Local Server Caching for Software Update Distribution
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current data transfer mechanisms in networks are inefficient, particularly for software updates, as they rely on each machine obtaining separate copies from external servers, leading to network congestion and high bandwidth usage.
Innovation Solution
A method of data caching and distribution within a linear communication orbit, where machines execute software programs, a local server receives update metadata, and an update module evaluates software version information to determine necessary updates, requesting and applying software update files through the orbit.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If each computational machine separately obtains copies of software applications or updates from external servers, then each machine can receive updates independently, but network bandwidth consumption increases and network congestion occurs
Solution Approach 1:
The patent implements preliminary action by having machines contact external servers to retrieve software update metadata before actually downloading update files. This preliminary metadata retrieval allows machines to determine which updates are needed and enables subsequent efficient file distribution through the linear communication orbit, reducing redundant bandwidth consumption while ensuring reliable update delivery.
Solution Approach 2:
The patent introduces a linear communication orbit as an intermediary mechanism between external servers and computational machines. This orbit acts as a mediator that receives update metadata from external servers and distributes it efficiently among multiple machines, reducing direct bandwidth consumption from external servers while maintaining update reliability.
2Productivity
If hundreds or thousands of computational machines contact the same external server simultaneously for updates, then updates can be distributed to all machines, but network congestion occurs and update process efficiency decreases
Solution Approach 1:
The patent segments the update distribution process into two distinct phases: (1) metadata retrieval phase where machines obtain update information from external servers, and (2) file distribution phase where actual update files are exchanged through the linear communication orbit. This segmentation prevents simultaneous heavy traffic on external server connections while enabling rapid local distribution.
Solution Approach 2:
The patent transitions from a single-point update distribution model (all machines contacting one external server) to a distributed linear communication orbit model. This dimensional change from centralized to distributed architecture allows parallel update distribution across multiple machines without creating network congestion at a single bottleneck point.
3Loss of energy
If a linear communication orbit is used for data distribution, then network bandwidth consumption is reduced, but system complexity increases
Solution Approach 1:
The patent implements self-service by enabling computational machines to autonomously evaluate software version information against update metadata and independently determine which updates they need. Each machine actively participates in the update process by requesting only the updates it requires, reducing the need for complex centralized coordination while maintaining low bandwidth consumption through the linear communication orbit.
Data Source
AI summary
Software updates by computational machines in a distributed system are facilitated by a server system that receives a master update from a remote server outside to the distributed system. The server system, which is in the distributed system, generates from the master update a first group of update metadata comprising less than all of master update and sends to a first plurality of computational machines in the distributed system first update materials, including the first group of update metadata and first filter criteria specifying characteristics of computational machines that are to locally store the first group of update metadata. A respective computational machine that satisfies the first filter criteria locally stores the first group of update metadata, obtains first update files based on the first group of update metadata, and updates one or more software programs using the first update files.