Cloud Storage Performance-Based Provisioning
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Cloud computing providers lack a mechanism to accurately provision and market storage space based on performance parameters such as throughput and latency, which are crucial for efficient storage management in cloud computing environments.
Innovation Solution
A system that includes a performance manager and a cloud manager, which present clients with options associated with specific latency and throughput targets, allowing clients to select based on performance requirements and enabling the cloud provider to charge accordingly, using quality of service (QOS) data structures to prioritize and manage storage operations.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If storage space is presented to clients in cloud computing environment without performance parameters, then storage provisioning is simple, but clients cannot select storage based on specific throughput and latency requirements
Solution Approach 1:
The patent introduces performance parameters (throughput and latency) as new dimensions for storage selection. Storage volumes are characterized by specific IOPS ranges and latency targets, transforming the storage selection process from simple capacity-based to performance-parameter-based, enabling clients to match storage options with their specific performance requirements
Solution Approach 2:
The patent introduces a cloud manager as an intermediary component that maintains data structures mapping storage volumes to their performance characteristics. This intermediary layer translates client performance requirements into appropriate storage volume selections, resolving the contradiction between simple provisioning and performance-based selection
2Loss of energy
If cloud providers license storage space without performance differentiation, then pricing is straightforward, but providers cannot monetize performance parameters like throughput and latency
Solution Approach 1:
The patent transforms the pricing model by introducing performance parameters (IOPS ranges, latency targets) as billing dimensions. Different storage volumes with different performance characteristics can be priced differently, enabling providers to capture value from performance differentiation while maintaining clear pricing structures through defined performance tiers
Solution Approach 2:
The cloud manager data structures serve multiple functions: they track storage volume performance characteristics, match client requirements to appropriate volumes, and enable performance-based pricing. This multi-functional approach allows the system to handle both technical allocation and commercial pricing through a unified mechanism
3Productivity
If storage operations are managed without QOS data structures, then system complexity is reduced, but storage operations cannot be prioritized or managed efficiently
Solution Approach 1:
The patent implements feedback mechanisms where the cloud manager continuously monitors storage volume performance against target metrics and adjusts resource allocation accordingly. QOS data structures store performance targets and actual measurements, enabling the system to respond to performance deviations and maintain service level agreements through automated feedback loops
Data Source
AI summary
Methods and systems for presenting a plurality of options to a client for using storage space in a cloud computing environment are provided. Each option is associated with a latency target and/or a throughput target. The latency target provides a delay in processing input/output (I/O) requests and the throughput target provides a number of I/O requests that are processed within a unit of time. An existing volume is assigned to the client when the existing volume meets a guaranteed latency target and/or a guaranteed throughput target for an option selected from the plurality of options, otherwise a new volume is allocated.