NAS Slice Balancing for Parallel Backup Streams
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Network Attached Storage (NAS) systems face challenges in backup performance due to their large storage capacity, which results in time-consuming and resource-intensive backup processes, especially since they are not configured to support multiple backup streams.
Innovation Solution
The solution involves slicing a dataset in an NAS share into independent slices, each sized by maximum file count or data volume, allowing for parallel stream backups and optimizing data grouping to improve backup efficiency.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If a single backup stream is used for NAS shares, then the backup process is simple to implement, but the backup speed is slow and resource-intensive
Solution Approach 1:
The patent divides a large NAS share into multiple smaller slices, each of which can be backed up independently using separate backup streams. This segmentation allows parallel processing of multiple slices simultaneously, significantly improving backup throughput while maintaining manageable complexity through automated slice management
2Productivity
If multiple backup streams are implemented for NAS shares, then backup throughput is improved, but the system complexity increases significantly
Solution Approach 1:
The system automatically performs slice creation, monitoring, and re-balancing operations without requiring manual intervention. The backup platform autonomously manages the complexity of multiple streams by dynamically creating and adjusting slices based on performance metrics, eliminating the need for complex manual configuration while maintaining high throughput
3Productivity
If slice size is increased to reduce the number of slices, then management overhead is reduced, but backup parallelism and speed are diminished
Solution Approach 1:
The system dynamically adjusts slice configurations based on real-time performance monitoring and changing data conditions. Slice sizes and numbers are automatically optimized to maintain optimal parallelism, with the system adapting slice management complexity only when performance degradation is detected, thereby balancing parallelism benefits against management overhead
Data Source
AI summary
One example method includes, in a network attached storage platform, determining a size of a slice of data that is included in a share of the network attached storage platform, when the size of the slice exceeds a threshold, re-slicing the slice to generate multiple new slices, and each of the new slices has a size that is smaller than the threshold, and when all slices in the share have been evaluated in terms of their respective sizes, generating a slice list that includes the new slices.