Link-Server Caching for Cloud Storage Bandwidth Optimization
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Cloud storage providers face high costs due to bandwidth and hardware requirements for maintaining redundant storage arrays and large bandwidth to facilitate quick file downloads, necessitating efficient bandwidth optimization solutions.
Innovation Solution
A computer system optimizes bandwidth usage by intelligently caching and decoding digital files within a local-area network, dynamically adjusting thresholds based on connection speed and latency to local storage nodes, and using parity information for error correction, thereby reducing reliance on external bandwidth.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If cloud providers maintain redundant storage arrays and purchase large amounts of bandwidth to facilitate quick file downloads, then service reliability and download speed are improved, but hardware costs and bandwidth costs increase significantly
Solution Approach 1:
The patent segments files into multiple data blocks and distributes them across multiple storage nodes. This allows the system to maintain reliability through redundancy while reducing the need for expensive dedicated hardware at each node, as each node stores only a portion of the data
Solution Approach 2:
The patent creates multiple copies of data blocks across different storage nodes and uses parity information for redundancy. This copying approach enables reliable data retrieval without requiring expensive dedicated hardware at each location
2Speed
If cloud providers purchase large amounts of bandwidth to facilitate quick file downloads, then download speed is improved, but bandwidth costs increase significantly
Solution Approach 1:
By segmenting files into data blocks distributed across multiple nodes, the system enables parallel download of multiple blocks simultaneously, achieving high download speeds without requiring a single high-bandwidth connection
Solution Approach 2:
The patent allows clients to download only the specific data blocks they need rather than entire files, and enables recovery of complete files from partial block sets using parity information, reducing overall bandwidth consumption
3Reliability
If cloud providers maintain redundant storage arrays, then data protection and reliability are improved, but hardware costs and system complexity increase
Solution Approach 1:
The patent introduces a coordinator entity that manages the distributed storage system, handling block allocation, parity generation, and data recovery coordination. This intermediary approach simplifies the overall system architecture compared to fully distributed self-organizing systems while maintaining redundancy and reliability
Solution Approach 2:
The system uses parity information and erasure coding parameters to enable data recovery from partial block sets, allowing flexible redundancy configurations that balance reliability requirements with system complexity and hardware costs
Data Source
AI summary
A computer system for optimizing bandwidth usage within a cloud storage system identifies distinct requests to access a particular digital file. When the number of requests exceeds a threshold, the computer system queries a remote directory server for the particular digital file. The query system then receives, from the remote directory server, addresses to data blocks stored within multiple remote storage nodes. The system requests at least a portion of the addressed data blocks. Upon receiving the blocks, the system decodes the particular digital file from the portion of the addressed data blocks. The computer system then stores the decoded digital file within local memory and communicates the local-area network address to the remote directory server.