Spool Data Policy Automation for Storage Cost Control
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Large scale storage systems face challenges with unchecked data volume growth leading to storage cost overruns, immaturity of big data platforms, and the need to derive insights from stored data, despite declining storage hardware costs.
Innovation Solution
A computer-implemented method that collects historical and real-time information to train sub models for predicting user actions with spool usage and job statuses, integrates these models into a meta model for policy script generation, and automatically executes commands for intelligent spool data management.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Quantity of substance
If data volumes continue to grow unchecked to meet increasing storage needs, then storage capacity is improved, but storage costs increase and become unsustainable
Solution Approach 1:
The system dynamically changes spool management parameters based on job characteristics and historical data. The meta model adjusts spool dataset allocation, retention policies, and cleanup timing by changing operational parameters rather than fixed rules, optimizing storage utilization while controlling costs through intelligent, adaptive parameter adjustment.
Solution Approach 2:
The spool management system performs self-service through automated machine learning models that continuously learn from historical data and make autonomous decisions about spool dataset management. The system self-optimizes storage allocation, automatically identifies candidates for data removal, and executes management actions without manual intervention, reducing operational overhead and improving efficiency.
2Reliability
If all data is retained to ensure data availability and compliance, then data accessibility is improved, but storage space is wasted and costs increase
Solution Approach 1:
The system performs preliminary actions by proactively identifying spool datasets that are candidates for removal before they consume excessive storage space. The machine learning models predict which datasets will become obsolete based on historical patterns and job characteristics, enabling preemptive data removal that maintains availability of needed data while preventing storage waste from retained obsolete data.
Solution Approach 2:
The system implements intelligent discarding and recovering by selectively removing spool datasets that are no longer needed while preserving those that remain valuable. The meta model determines optimal retention periods and identifies candidates for removal based on job status, user behavior patterns, and data accessibility requirements, enabling space recovery without compromising data availability when needed.
3Loss of energy
If manual spool management procedures are used to control storage usage, then storage cost control is improved, but operational complexity and time consumption increase
Solution Approach 1:
The system replaces manual mechanical spool management procedures with automated machine learning-based intelligence. Instead of relying on human operators to manually review and manage spool datasets, the system uses trained models to automatically analyze job data, predict retention needs, and execute management actions, substituting automated computational processes for manual operational complexity.
Solution Approach 2:
The system implements continuous feedback loops where the meta model learns from historical spool management outcomes and job data patterns. The system collects feedback on data accessibility, storage utilization, and job completion patterns, continuously refining its predictions and management decisions to optimize storage cost control while reducing operational complexity through increasingly accurate automated decisions.
Data Source
AI summary
A computer-implemented method is provided for intelligently managing simultaneous peripheral operations online (spool) data sets. The computer-implemented method includes collecting historical information and real-time information, training, based on the historical information and the real-time information, first sub models to predict user actions with different spool usage and second sub models to predict user actions with various job statuses, integrating the first and second sub models into a meta model configured to generate policy scripts for system spool management, transforming the policy scripts to executive commands for the system spool management and automatically issuing and executing the executive commands for the system spool management.