Integrated Parcel-to-Drive Mapping for Incremental RAID Expansion
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing storage systems face challenges in efficiently adding drives in increments of one while maintaining drive box failure resistance, necessitating complex and time-consuming configuration changes between distributed RAID and drive box failure resistance functions.
Innovation Solution
A storage system with integrated maps that manage correspondence relations between parcels and drives, allowing seamless addition or removal of drives while ensuring drive box failure resistance, by using a single map that combines both functions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of manufacture
If distributed RAID is implemented to enable drive addition in increments of one, then investment efficiency is improved, but drive box failure resistance cannot be guaranteed
Solution Approach 1:
The patent implements dynamic configuration where the storage system can transition between different RAID levels based on the number of drives. When the number of drives is a multiple of the stripe width, the system automatically configures as RAID 5 for failure resistance. When drives are added individually, the system dynamically adjusts to accommodate the new drive count, maintaining flexibility while ensuring failure resistance when conditions permit.
Solution Approach 2:
The system changes the RAID configuration parameter based on the number of drives present. The control unit determines whether to operate in RAID 5 mode (when drive count is divisible by stripe width) or flexible addition mode (when drives can be added individually), allowing the system to adapt its parameters to match the current hardware state and achieve both failure resistance and flexibility as needed.
2Reliability
If drive box failure resistance function is enabled requiring same number of drives in all drive boxes, then reliability is improved, but drive addition flexibility is reduced
Solution Approach 1:
The system dynamically adjusts the RAID configuration based on the number of drives. When drives are added and the total count allows for equal distribution across drive boxes, the system transitions to RAID 5 configuration for failure resistance. When drives cannot be evenly distributed, the system maintains flexible addition mode, allowing the configuration to change adaptively rather than being locked into a single mode.
3Reliability
If configuration changes are made to enable drive box failure resistance function, then reliability is improved, but complexity and time of configuration changes increase
Solution Approach 1:
The control unit automatically determines the appropriate RAID configuration based on the number of drives present in the storage system. When the drive count is a multiple of the stripe width, the system automatically configures as RAID 5 without requiring manual intervention. This self-service approach eliminates complex manual configuration changes while ensuring failure resistance is enabled when conditions permit.
Solution Approach 2:
The system prepares multiple configuration states in advance (RAID 5 mode and flexible addition mode) and automatically selects the appropriate one based on the current drive count. This preliminary preparation of configuration options eliminates the need for complex real-time configuration changes, reducing both complexity and time required to enable failure resistance functionality.
Data Source
AI summary
A storage system manages parcels of a parcel group including parcels of user data and redundant data, and includes a storage device that stores an integrated map including showing a correspondence relation between a parcel and a drive. A first map corresponding to a first count for which the number of drives in each drive box is the same has a failure resistance correspondence relation having failure resistance by which the user data is not lost even if a failure occurs in the drive boxes of a predetermined number or less. The first map is configured to have a relation that is implemented by moving data of one parcel to a drive box in which a drive is added, from a storage state of the parcel in the drive according to a second map corresponding to a count of the drive less than the first drive by one.