Unified Data Path for Block and File Storage Reuse
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Block-based and file-based data storage systems often operate independently, making it difficult to transfer advancements between the two, leading to inefficiencies such as stranded storage where resources are underutilized due to the inability to reuse storage allocated to one type for the other.
Innovation Solution
A unified data path architecture that expresses both block-based and file-based objects as files within an underlying file system, allowing for common services like replication, deduplication, and storage reallocation across both types, thereby eliminating stranded storage inefficiencies.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If block-based and file-based storage systems operate independently with separate architectures, then each system can be optimized for its specific protocol and operations, but storage resources become stranded and cannot be reused across different object types
Solution Approach 1:
The patent merges block-based and file-based storage operations into a unified data path architecture. Both block objects (LUNs) and file objects are expressed as files in an underlying file system, allowing storage resources to be shared and reused across different object types while maintaining protocol-specific optimizations through separate processing paths.
Solution Approach 2:
The unified data path architecture provides universal storage resource management that serves both block-based and file-based protocols. The underlying file system acts as a universal layer that can allocate and manage storage blocks for any object type, eliminating stranded storage while maintaining specialized functionality for each protocol type.
2Reliability
If separate processing paths are maintained for block-based and file-based storage, then protocol-specific operations can be optimized, but system complexity increases and maintenance becomes difficult
Solution Approach 1:
The patent combines block-based and file-based processing paths into a unified data path that shares common infrastructure components including storage resource management, data movement operations, and underlying file system access. This reduces system complexity while maintaining protocol-specific optimization through targeted processing layers.
3Reliability
If storage resources are allocated separately for block-based and file-based objects, then each protocol can guarantee its performance requirements, but valuable storage resources remain underutilized due to stranded storage
Solution Approach 1:
The unified data path architecture implements universal storage resource allocation where the underlying file system manages a common pool of storage blocks that can be dynamically allocated to either block-based or file-based objects as needed. This eliminates stranded storage while maintaining performance guarantees through controlled allocation policies.
Solution Approach 2:
The storage resource allocation is made dynamic, allowing blocks to be reallocated from unused or freed objects to new objects of any type. The system continuously monitors storage usage and automatically reallocates resources to maintain high utilization efficiency while meeting performance requirements for active objects.
Data Source
AI summary
A technique forms contents of a compression container from a standalone container file system stored on the same data storage array on which the compression container is stored. Such a standalone container includes a container file which is formed by standalone blocks. The standalone blocks store host data and which are hierarchically related to each other. Upon receiving a compression command, the storage processor carries out a set of compression operations which include creating and storing the compression container in the data storage array. The compression container includes container files which are formed by compression blocks storing the host data and which are hierarchically related to each other.