Multi-Tier Storage Configuration Adviser for SSD-HDD Cost Optimization
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current commercial SSD-based multi-tier storage systems face challenges in determining the optimal mix of devices to achieve performance and cost efficiency due to complexity in workload variability and incomplete information, leading to over-provisioning and failure to meet customer expectations.
Innovation Solution
A configuration adviser is implemented to optimize storage extent placement by determining the lowest cost storage tier for each data extent based on data access and resource information, selecting the minimum number of devices needed to meet performance and capacity requirements, and generating configuration options with correlated response times and costs.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Speed
If storage systems use only SSDs to provide high performance storage, then performance requirements are met, but capital cost increases significantly
Solution Approach 1:
The storage system is segmented into multiple tiers with different performance characteristics. Hot data requiring high performance is placed on SSDs, while cold data with lower performance requirements is placed on HDDs. This segmentation allows the system to meet performance requirements for critical operations while reducing overall capital cost by not provisioning entire systems with expensive SSDs.
Solution Approach 2:
Different storage media with different quality characteristics are used in different locations within the storage system. SSDs are deployed locally for storing frequently accessed data that requires high random access capability, while HDDs are used locally for storing less frequently accessed data. This local quality approach ensures that high performance is provided only where needed, optimizing the balance between performance and cost.
2Quantity of substance
If storage systems use only HDDs to reduce capital cost, then capital cost decreases, but performance requirements cannot be met
Solution Approach 1:
The storage system is divided into performance tiers where SSD-based tiers handle workloads requiring high random access capability while HDD-based tiers handle workloads with lower performance requirements. This segmentation ensures that performance-critical operations are executed on fast storage media while maintaining overall system cost-effectiveness.
Solution Approach 2:
High-performance SSD storage is deployed locally for specific data sets that require fast random access, while standard HDD storage is used locally for other data sets. This local quality differentiation ensures that performance requirements are met for critical operations without incurring the high capital cost of provisioning entire systems with SSDs.
3Reliability
If storage teams over-provision systems based on incomplete workload information, then performance requirements may be met, but capital cost increases due to unnecessary capacity
Solution Approach 1:
The system performs preliminary analysis of workload characteristics and data access patterns before finalizing the storage configuration. By proactively analyzing workload information and predicting performance requirements, the system can provision storage capacity more accurately, avoiding both over-provisioning and under-provisioning scenarios.
Solution Approach 2:
The system continuously monitors actual workload performance and storage system performance, using this feedback to optimize the configuration. By comparing actual performance metrics against requirements and adjusting the storage mix accordingly, the system can maintain performance guarantees while minimizing capital cost by removing unnecessary capacity.
Data Source
AI summary
In one embodiment, a configuration adviser provides configuration options for a multi-tiered storage system that includes a number of different storage tiers, each of which include storage devices of a particular storage type. Data access information for storage extents to be stored in the storage system are received. Resource information for available storage tiers in the storage system to place the storage extents on are also received. A cost incurred by the storage system for placing each of the storage extents on each of the storage tiers is determined. Each storage extent is assigned to a particular one of the storage tiers that would incur the lowest cost to the storage system for storing the storage extent. For each storage tier, a minimum number of storage devices are selected, within the assigned storage tier, that would satisfy data access and capacity requirements for all storage extents assigned to that tier.