Distributed RAID Virtualization for Resource Consolidation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current RAID systems face challenges due to redundancy in physical components and resource consumption, leading to increased hardware expenditure, communication bottlenecks, and space requirements, as both hosts and RAID systems require similar resources like power and cooling, and all communications must be addressed to a single server.
Innovation Solution
Implementing a distributed RAID system that allows various applications to be executed on the same set of computing devices using a virtualization layer, where applications and their contexts are stored in volumes within the RAID system, enabling efficient resource consolidation and fault tolerance by distributing data across multiple data banks.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If a traditional RAID system is implemented with separate host and RAID server components, then data redundancy and reliability are improved, but hardware expenditure, physical space requirements, and resource consumption increase
Solution Approach 1:
The patent combines the host system and RAID system into a single integrated computing device. The host processor executes both host applications and RAID applications, while a single set of storage devices serves both as host storage and RAID array. This merging eliminates the need for separate RAID server hardware, reducing hardware expenditure while maintaining data redundancy through software-based RAID implementation.
Solution Approach 2:
The computing device is designed to perform multiple functions: it serves as both a host system for running applications and a RAID system for data redundancy. The same processors, memory, and storage devices are utilized for both host operations and RAID operations, making the system universal and eliminating the need for dedicated separate hardware for each function.
2Reliability
If a traditional RAID system is implemented with separate host and RAID server components, then data redundancy is improved, but physical space requirements increase
Solution Approach 1:
By merging the host and RAID systems into a single computing device, the patent eliminates the need for separate physical servers and storage arrays. The integrated system uses the same physical components for both host storage and RAID array functions, significantly reducing the physical space required while maintaining full data redundancy capabilities.
3Reliability
If a traditional RAID system is implemented with separate host and RAID server components, then data redundancy is improved, but communication bottlenecks occur due to single-server addressing
Solution Approach 1:
The patent segments the RAID system into multiple independent RAID applications running as separate processes on the host system. Each RAID application can manage different volumes or storage tasks independently, allowing parallel processing and eliminating communication bottlenecks. The host processor can switch between different RAID applications without requiring communication through a single centralized RAID server.
4Device complexity
If both host and RAID systems are implemented on the same computing devices, then resource consolidation and cost reduction are achieved, but fault tolerance and reliability may be compromised
Solution Approach 1:
The patent implements fault tolerance by segmenting the system into distinct host and RAID application processes that run independently on the same physical hardware. This process-level segmentation allows the RAID functions to fail independently from host applications, and vice versa. The virtualization layer provides isolation between host and RAID operations, ensuring that failures in one do not necessarily compromise the other, thus maintaining reliability while achieving resource consolidation.
Solution Approach 2:
The patent introduces a virtualization layer as an intermediary between the host system and RAID applications. This virtualization layer provides isolation and protection, allowing the host and RAID systems to share physical resources while maintaining independent failure domains. The virtualization layer ensures that faults in the host system do not directly propagate to the RAID system and vice versa, maintaining fault tolerance despite resource consolidation.
Data Source
AI summary
Systems and methods are disclosed which allow various applications which may utilize a RAID system (or other types of applications) to be executed on the same set of computing devices which implement that RAID system. More particularly, in certain embodiments a virtualization layer may be executed on a data bank. A set of desired application programs may be executed using this virtualization layer, where the context for each instance of the applications executing on the virtualization layer may be stored in a volume kept utilizing the RAID system.