Auto-Tiering Storage Pool Design Using Workload Density Functions
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The design of auto-tiering data systems is complex and often based on speculation and estimation, lacking a systematic approach to optimize data distribution across different storage tiers based on workload requirements.
Innovation Solution
A computer-implemented method that defines a storage pool with multiple tiers of varying performance, where workloads are defined with target skew factors, capacity requirements, and IOPS requirements, and a density function is generated to determine the optimal distribution of content across these tiers, using historical or default information to simulate real-world workloads and generate consolidated system loading reports.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If auto-tiering systems are designed based on speculation and estimation, then system design can be completed, but the design accuracy and optimization of data distribution are poor
Solution Approach 1:
The patent applies preliminary action by defining workload characteristics and generating density functions before actual system deployment. The design process establishes target skew factors, capacity requirements, and IOPS requirements in advance, allowing the system to be optimized based on predicted rather than speculative data distribution patterns.
Solution Approach 2:
The patent utilizes parameter changes by introducing target skew factors as a key design parameter that quantifies data distribution characteristics across storage tiers. By varying this parameter based on different workload types, the system can optimize data placement without requiring complex speculative design approaches.
2Productivity
If multiple storage tiers with different performance levels are used, then system performance can be optimized, but determining optimal data distribution becomes complex
Solution Approach 1:
The patent applies feedback by using density functions that mathematically model the relationship between storage tier performance and data distribution. The target skew factor serves as a feedback mechanism that guides data placement decisions, allowing the system to achieve optimal performance across multiple tiers without complex manual configuration.
Solution Approach 2:
The patent utilizes dynamics by making the data distribution model adaptable to different workload characteristics. The density functions can be regenerated for different workload types, allowing the system to dynamically adjust optimal data distribution strategies based on actual usage patterns rather than static configurations.
3Reliability
If workloads are defined with detailed requirements including target skew factor, capacity requirement, and IOPS requirement, then workload optimization improves, but the definition process becomes more complex
Solution Approach 1:
The patent applies segmentation by breaking down workload definitions into distinct, manageable components: target skew factor, capacity requirement, and IOPS requirement. This segmentation allows each parameter to be independently optimized and understood, reducing the overall complexity of workload definition while improving reliability.
Data Source
AI summary
A method, computer program product, and computing system for defining a storage pool for a storage system being designed that includes a plurality of storage tiers. Each storage tier has a different level of performance. A plurality of workloads are defined for the storage system, wherein each of the plurality of workloads includes: a target skew factor, a capacity requirement, and a IOPS requirement. A density function is generated for each of the plurality of workloads based, at least in part, upon the target skew factor for each of the plurality of workloads. A consolidated system loading report is generated based, at least in part, upon the target skew factor for each of the plurality of workloads.