Read Ahead Caching for Non-Sequential Cloud Storage Access
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional file storage methods are inefficient in handling non-sequential I/O requests and do not effectively utilize tiered storage systems, particularly in cloud storage environments where files with limited lifespan clutter storage and require efficient retrieval.
Innovation Solution
A tiered storage system that moves infrequently accessed files from high-tier storage to lower-tier cloud storage, maintaining stub files in the high-tier system to allow for efficient retrieval based on non-sequential access patterns, optimizing storage space and performance.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Speed
If files are stored in high-tier storage for frequent access, then access speed is improved, but storage cost increases and space is wasted on infrequently accessed files
Solution Approach 1:
The patent segments files into frequently accessed and infrequently accessed categories, storing different segments in different storage tiers. The file system monitors access patterns and automatically moves files between high-tier and low-tier storage based on their access frequency, ensuring that only files requiring fast access occupy expensive high-tier storage space.
Solution Approach 2:
The patent implements preliminary actions by pre-loading file data into high-tier storage before actual access occurs. The system monitors access patterns and proactively transfers files from low-tier to high-tier storage in advance, so when a file is needed, it is already in the fast storage tier, maintaining access speed without permanently occupying high-tier space.
2Quantity of substance
If cloud storage is used for low-tier storage, then storage cost is reduced, but access time increases due to network latency
Solution Approach 1:
The patent applies preliminary action by anticipating future file access needs and pre-transferring files from cloud storage to local high-tier storage before they are actually accessed. The system analyzes access patterns and proactively loads files into local storage, eliminating network latency when files are needed while still maintaining the ability to use inexpensive cloud storage for archival purposes.
Solution Approach 2:
The patent introduces a local high-tier storage system as an intermediary between cloud storage and the file access points. This intermediary layer caches frequently accessed files locally, so when files are needed, they can be retrieved from the local high-tier storage without incurring cloud access latency, while less frequently accessed files remain in the cloud.
3Productivity
If sequential data loading is used, then I/O efficiency is improved for sequential access patterns, but efficiency drops for non-sequential access patterns
Solution Approach 1:
The patent implements dynamic data loading strategies that adapt to the actual access patterns observed. Instead of always loading sequential blocks, the system monitors how files are accessed and dynamically adjusts the loading strategy - loading sequential blocks when sequential access is detected, and loading non-sequential blocks when random access patterns are observed, thereby maintaining high I/O efficiency across different access scenarios.
Solution Approach 2:
The patent employs feedback mechanisms where the system continuously monitors file access patterns and uses this information to optimize data loading strategies. When non-sequential access patterns are detected, the system receives feedback and adjusts by loading only the specific non-sequential blocks that are actually needed, rather than wasting I/O resources loading sequential blocks that won't be accessed.
Data Source
AI summary
A high tier storage area stores a stub file and a lower tier cloud storage area stores the file corresponding to the stub file. When a client apparatus requests segments of the file from the high tier storage area, reference is made to the stub file to determine a predicted non-sequential pattern of requests to the segments by the client apparatus. The high tier storage area follows the predicted non-sequential pattern of requests to retrieve the segments of the file from the cloud prior to the client apparatus actually requesting the segments. As such, the file may be efficiently provided to the client apparatus while also efficiently storing the file on the lower tier cloud storage area.