Distributed Storage Resource Partitioning for IOPS and Capacity
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current distributed storage systems face limitations in efficiently managing resources due to fixed capacity allocations, which restrict flexibility and fail to provide guaranteed quality of service, as they often assign equal performance to all object owners, leading to suboptimal utilization of resources with varying IOPS and capacity costs.
Innovation Solution
A method for partitioning resource objects into multiple components in a distributed system, where resource allocations are determined based on specific requirements and available configurations, optimizing resource utilization by selecting the most suitable resource configuration that balances resource types, such as IOPS and capacity, across host computer nodes.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If fixed capacity allocations are used to partition disks, then object flexibility is reduced and resource utilization is suboptimal, but implementation simplicity is maintained
Solution Approach 1:
The patent segments storage resources into multiple resource types (capacity, IOPS, latency) and partitions them independently using separate partitioning factors for each resource type. This allows flexible allocation of different resource types to objects based on their specific requirements, resolving the contradiction by enabling adaptability through multi-dimensional segmentation while maintaining manageable complexity through systematic partitioning rules.
Solution Approach 2:
The patent changes the allocation parameters from fixed capacity-only partitions to dynamic multi-parameter partitions that include capacity, IOPS, and latency characteristics. By introducing partitioning factors that can be adjusted independently for each resource type, the system achieves greater object flexibility and optimized resource utilization without overwhelming complexity.
2Reliability
If equal performance allocation is provided to all object owners, then implementation is simple, but quality of service guarantees cannot be enforced
Solution Approach 1:
The patent applies local quality by allocating different performance characteristics to different objects based on their specific needs. Each object receives customized partitions with tailored capacity, IOPS, and latency properties rather than uniform allocation. This enables quality of service guarantees for specific objects while maintaining overall system manageability through consistent partitioning methodologies.
Solution Approach 2:
The patent introduces dynamic allocation where partitioning factors and performance characteristics can be adjusted based on object requirements and system conditions. The system transitions from static equal allocation to dynamic differentiated allocation, enabling quality of service guarantees while managing complexity through automated partitioning calculations and resource type independence.
3Productivity
If commodity disks with varying performance characteristics are used, then cost-effectiveness and scalability are improved, but resource utilization efficiency decreases
Solution Approach 1:
The patent addresses heterogeneous disk performance by changing the allocation parameters to include multiple resource types (capacity, IOPS, latency) with adjustable partitioning factors. This allows the system to match objects with appropriate disk resources based on their specific performance requirements, optimizing resource utilization across commodity disks with varying characteristics while maintaining cost-effectiveness and scalability.
Solution Approach 2:
The patent segments the heterogeneous disk pool into distinct resource type partitions (capacity partitions, IOPS partitions, latency partitions) that can be independently allocated. This segmentation approach enables efficient matching of objects with suitable disk resources based on their specific performance needs, improving overall resource utilization while preserving the cost benefits of using commodity hardware.
Data Source
AI summary
Embodiments of the disclosure provide techniques for partitioning a resource object into multiple resource components of a cluster of host computer nodes in a distributed resources system. The distributed resources system translates high-level policy requirements into a resource configuration that the system accommodates. The system determines an allocation based on the policy requirements and identifies resource configurations that are available. Upon selecting a resource configuration, the distributed resources system assigns the allocation and associated values to the selected configuration and publishes the new configuration to other host computer nodes in the cluster.