Cold Storage Array Layout for Low-Power Data Redundancy
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Cold storage systems face challenges in minimizing power consumption and component wear while maintaining reasonable access times and data redundancy, especially when dealing with large volumes of rarely accessed data.
Innovation Solution
A redundant storage array scheme is employed, where data blocks are written sequentially to a single data device, and protection blocks are generated and stored on separate protection devices, allowing for efficient data recovery and reduced device activation during read/write operations, thus conserving power and reducing wear.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If data blocks are distributed across multiple storage devices in traditional RAID schemes, then data redundancy and fault tolerance are improved, but power consumption and component wear increase due to frequent access to multiple devices
Solution Approach 1:
The storage array is segmented into two distinct functional groups: data devices (k storage devices) that store data blocks, and protection devices (R storage devices) that store protection blocks. This segmentation allows the system to access only data devices during normal read operations, minimizing power consumption and wear, while protection devices remain inactive unless needed for data recovery.
Solution Approach 2:
The protection function is extracted from the data storage function. Instead of distributing protection blocks across all storage devices as in traditional RAID, the patent extracts protection storage to dedicated protection devices. This extraction enables independent operation of data devices during normal access, reducing unnecessary power consumption and component wear on protection devices.
2Reliability
If protection blocks are distributed across all storage devices in traditional RAID, then data recovery capability is improved, but device complexity and access time increase due to coordinated access requirements
Solution Approach 1:
The system is segmented into data devices and protection devices with clearly defined roles. Data devices handle normal read/write operations independently, while protection devices are dedicated to data recovery. This segmentation simplifies the control logic compared to traditional RAID schemes that require coordinated access to multiple devices for both normal operations and recovery.
Solution Approach 2:
Protection devices act as intermediaries specifically for data recovery operations. Instead of requiring direct coordinated access between data devices during recovery, the protection devices serve as intermediaries that store pre-computed protection blocks, simplifying the recovery process and reducing system complexity.
3Speed
If all storage devices remain active for fast access, then access time is reduced, but power consumption and component wear increase significantly
Solution Approach 1:
Storage devices operate in periodic cycles rather than continuously. Data devices remain active during normal read/write operations and can enter low-power states during idle periods. Protection devices remain inactive during normal operations and are activated only periodically when data recovery is needed, significantly reducing overall power consumption while maintaining fast access times.
Solution Approach 2:
Data devices are self-sufficient for normal read operations, requiring no assistance from protection devices. This self-service capability allows data devices to operate independently and efficiently during normal access, minimizing power consumption. Protection devices remain dormant unless data recovery is required, optimizing the balance between access speed and energy efficiency.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
A data storage apparatus comprising a plurality of data storage devices configured to store data blocks, and one or more protection devices configured to store protection blocks, wherein the data devices and the protection devices are associated by a plurality of stripes, wherein each stripe comprises a memory block on each data device or protection device, and wherein each protection block in a protection device comprises a value for reconstructing storage blocks in the same stripe, and a controller configured to select a data device, store data blocks sequentially to the memory blocks in the selected data devices, store protection blocks in the protection devices for each updated stripe, read data blocks from a selected data device, and reconstruct damaged storage devices. It operates a cold storage system with less power consumption, low component wear, and flexible in capacity expansion.