Multi-Tier Storage Workload Distribution with Unknown Performance
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current data storage systems face challenges in optimizing data placement across multiple storage tiers with varying performance characteristics, particularly when the performance of certain tiers is incomplete, unknown, or cannot be predictively modeled, leading to inefficiencies in response time and workload management.
Innovation Solution
A method is introduced to determine a reserved workload for a tier with incomplete or unknown performance characteristics by modeling the performance of other tiers and performing data movements to optimize response time, involving the selection of a tier with the minimum classification ranking and calculating data movements based on observed and modeled response times, using performance curves and histograms to promote or demote data across tiers.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Quantity of substance
If data is placed on storage tiers with incomplete or unknown performance characteristics, then storage capacity is utilized, but response time performance cannot be predictively modeled
Solution Approach 1:
The patent introduces an intermediary storage tier (the first storage tier with incomplete performance characteristics) between the host processor and the remaining storage tiers. This intermediary tier allows the system to utilize additional storage capacity while isolating the unpredictability of its performance characteristics from the overall system performance modeling, enabling response time to be modeled for the remaining tiers without being affected by the first tier's unknown characteristics.
Solution Approach 2:
The storage system is segmented into two distinct parts: the first storage tier with incomplete performance characteristics and the remaining storage tiers with known performance characteristics. This segmentation allows the system to separately handle and model each tier's performance, enabling predictable response time modeling for the majority of storage operations while still utilizing the capacity of the unpredictable tier.
2Quantity of substance
If workload is distributed across multiple storage tiers with varying performance characteristics, then storage capacity is maximized, but workload management complexity increases
Solution Approach 1:
The patent extracts the complexity of managing the first storage tier with incomplete performance characteristics by separating it from the main workload distribution logic. The host processor or storage management system can apply simplified workload distribution rules to the remaining tiers with known characteristics, while the first tier is handled as a separate, isolated component, thereby reducing overall workload management complexity.
3Speed
If data movements are performed to optimize response time, then performance is improved, but data movement overhead increases
Solution Approach 1:
The patent performs preliminary actions by pre-characterizing storage tiers and establishing performance models before actual data placement decisions are made. By预先 determining which tiers have predictable performance characteristics and modeling their response times, the system can make informed data placement decisions that optimize response time while avoiding unnecessary data movements, thereby reducing the overhead associated with frequent data relocation.
Data Source
AI summary
Described are techniques for performing data storage optimizations. A reserved workload for a first of a plurality of storage tiers is determined. Each of the plurality of storage tiers is characterized by a set of one or more attributes. The first storage tier includes performance characteristics which are any of incomplete, unknown, and unable to be predictively modeled for various workloads. The plurality of storage tiers includes the first tier and a remaining set of additional storage tier(s). Performance is modeled of a first workload distributed among the remaining set of storage tiers. The first workload represents a total workload less the reserved workload. One or more data movements are determined in accordance with the modeling. Each data movement moves a data portion from a first physical device of one of the plurality of storage tiers to a second physical device of another of the plurality of storage tiers.