Distributed QoS Storage Management via Resource Pooling
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional Quality of Service (QoS) techniques in resource management lack sufficient control for enterprises with different infrastructures and objectives, often requiring a centralized scheduler that increases complexity and susceptibility to system-wide failures.
Innovation Solution
A distributed QoS management system using a resource pool module and local scheduler in each host computer, which computes and allocates storage resource entitlements based on global average latency and client demands, allowing for robust QoS control without a centralized scheduler.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If a centralized scheduler is used for QoS management, then resource allocation control is improved, but system complexity and susceptibility to system-wide failures increase
Solution Approach 1:
The patent segments the centralized QoS management function into distributed components. Each host computer runs its own resource pool module and local scheduler, dividing the monolithic centralized scheduler into multiple independent units. This segmentation eliminates the single point of failure inherent in centralized systems while maintaining QoS control capabilities at each segment.
Solution Approach 2:
The patent extracts the scheduling function from a centralized entity and embeds it within each host computer's resource pool module. By taking out the centralized scheduler and distributing its functionality to individual hosts, the system eliminates the complexity and failure risks associated with a single centralized component while preserving resource allocation control.
2Ease of operation
If a centralized scheduler is used for QoS management, then resource allocation control is improved, but the risk of system-wide failures increases
Solution Approach 1:
The system segments resource allocation control into distributed decisions made by local schedulers at each host. Each host independently manages its own resource pool and makes allocation decisions based on global average latency metrics, eliminating the single point of failure in centralized systems while maintaining operational control through coordinated distributed decision-making.
Solution Approach 2:
The patent implements feedback mechanisms where each host's resource pool module monitors global average latency and adjusts local resource allocation accordingly. This feedback loop enables distributed hosts to maintain coordinated resource management without requiring a centralized scheduler, improving reliability while preserving control through adaptive response to system conditions.
3Reliability
If distributed resource allocation is implemented, then system reliability is improved, but coordination complexity among hosts increases
Solution Approach 1:
The patent implements a universal resource pool module design that can be deployed identically on each host computer. This universal module handles both local resource management and global coordination functions, eliminating the need for complex specialized coordination mechanisms between hosts. Each host performs the same functions, simplifying the overall system architecture despite the distributed nature.
Solution Approach 2:
The patent creates equipotential conditions by having all hosts use the same resource pool module implementation and the same latency-based allocation algorithm. This uniformity across all hosts eliminates coordination complexity that would arise from heterogeneous systems, as each host operates under identical rules and can independently make consistent decisions without complex inter-host negotiation.
Data Source
AI summary
A system and method for providing quality of service (QoS) for clients running on host computers to access a common resource uses a resource pool module and a local scheduler in at least one of the host computers. The resource pool module operates to compute an entitlement of each client for the common resource based on a current capacity for the common resource and demands of the clients for the common resource. In addition, the resource pool module operates to assign a portion of the computed current capacity for the common resource to a particular host computer using the computed entitlement of each client running on the particular host computer. The local scheduler operates to allocate the portion of the computed current capacity among the clients running on the particular host computer.