User-Based Data Tiering via Inode Extended Attributes
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing storage tiering systems lack the ability to differentiate between files modified by different users, leading to equal treatment of files irrespective of user criticality, resulting in inadequate quality of service (QoS) for premium users and inefficient data movement across storage tiers.
Innovation Solution
Implementing user-based data tiering by assigning priorities to users and using extended attributes to track recent file modifications, allowing for finer control over data movement and storage tier allocation based on user priority, with policies defining when and how files are moved between storage tiers.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of energy
If storage tiering is implemented without user differentiation, then storage cost is reduced through automated data movement, but quality of service for premium users deteriorates due to equal treatment of all files
Solution Approach 1:
The patent applies local quality by differentiating storage treatment based on user characteristics. Premium users receive preferential treatment with their files retained on high-performant storage tiers, while standard users' files are moved to archival tiers. This is implemented through user priority attributes and differentiated tiering policies that apply different quality standards to different user segments, resolving the contradiction between cost reduction and service quality maintenance.
2Extent of automation
If traditional storage tiering is used, then automated data movement occurs based on file attributes, but differentiation capability deteriorates because user criticality cannot be distinguished
Solution Approach 1:
The patent introduces a new dimension for storage tiering by incorporating user identity and priority attributes into the tiering decision process. Instead of relying solely on file attributes, the system adds user-based dimensions (user ID, user priority level) to the tiering criteria. This enables automated data movement to differentiate between users, allowing premium users to maintain files on high-performant tiers while standard users' files are archived, thus enhancing differentiation capability while maintaining automation.
3Device complexity
If all files are treated equally in storage tiering, then system complexity is reduced through uniform policies, but service quality deteriorates due to inability to provide premium service to critical users
Solution Approach 1:
The patent applies segmentation by dividing the user base into distinct segments (premium users and standard users) with different priority levels. This segmentation enables the system to apply different tiering policies to different user segments, allowing premium users to receive enhanced service with their files retained on high-performant storage while standard users receive standard service. The segmentation approach maintains manageable system complexity through clear policy definitions for each segment while significantly improving service quality differentiation.
Data Source
AI summary
Techniques are provided for user-based data tiering. In an example, a computer maintains a first-in-first-out queue that logs a finite set of users that have most-recently modified a file. This queue can be maintained in an extended attribute of an inode that corresponds to a file. A computer can also maintain a policy that defines how to perform storage tiering on a file based on which users have accessed the file. When a tiering operation is performed, the files specified by a corresponding tiering policy can be evaluated for which users have recently accessed them. When a user specified by the tiering policy has recently modified a file, the file can be placed in tiering queue for tiering.