Software Update Distribution via Local Sub-file Broadcasting
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current software update distribution methods for local machines are inefficient, leading to high network utilization due to redundant downloads and direct one-to-one connections, which can overwhelm network resources.
Innovation Solution
The method involves breaking software updates into sub-files with a descriptor file that includes hash codes for assembly, allowing one machine to download and broadcast these sub-files to others, reducing network load by utilizing local caching and peer-to-peer distribution.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If direct download methods (ftp/http) are used for software updates, then updates can be delivered to individual computers, but network traffic and network load increase significantly
Solution Approach 1:
The software update is divided into multiple sub-files (first sub-file, second sub-file, third sub-file, etc.). Each sub-file is distributed separately through different channels: some computers download from the network while others receive via local broadcasting. This segmentation allows parallel distribution and reduces the burden on any single network connection.
Solution Approach 2:
Update sub-files are broadcast through intermediate computers (second computer, third computer, etc.) that already have the sub-files in their memory. These intermediaries re-broadcast the sub-files to other computers in the network, creating a multi-hop distribution system that reduces direct network traffic between the source and all destination computers.
2Reliability
If one-to-one connections are used for update distribution, then each computer receives updates individually, but the number of connections and network overhead increases
Solution Approach 1:
Computers that receive update sub-files (either through network download or local broadcast) automatically become distribution nodes themselves. They store sub-files in their memory and broadcast them to other computers without requiring centralized coordination. This self-service approach simplifies connection management while maintaining reliable update delivery.
Solution Approach 2:
Multiple distribution channels are merged into a single coordinated system. Computers can receive updates through either network download or local broadcasting, and both methods work together seamlessly. The system combines the reliability of direct network delivery with the efficiency of local peer-to-peer sharing.
3Speed
If all computers download updates simultaneously from the network, then update speed is maximized for each computer, but network bandwidth is overwhelmed
Solution Approach 1:
The update distribution is segmented into multiple phases and channels. Some computers download sub-files from the network while others receive via local broadcasts. This segmentation allows simultaneous updates across multiple computers without overwhelming network bandwidth, as the bulk transfer happens through local broadcasting rather than repeated network downloads.
Solution Approach 2:
Certain computers (second computer, third computer, etc.) are pre-loaded with update sub-files through network download before the broadcasting phase begins. These computers then serve as local distribution nodes, enabling other computers to receive updates quickly via local broadcasting without consuming additional network bandwidth.
Data Source
AI summary
In embodiments of the present invention improved capabilities are described for a updating software in a plurality of devices coupled to one another in a communicating relationship through a local network, the method comprising receiving a descriptor file for a software update at the first device from a remote source outside the local network, the descriptor file including a hash code for each of a plurality of update sub-files and an order for assembling the update sub-files into the software update, downloading the plurality of update sub-files to the first device from a remote source outside the local network until the sub-files identified in the descriptor file are present on the first device, where a presence of the sub-files is evaluated using the hash codes in the descriptor file, and broadcasting every one of the plurality of update sub-files from the first device to the number of other devices.