IPv6 Distributed Storage System Architecture
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Traditional distributed storage systems face inefficiencies due to heavy layering, complexity, and scalability issues, with bottlenecks shifting from disk performance to network bandwidth and CPU, requiring improved data management processes.
Innovation Solution
A distributed storage system with a minimal number of layers, utilizing IPv6 addresses for metadata and storage nodes, allowing direct communication between the application layer and storage system, and leveraging segment routing for flexible and scalable data management.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If traditional layered storage system architecture is used, then data management functionality is comprehensive, but system complexity increases and scalability is reduced
Solution Approach 1:
The patent extracts and removes intermediate software layers (filesystem layer, protocol layers) from the traditional storage architecture, allowing applications to communicate directly with storage nodes using native IPv6 addresses. This extraction eliminates the complexity of multiple software layers while preserving the essential data management functionality through direct peer-to-peer communication between clients and storage nodes.
Solution Approach 2:
The patent segments the monolithic layered architecture into independent, loosely-coupled components: metadata servers that provide namespace management, storage nodes that handle actual data operations, and clients that communicate directly using IPv6 addresses. This segmentation allows each component to be scaled and optimized independently, improving both simplicity and scalability.
2Productivity
If multiple software layers are used for data management, then functionality is enhanced, but communication efficiency decreases due to additional processing overhead
Solution Approach 1:
The patent removes intermediate software layers that add processing overhead, enabling direct communication between clients and storage nodes. By eliminating the filesystem layer and protocol translation layers, data can be read and written directly to storage nodes using native IPv6 addresses, significantly improving communication efficiency while reducing the number of software layers involved in data operations.
3Speed
If disk performance is optimized through software engineering, then disk I/O performance improves, but the system assumes disks remain the bottleneck which may no longer be true with SSD technology
Solution Approach 1:
The patent creates a universal storage architecture that works equally well with different storage technologies (HDD, SSD, etc.) by removing assumptions about disk performance being the bottleneck. The system uses native IPv6 addressing and direct communication protocols that are technology-agnostic, allowing the same architecture to optimize for SSD performance without requiring software engineering solutions designed for mechanical disk bottlenecks.
4Ease of operation
If traditional storage architecture with many layers is used, then data management is comprehensive, but configuration and optimization efforts are required
Solution Approach 1:
The patent extracts and eliminates the complex layered software architecture that requires extensive configuration and optimization. By allowing direct IPv6-based communication between clients and storage nodes, the system removes the need for filesystem layer configuration, protocol translation settings, and multiple intermediate software components, dramatically reducing configuration effort while maintaining comprehensive data management capabilities.
Data Source
AI summary
The disclosure provides a system, method and computer-readable storage device embodiments. Some embodiments can include an IPv6-centric distributed storage system. An example method includes receiving, at a computing device, a request to create metadata associated with an object from a client, creating the metadata based on the request and transmitting the metadata and an acknowledgment to the client, wherein the metadata contains an address in a storage system for each replica of the object and wherein the metadata can be used to write data to the storage system and read the data from the storage system. There is no file system layer between an application layer and a storage system layer.