Peer-to-Peer Software Update Staging via BMC
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The existing process of updating software in computing systems is time-consuming and inefficient, with a significant portion of the update process requiring downtime and high network bandwidth, leading to increased load on management controllers and potential single points of failure.
Innovation Solution
Implementing a method where multiple computing systems can stage and share software updates with each other via a network protocol, allowing them to organize and pre-stage updates before the actual installation, thereby reducing the load on management controllers and minimizing downtime.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If software updates are staged from a central management controller, then updates can be distributed to computing systems, but the management controller experiences high load and becomes a single point of failure
Solution Approach 1:
The patent segments the centralized update distribution function into distributed peer-to-peer update sharing among computing systems. Each computing system can stage and share updates with others directly via network protocols, eliminating the single-point bottleneck at the management controller while maintaining update distribution capability.
Solution Approach 2:
The patent introduces an intermediary network protocol layer that enables direct communication and file sharing between computing systems. This intermediary mechanism allows systems to exchange update files peer-to-peer without overloading the management controller, reducing its role to coordination rather than actual file transfer.
2Productivity
If software updates are installed immediately, then systems receive current functionality, but significant downtime is required during the update process
Solution Approach 1:
The patent implements preliminary action by enabling computing systems to stage update files in advance during idle periods or low-utilization times. Systems can download and prepare update files before they are actually needed, so when installation is required, the files are already ready and local, minimizing installation downtime.
Solution Approach 2:
The patent maintains continuity of useful action by allowing update staging to occur during system idle periods or low-utilization windows. Rather than stopping operations for updates, the system continuously performs useful work (staging updates) during otherwise idle time, ensuring updates are ready when needed without interrupting productive operations.
3Productivity
If large update files are transferred over the network, then systems can be updated, but high network bandwidth is consumed
Solution Approach 1:
The patent merges the update distribution function into the existing peer-to-peer network infrastructure already present in data centers. By utilizing existing network connections and protocols between computing systems, the solution avoids dedicated high-bandwidth update channels while effectively distributing large update files through combined network resources.
Solution Approach 2:
The patent implements copying by having computing systems create local copies of update files and share them with peer systems. Instead of repeatedly transmitting the same large files from a central source, each system maintains a local copy and can serve as a distribution point for others, reducing overall network bandwidth consumption through efficient file replication.
Data Source
AI summary
Example implementations relate to software updates of a plurality of computing systems. An example system includes a plurality of computing systems including a first computing system and a second computing system. The first computing system includes a first board management controller and the second computing system includes a second board management controller. The first computing system and the second computing system are capable of staging from each other through the first board management controller and the second board management controller via a network protocol.