Mapping Client Integration for Shared Storage I/O Elimination
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In storage virtualization environments, accessing shared data sets by multiple nodes simultaneously is inefficient due to high I/O requirements and storage redundancy, especially as the number of virtual machines increases, leading to performance and resource utilization challenges.
Innovation Solution
A system and method that integrates a mapping client with a data cache in the client's operating system, allowing applications to access shared storage objects as logically addressable objects, reducing the need for physical I/O operations by caching and deduplicating data, thereby streamlining access and storage efficiency.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Quantity of substance
If multiple virtual machines access shared storage simultaneously through traditional file server architecture, then storage capacity and network bandwidth are increased, but I/O operations multiply and performance degrades
Solution Approach 1:
The patent merges multiple virtual machine I/O operations into a single consolidated I/O path by introducing a storage virtualization layer. This layer aggregates I/O requests from multiple VMs and presents them as unified operations to the physical storage array, thereby maintaining storage capacity while preventing I/O performance degradation that would occur with traditional separate access paths for each VM.
Solution Approach 2:
The patent introduces a storage virtualization intermediary layer that sits between the virtual machines and the physical storage array. This intermediary consolidates and manages I/O operations, translating multiple VM I/O requests into optimized physical I/O operations, thereby decoupling the quantity of storage capacity from the multiplicative effect on I/O operations.
2Speed
If storage data is duplicated for each virtual machine instance, then data access speed is improved, but storage redundancy increases
Solution Approach 1:
The patent implements intelligent copying mechanisms where data blocks are copied and cached in the virtualization layer's buffer memory when first accessed. Subsequent accesses to the same data blocks by different virtual machines are served from this cache rather than requiring duplicate physical copies, thereby maintaining fast data access while eliminating storage redundancy.
Solution Approach 2:
The patent performs preliminary data caching and pre-fetching operations in the virtualization layer. Data is proactively cached in buffer memory before being requested by multiple VMs, and frequently accessed blocks are pre-positioned in optimal locations. This preliminary action enables fast data access for multiple VMs without requiring duplicate physical storage copies.
3Reliability
If physical I/O operations are performed for every storage access request, then data consistency is maintained, but I/O operation count increases
Solution Approach 1:
The patent performs preliminary data validation and consistency checks in the virtualization layer before data is accessed by virtual machines. Data integrity is verified and cached metadata is prepared in advance, allowing subsequent access requests to be served from the cached information without requiring repeated physical I/O operations, thereby maintaining data consistency while reducing I/O counts.
Solution Approach 2:
The patent implements a caching buffer in the virtualization layer that acts as a cushion between physical storage and virtual machine access. This buffer pre-loads and holds data blocks, allowing multiple VMs to access the same data without triggering repeated physical I/O operations. The cushioning effect maintains data consistency while dramatically reducing the number of physical I/O operations required.
Data Source
AI summary
This invention is a system and a method for performing an I/O in a virtual data storage environment using a new architecture. The system of performing an I/O includes a mapping client integrated into a client of the storage server which in communication with the mapping server included in the storage server retrieves the mapping of the special data sharing storage objects and caches the shared objects in the data cache include in the client environment. The method of accessing the data sharing storage objects by one or more applications running on a client reduces the number of I/O on the storage objects by caching the storage objects in the data cache and bringing the knowledge of data sharing into the client environment.