ROC Cluster Workload Negotiation for RAID Throughput Scaling
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
RAID systems face limitations in scalability and flexibility due to bandwidth constraints of RAID controllers, which can lead to bottlenecks in IO throughput and limitations on the number of virtual disks that can be managed, affecting availability and performance.
Innovation Solution
Implementing a RAID-on-chip (ROC) cluster that includes multiple RAID devices and IO controllers, allowing for distributed workload management and communication among heterogeneous devices, enabling parallel processing and offloading physical disk management to IOC devices, thereby enhancing scalability and flexibility.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If a single RAID controller manages all virtual disks, then device complexity is reduced, but bandwidth limitations cause IO throughput bottlenecks and limit scalability
Solution Approach 1:
The patent divides the RAID controller functionality into multiple independent RAID-on-chip devices that can operate in parallel. Each device manages a subset of virtual disks, allowing the system to handle higher IO throughput by distributing the workload across multiple controllers rather than relying on a single bottleneck-prone controller.
Solution Approach 2:
The patent introduces an intermediary communication layer between multiple RAID-on-chip devices and the host system. This intermediary mechanism enables coordinated workload distribution and data management across multiple controllers while maintaining a simplified interface for the host, thus resolving the complexity-throughput tradeoff.
2Quantity of substance
If more virtual disks are managed by a single controller, then device complexity remains low, but the number of manageable disks is limited by controller bandwidth
Solution Approach 1:
The patent segments the management of virtual disks across multiple RAID-on-chip devices. Each controller manages a subset of disks, allowing the system to support a larger total number of virtual disks while maintaining acceptable IO throughput performance through parallel processing across multiple devices.
3Reliability
If a single RAID controller is used, then cost is reduced, but availability is compromised when the controller experiences problems
Solution Approach 1:
The patent implements local quality by making each RAID-on-chip device independently manageable and fault-isolatable. When one controller experiences a problem, only the virtual disks managed by that specific device are affected, while other devices continue to operate normally. This localized fault containment improves overall system availability without requiring complex inter-controller redundancy mechanisms.
4Adaptability or versatility
If RAID systems are extended to support more virtual disks, then adaptability improves, but bandwidth constraints create performance penalties
Solution Approach 1:
The patent segments the virtual disk management workload across multiple RAID-on-chip devices, each capable of independently configuring and managing its assigned disks. This segmentation enables the system to support a larger number of virtual disks with diverse configurations while maintaining high IO throughput through parallel processing, thus achieving both adaptability and performance improvements.
Data Source
AI summary
Devices, systems, and corresponding methods, including without limitation RAID devices, I/O controllers, and RAID clusters, that can provide enhanced storage solutions. A RAID device might be a RAID-on-chip device. A RAID cluster can comprise a plurality of RAID devices and/or I/O controllers that can provide one or more virtual disks for use by a host. In some cases, a RAID cluster can be a ROC cluster, which includes one or more ROC devices.