Distributed Object Storage Format Conversion via Disk Evacuation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In large distributed data storage systems, completely reformatting disk clusters to enable or disable encryption, or to convert the format, requires significant time and system resources due to the need for multiple write operations to transfer data across disks.
Innovation Solution
A computerized method that identifies the emptiest disks in a disk cluster, evacuates data from these disks to other disks, reforms them according to a format conversion instruction, and then writes converted data objects back to the reformatted disks, thereby reducing the number of write operations required.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If traditional complete reformatting method is used, then format conversion is achieved, but significant time and system resources are consumed due to multiple write operations
Solution Approach 1:
The patent divides the disk cluster into multiple groups and processes format conversion in segments rather than all at once. By evacuating data from one group of disks, reformatting that group, and then writing converted data back, the system achieves format conversion through multiple smaller write operations instead of one large operation, improving efficiency and reducing total time.
Solution Approach 2:
The patent performs preliminary data evacuation from selected disks before reformatting. By proactively moving data out of the way and creating empty space on target disks, the system prepares the necessary conditions for efficient reformatting and subsequent data writing, avoiding the need for multiple read-write cycles during the conversion process.
2Ease of manufacture
If traditional reformatting method is used, then format conversion is completed, but large quantities of write operations are required transferring data multiple times
Solution Approach 1:
The patent uses an intermediary approach by selecting specific disks as target disks to receive evacuated data during reformatting. These intermediary disks serve as temporary storage locations, allowing the system to perform reformatting on source disks without requiring data to be written back multiple times. The converted data is written once to the intermediary target disks, significantly reducing the total number of write operations.
3Reliability
If complete reformatting is performed on all disks, then format conversion is achieved, but system resources are significantly expended
Solution Approach 1:
The patent applies local quality by performing reformatting operations on specific groups of disks rather than uniformly processing all disks in the cluster. By identifying and processing subsets of disks that need conversion while leaving others unchanged or processing them in subsequent batches, the system achieves necessary format conversion with reduced system resource expenditure, focusing computational power only where needed.
Data Source
AI summary
The disclosure herein describes converting a disk cluster to a different format. A format conversion instruction associated with a disk cluster is received. A first subgroup of disks of the disk cluster that are the emptiest disks of the disk cluster are identified and all data is evacuated from the first subgroup of disks to other disks of the disk cluster. The first subgroup of disks is reformatted based on the received format conversion instruction. A group of data objects stored in the disk cluster is converted based on the format conversion instruction and the converted group of data objects are written to the reformatted first subgroup of disks. The process iterates through the disks of the disk cluster to reformat all disks and convert all data objects based on the received format conversion instruction. The process reduces the write operations required to convert the format of the disk cluster.