Multi-Tier Cloud Storage Provisioning via Common API
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current data storage solutions fail to balance capacity, performance, and cost effectively, particularly for I/O intensive applications, leading to unbalanced systems and high costs for high-performance storage, with cloud storage offering inadequate performance and custom-built solutions being prohibitively expensive.
Innovation Solution
A cloud-computing environment provides a flexible storage solution using a common API to provision multiple tiers of storage with varying performance levels, allowing customers to design custom storage blends that optimize capacity, performance, and cost without requiring rebooting of host computers.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Quantity of substance
If cloud storage uses thin provisioning to share storage capacity among many tenants, then storage capacity utilization is improved, but I/O performance deteriorates
Solution Approach 1:
The storage system is segmented into multiple performance tiers (e.g., high-performance SSD tier, standard HDD tier, archival tier). Each tier is independently managed and can be allocated to different tenants or workloads based on performance requirements. This segmentation allows the system to provide high I/O performance to specific tenants without requiring all storage capacity to be high-performance, thus resolving the contradiction between capacity utilization and performance.
2Productivity
If custom-built high-performance storage solutions are implemented, then I/O performance is improved, but cost increases
Solution Approach 1:
Instead of making the entire storage system high-performance, the patent applies high-performance storage resources locally to specific tenants or workloads that require high I/O performance. The multi-tier architecture allows high-performance SSDs to be allocated only where needed, while other tenants use standard or archival storage. This local application of quality resolves the contradiction by providing high performance only where necessary, reducing overall cost.
3Quantity of substance
If traditional cloud storage shares performance proportionally among tenants, then storage capacity is maximized, but performance for individual tenants deteriorates
Solution Approach 1:
The storage system implements dynamic performance allocation where each tenant can select their desired performance level. The system dynamically provisions storage resources from different performance tiers based on individual tenant requirements rather than uniformly sharing all resources. This dynamic approach allows the system to maintain high capacity utilization while ensuring each tenant receives the performance level they need.
Data Source
AI summary
This document discloses high performance data storage solutions. In an aspect, some solutions can be employed in a cloud-computing environment that provides shared storage for a plurality of customers. In other aspects, the solutions provided by some embodiments can provide multiple tiers of storage, each having a different performance level. This feature can provide a customer with the ability (e.g., through a web portal) to design its own custom storage solution that blends multiple tiers of storage at different capacity and performance levels to attain an optimal level of capacity, performance, and cost. Some disclosed solutions provide a novel arrangement of different types of storage devices that are provisioned through a common API, which can access each storage device's native management interface, as well as the host computer for which the storage will be provisioned, to allocate multiple tiers of storage to the host computer in an automated manner.