Adaptive Persistence Levels for Storage I/O Requests
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current storage systems treat all input/output (I/O) requests equally, leading to reduced performance as they fail to differentiate between data that needs long-term preservation and data that is temporary, resulting in inefficient use of resources.
Innovation Solution
The method involves identifying and selecting adaptive persistence levels for I/O requests based on their properties, such as file identifiers and application associations, to determine the appropriate storage resources and modes, including ephemeral caching mode for temporary data, which does not require primary storage access.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If all I/O requests are treated equally with the same persistence level, then data reliability is maintained, but I/O performance deteriorates due to unnecessary storage operations on temporary data
Solution Approach 1:
The patent segments I/O requests into different persistence levels (e.g., ephemeral, short-term, long-term, permanent) based on data characteristics. This segmentation allows the system to apply different storage strategies to different data types, improving performance for temporary data while maintaining reliability for critical data through differentiated handling.
Solution Approach 2:
The patent applies local quality by assigning different persistence levels and storage policies to different data based on their specific requirements. Critical data receives high-persistence treatment with redundant storage, while temporary data receives low-persistence treatment with ephemeral caching, optimizing both performance and reliability locally for each data type.
2Reliability
If persistent storage is used for all I/O requests, then data preservation is ensured, but resource utilization deteriorates due to wasted storage capacity and bandwidth on temporary data
Solution Approach 1:
The patent changes the persistence parameter dynamically based on data characteristics. By adjusting the persistence level parameter according to data type, application requirements, and access patterns, the system ensures data preservation when needed while reducing storage resource consumption for temporary data through ephemeral caching strategies.
Solution Approach 2:
The patent introduces dynamic persistence levels that can change over time based on data age, access frequency, and system conditions. Data can transition between persistence levels (e.g., from ephemeral to short-term to long-term), allowing the system to optimize resource utilization while maintaining data preservation requirements.
3Loss of time
If ephemeral caching mode is used for temporary data, then I/O latency is reduced, but data loss risk increases for non-persistent data
Solution Approach 1:
The patent segments data into different persistence categories, explicitly identifying which data should use ephemeral caching (with accepted loss risk) and which should use persistent storage. This segmentation allows the system to achieve low latency for temporary data while maintaining reliability for critical data through differentiated persistence strategies.
Data Source
AI summary
A storage module may be configured to service I/O requests according to different persistence levels. The persistence level of an I/O request may relate to the storage resource(s) used to service the I/O request, the configuration of the storage resource(s), the storage mode of the resources, and so on. In some embodiments, a persistence level may relate to a cache mode of an I/O request. I/O requests pertaining to temporary or disposable data may be serviced using an ephemeral cache mode. An ephemeral cache mode may comprise storing I/O request data in cache storage without writing the data through (or back) to primary storage. Ephemeral cache data may be transferred between hosts in response to virtual machine migration.