WAN Accelerator Lock-Aware Prefetching
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
WAN accelerators operate conservatively due to locks and access control restrictions, reducing network performance by limiting data access and transmission in wide-area networks, as they must respect existing locks and restrictions to prevent data corruption.
Innovation Solution
A method to optimize network traffic by identifying application-specific behavior in data access requests, allowing for aggressive prefetching and buffering of data not subject to access restrictions, and modifying requests that are inconsistent with expected behavior to ensure data integrity.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If WAN accelerators respect locks and access control restrictions to prevent data corruption, then data integrity is maintained, but network performance deteriorates due to conservative operation and limited data access
Solution Approach 1:
The patent applies local quality by differentiating between locked and unlocked portions of files. WAN accelerators identify specific file regions that are not subject to access restrictions and apply aggressive optimization techniques (prefetching, buffering, compression) only to those unlocked portions, while respecting locks on other portions. This allows simultaneous data integrity protection and performance optimization by treating different parts of the data differently based on their lock status.
2Reliability
If WAN accelerators operate conservatively to respect access control restrictions, then data corruption is prevented, but network throughput decreases due to reduced data transmission optimization
Solution Approach 1:
The patent implements segmentation by dividing file access into locked and unlocked segments. The WAN accelerator analyzes file access patterns and identifies contiguous unlocked regions, then applies aggressive optimization (prefetching, buffering, compression) to these segments while maintaining conservative behavior for locked segments. This segmentation enables parallel optimization paths that preserve data integrity while maximizing throughput for safe-to-optimize data.
Solution Approach 2:
The patent applies preliminary action by prefetching and buffering data from unlocked file portions before they are actually requested by applications. The WAN accelerator proactively retrieves and prepares data in advance based on predicted access patterns, reducing latency and improving throughput. This preliminary optimization is applied selectively only to unlocked portions, ensuring data integrity while enhancing performance.
3Reliability
If WAN accelerators disable acceleration for locked files, then access control restrictions are respected, but application performance suffers due to lack of optimization
Solution Approach 1:
The patent applies local quality by enabling WAN acceleration for specific unlocked portions of files while maintaining access control compliance for locked portions. Instead of disabling acceleration entirely for locked files, the system identifies and accelerates only the unlocked segments, providing partial optimization that improves application performance without violating access control restrictions.
Data Source
AI summary
A data access request from an application for access to a data resource is received from a first application. The data access request is analyzed to identify application-specific behavior indicating a type of data access for the data resource. The WAN acceleration functionality of a first device is configured for network traffic optimization based on the type of data access for the data resource. The analysis of the data access request may be based on attributes of the data access request, the data resource, and/or an access control restriction. The network traffic optimization may be adapted to optimize network traffic to a first portion of the data resource that is different than a second portion of the data resource associated with the data access request or to a second data resource separate from the data resource associated with the data access request.