Scalable Index Store for Object Storage Systems
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional object systems are not scalable and typically do not offer file access, and existing file systems cannot switch between file and object access efficiently.
Innovation Solution
A method and system for providing a scalable set of indexed key-value pairs that enable storage of objects across multiple arrays in a data storage environment, allowing for dynamic addition of arrays and balancing of data, using a Virtual Storage Pool and indexing service that supports fault tolerance, load balancing, and multiversion concurrency control.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If conventional object systems are used, then object storage is provided, but scalability is limited and file access is not offered
Solution Approach 1:
The patent implements a unified storage system that provides both object storage and file access capabilities through a single architecture. The indexing service and storage pools work together to enable multiple access methods (object-oriented and file-oriented) without requiring separate systems, thus achieving multi-functionality while managing complexity through integration rather than multiplication of components.
Solution Approach 2:
The patent introduces an indexing service as an intermediary layer between the storage pools and access interfaces. This indexing service manages the complexity of underlying storage structures and provides simplified access paths for both object and file operations, acting as a mediator that handles the complexity internally while presenting unified access to users.
2Reliability
If data is stored across multiple arrays, then scalability and fault tolerance improve, but system complexity increases
Solution Approach 1:
The patent implements self-service mechanisms where the indexing service automatically manages data distribution across storage pools without requiring manual intervention. The system performs automated metadata management, data placement, and recovery operations, allowing the distributed system to maintain itself and reducing the operational complexity despite the distributed architecture.
Solution Approach 2:
The patent divides the storage system into separate indexing services and storage pools, with further segmentation of data across multiple arrays within pools. This segmentation allows independent management and failure isolation - when one array fails, only the affected segment is impacted while other segments continue operating, maintaining reliability while managing complexity through modular design.
3Productivity
If dynamic addition of arrays is allowed, then system scalability improves, but data balancing complexity increases
Solution Approach 1:
The patent implements dynamic load balancing where the indexing service continuously monitors and adjusts data distribution as arrays are added or removed from the system. The system adapts its data placement strategies in real-time based on current system state, automatically rebalancing data when capacity changes occur without requiring manual reconfiguration or complex static planning.
Solution Approach 2:
The patent incorporates feedback mechanisms where the indexing service monitors the state of storage pools and automatically adjusts data placement decisions. When arrays are dynamically added, the system receives feedback about available capacity and redistributes data accordingly, maintaining balance through continuous monitoring and adjustment rather than complex pre-planning.
Data Source
AI summary
A method, system, and computer program product for providing, via a provisioning engine, a scalable set of indexed key-value pairs enabled to store objects in a data storage environment; wherein the data representing the objects is enabled to be spread across arrays in the data storage environment; wherein additional arrays are enabled to be added to the data storage environment and included in the indexed key-value pairs; wherein the data stored across the arrays may be balanced.