Virtual Storage Mechanism Optimizing Named Object Fetch Performance
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for fetching named objects in enterprise systems are inefficient, leading to high CPU usage and increased input/output processing, as they require frequent access to direct access storage devices (DASDs).
Innovation Solution
A computer-implemented method that enhances a virtual storage mechanism by using a machine learning model to place named objects in a virtual data space, allowing for retrieval without accessing the storage system, thereby reducing CPU usage and I/O processing.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If objects are retrieved from direct access storage devices (DASDs) for every user request, then data availability is ensured, but CPU time and I/O processing increase significantly
Solution Approach 1:
The system pre-loads objects from DASD into virtual storage before they are actually requested by users. The machine learning model predicts which objects will be needed and stages them in advance in virtual storage, so when a request arrives, the object is already available in memory rather than requiring I/O from DASD at request time.
Solution Approach 2:
Virtual storage acts as an intermediary layer between DASD and the application. Instead of direct DASD access for every request, the system uses virtual storage as a buffer that holds copies of frequently accessed objects, reducing the frequency and volume of I/O operations to DASD while maintaining data availability.
2Productivity
If virtual storage is used to cache objects, then I/O processing is reduced, but storage management complexity increases
Solution Approach 1:
The machine learning model automatically monitors access patterns, predicts future object needs, and manages the loading and eviction of objects from virtual storage without human intervention. The system self-adjusts its caching strategy based on observed usage patterns, eliminating the need for manual storage management while optimizing I/O efficiency.
Solution Approach 2:
The system continuously monitors actual object access patterns and uses this feedback to refine its machine learning model's predictions. By comparing predicted versus actual access patterns, the system learns and adapts its caching strategy over time, automatically optimizing storage management without increasing operational complexity.
3Productivity
If machine learning models are used to predict object placement, then fetch performance is optimized, but system resource consumption increases
Solution Approach 1:
The machine learning model does not attempt to predict or cache all possible objects, but rather focuses on predicting only those objects that are likely to be accessed soon based on observed patterns. This partial action approach - caching only the predicted necessary subset - optimizes fetch performance for critical objects without consuming excessive CPU resources trying to manage all objects.
Data Source
AI summary
Embodiments relate to enhancing a virtual storage mechanism to optimize fetch performance of named objects in enterprise systems. A technique includes receiving from a requester a request to retrieve an object and determining that the object is available for retrieval in a virtual data space, where a machine learning model is configured to cause the object to be placed in the virtual data space, the machine learning model being trained on fetch data for objects. The technique includes retrieving the object from the virtual data space instead of a storage system and providing the object to the requester for execution.