Dynamic Data Collection Scheduling for Storage Systems
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional data collection mechanisms in storage systems face challenges such as delayed or incomplete data collection, leading to inefficiencies in problem analysis and maintenance, particularly with partial data collection being insufficient and full data collection impacting system performance.
Innovation Solution
A method that dynamically adjusts data collection schedules based on grade ranges and relevancies of running parameters, such as system health, CPU usage, and time intervals, to ensure timely and relevant data collection.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of information
If full data collection is performed, then data completeness is improved, but system performance deteriorates
Solution Approach 1:
The patent applies partial action by collecting only the necessary subset of data based on system state. Instead of always performing full data collection, the system dynamically determines the appropriate data collection scope based on running parameters and grade ranges, collecting partial data when system state is normal and full data when problems are detected, thus balancing data completeness with system performance
2Loss of time
If data collection is performed frequently, then data timeliness is improved, but system resource consumption increases
Solution Approach 1:
The patent implements dynamic data collection scheduling where the collection frequency and timing are adjusted based on system state. The system monitors running parameters and dynamically determines when data collection should occur by comparing parameter values against grade ranges, making the data collection process adaptive rather than static, thus improving timeliness when needed while conserving resources during normal operation
Solution Approach 2:
The system changes operational parameters based on system state by monitoring running parameters (CPU usage, memory usage, I/O operations, etc.) and adjusting data collection timing accordingly. When parameters exceed thresholds defined by grade ranges, the system triggers data collection; otherwise, it delays collection, thus optimizing the balance between timeliness and resource consumption through parameter-driven dynamic adjustment
3Productivity
If partial data collection is performed, then system performance is maintained, but diagnostic capability deteriorates
Solution Approach 1:
The patent implements feedback mechanisms by continuously monitoring system running parameters and using this information to determine appropriate data collection scope. The system collects partial data during normal operation to maintain performance, but when feedback from parameter monitoring indicates problems (parameters exceeding grade range thresholds), the system automatically expands data collection to ensure complete diagnostic information is captured, thus resolving the contradiction between performance maintenance and diagnostic capability
Data Source
AI summary
Embodiments of the present disclosure provide a method, an electronic device, and a computer program product for scheduling data collection. The method includes acquiring a plurality of running parameters of a storage system. The method further includes determining a plurality of grade ranges of each of the plurality of running parameters, the grade ranges indicating degrees of impact on scheduling for data collection. The method further includes determining a plurality of relevancies of the plurality of running parameters for the plurality of grade ranges. The method further includes determining scheduling for the data collection based on the plurality of relevancies. The method can dynamically determine when to perform data collection, thus avoiding data loss.