data migration across multiple memory tiers, where data is initially stored in a first memory, then migrated to a second memory when full, and finally to a third memory when necessary, with the second memory configured to maintain a portion of the third memory in an inactive state to reduce power usage.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Quantity of substance
If data is stored in multiple memory devices to meet increasing storage demand, then storage capacity is improved, but power consumption increases
Solution Approach 1:
The storage system is segmented into multiple memory devices (first memory device, second memory device, third memory device) with different power characteristics. Data is distributed across these segmented storage units, allowing the system to scale capacity while selectively activating only the necessary portions to meet storage demands without powering all devices at full capacity.
Solution Approach 2:
The system changes the operational state parameter of memory devices between active and inactive states based on storage needs. When storage capacity needs to be increased, the system activates additional memory devices; when storage demand decreases, devices are transitioned to inactive states, thereby dynamically adjusting power consumption to match actual storage requirements.
2Quantity of substance
If more memory devices are activated to provide storage capacity, then storage availability is improved, but heat generation increases
Solution Approach 1:
By segmenting the storage system into multiple independent memory devices, the system can isolate heat generation to only those devices currently in active state. This segmentation allows thermal management to be applied locally to active devices rather than requiring cooling infrastructure for the entire storage subsystem, reducing overall heat generation while maintaining storage availability.
Solution Approach 2:
The system employs periodic monitoring of storage usage patterns and dynamically activates or deactivates memory devices based on current demands. This periodic adjustment ensures that storage availability is maintained when needed while minimizing the time that additional devices remain in high-power states, thereby reducing cumulative heat generation over time.
3Productivity
If storage sub-system operates continuously at high power, then performance is maintained, but device lifespan reduces
Solution Approach 1:
The storage sub-system transitions from a static high-power operational state to a dynamic state where memory devices can be switched between active and inactive modes. This dynamic operation allows the system to maintain high performance when storage demands require it, while extending device lifespan by periodically transitioning devices to lower-power states during periods of reduced demand, thereby reducing cumulative stress on the hardware.
Solution Approach 2:
The system changes operational parameters of memory devices by transitioning them between different power states (active, standby, inactive) based on storage workload. This parameter change allows performance to be maintained during high-demand periods while extending device lifespan during low-demand periods, optimizing the trade-off between productivity and durability over the long term.
Data Source
AI summary
The invention provides a method, apparatus and system for reducing power consumption involving data storage devices. One embodiment involves storing data in a first memory; in response to the first memory exceeding a first threshold, migrating the data from the first memory to a second memory; in response to the second memory exceeding a second threshold, then activating a third memory if the third memory is in active; and in response to the second memory exceeding a third threshold greater than the second threshold, migrating the data from the second memory to a third memory; wherein the second memory is sized and configured to store data targeted for the third memory to intelligently maintain a portion of the third memory in an inactive state.