Storage Provisioning via Performance Data Filtering
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Storage administrators face challenges in provisioning and managing storage resources in networked environments, particularly in determining if resources have enough performance capacity to meet latency goals, especially with varying workloads and demands across multiple resources.
Innovation Solution
A performance manager module interfaces with the storage operating system to collect Quality of Service (QOS) data and manage resources based on available performance capacity, using a provisioning engine to assign performance parameters and identify suitable resource pairs that can meet demand, while filtering historical performance data to account for transient events.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If storage administrators manually provision and manage storage resources in networked environments, then they can control resource allocation, but it becomes difficult to determine if resources have enough performance capacity to meet latency goals especially with varying workloads and demands
Solution Approach 1:
The patent introduces a performance manager module as an intermediary between storage administrators and the complex networked storage environment. This module automatically collects QOS data, analyzes performance capacity, and provides provisioning recommendations, eliminating the need for administrators to manually assess complex performance metrics across multiple resources with varying workloads.
Solution Approach 2:
The system enables self-service by automatically monitoring and analyzing performance data from storage resources. The performance manager module continuously collects QOS metrics, evaluates whether resources meet latency goals, and identifies provisioning opportunities without requiring manual administrator intervention, thus simplifying the management process while improving measurement accuracy.
2Measurement precision
If performance data is collected without filtering transient events, then data collection is simple, but performance capacity assessment becomes inaccurate due to transient events affecting the data
Solution Approach 1:
The patent extracts and removes transient events from the collected performance data through filtering operations. The performance manager module identifies and eliminates data points corresponding to transient events (such as temporary failures or anomalies) before analyzing performance capacity, ensuring that the assessment is based on stable, representative performance characteristics rather than distorted by temporary conditions.
Solution Approach 2:
The system performs preliminary filtering of performance data to remove transient events before conducting performance capacity analysis. By pre-processing the data to eliminate anomalies and transient conditions, the system ensures that subsequent performance assessments are based on clean, reliable data, improving measurement accuracy without significantly complicating the overall process.
3Productivity
If automated provisioning systems are implemented to manage storage resources, then resource allocation efficiency improves, but the system complexity increases
Solution Approach 1:
The performance manager module serves multiple functions within a single integrated system: it collects QOS data, analyzes performance capacity, identifies provisioning opportunities, and provides recommendations. This multi-functional approach improves provisioning efficiency by automating the entire workflow while containing system complexity within a single unified module rather than requiring multiple separate systems.
Data Source
AI summary
Methods and systems for a networked storage system are provided. A provisioning engine assigns a plurality of performance parameters in response to a provisioning request for provisioning a workload for storing data in a networked storage environment; identifies a demand for a plurality of resources of the networked storage environment for meeting the provisioning request, transforms historical available performance capacity by filtering any historical performance capacity data related to any transient event; and identifies at least a resource pair that can meet the identified demand based on the transformed historical performance capacity data.