Cache Page Sharing Table for Storage Volume Duplication
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing cache systems in storage systems inefficiently manage duplicated data across multiple persistent media, leading to unnecessary duplication and inefficient use of cache space by not recognizing and referencing shared data pages.
Innovation Solution
Implementing a page sharing table to map and manage data pages across multiple volumes, allowing the cache to store only one copy of each page, even if it is duplicated across different persistent media, and using this information to direct writes and updates across the appropriate volumes.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If the cache stores separate copies of data for each volume, then each volume can be accessed independently, but cache space is wasted due to duplication of identical data pages across multiple volumes
Solution Approach 1:
The patent merges the caching of multiple volumes by storing identical data pages from different volumes in a single shared cache location. The cache controller maintains a mapping structure that tracks which volumes share the same cache page, allowing multiple volumes to reference the same cached data without duplicating storage space.
Solution Approach 2:
The cache system is designed to serve multiple volumes simultaneously through a universal caching mechanism. A single cache entry can serve multiple volumes that contain identical data, making the cache subsystem multi-functional rather than dedicated to individual volumes.
2Quantity of substance
If the cache stores only one copy of shared data pages, then cache space is optimized, but the system complexity increases due to need for tracking and managing shared pages across volumes
Solution Approach 1:
The cache management system is segmented into distinct functional components: a caching subsystem that stores the actual data pages, and a separate mapping subsystem that tracks which volumes share which cache pages. This segmentation allows the complexity of tracking shared pages to be isolated in the mapping structure without affecting the simplicity of the caching operations.
Solution Approach 2:
The patent introduces a mapping structure as an intermediary between the cache subsystem and the volume subsystem. This mapping acts as a mediator that translates volume-specific page requests into cache location lookups, handling the complexity of shared page tracking without requiring changes to the fundamental caching mechanism.
3Quantity of substance
If the system tracks and manages shared data pages across volumes, then cache space is optimized by eliminating duplication, but the processing overhead increases due to additional mapping and update operations
Solution Approach 1:
The mapping between volumes and cache pages is established in advance when data is initially cached. When a volume's data page is cached, the system proactively creates and stores the mapping information indicating which other volumes share this cache page. This preliminary action eliminates the need for complex real-time analysis when handling subsequent access requests.
Solution Approach 2:
The system implements feedback mechanisms where the mapping structure provides information about shared pages back to the cache controller. When updating or evicting a cache page, the feedback from the mapping structure identifies all volumes affected, allowing the system to efficiently propagate changes only where necessary rather than checking all volumes.
4Speed
If the cache is unaware of page sharing between volumes, then the cache operations are simple and fast, but storage efficiency deteriorates due to unnecessary duplication of data
Solution Approach 1:
The mapping structure serves as an intermediary that provides page sharing information to the cache controller without requiring the cache controller to actively manage or track the sharing relationships. The mapping absorbs the complexity of tracking, allowing the cache operations to remain simple and fast while still achieving storage efficiency through informed caching decisions.
Data Source
AI summary
In an embodiment, a method can include storing a plurality of volumes on persistent media. A set of the volumes can store at least one portion of a same copy of data. The method can further include caching the set of the volumes as a single group. In an embodiment, the plurality of volumes can include at least one of drives, snapshots, clones and replicas.