Peer-Assisted Update Distribution in Device Management
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing device management platforms face high network traffic and prolonged deployment processes when updating large numbers of managed devices, due to the large size of updates and the resulting load on management servers, which can lead to failed downloads and extended update times.
Innovation Solution
Implementing a peer-assisted update deployment method where updated managed devices act as repositories to distribute updates to others, reducing the load on the management server and on-premise servers by allowing managed devices to share the update load among themselves.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If the management server sends update commands to all managed devices to download updates from the on-premise server, then all devices can receive updates, but network traffic and server load increase significantly
Solution Approach 1:
The system segments the update distribution task by dividing managed devices into multiple groups, where each group downloads updates from different master devices rather than all devices downloading from a single on-premise server. This segmentation reduces the concentration of network traffic and server load on any single source.
Solution Approach 2:
Updated managed devices serve as intermediary master devices that relay update content to other managed devices. These intermediary devices buffer and redistribute update data, reducing direct traffic to the on-premise server and managing network load more efficiently.
2Loss of energy
If the management server places a limit on the number of managed devices that can concurrently download an image, then server load is reduced, but the imaging process is prolonged
Solution Approach 1:
The system segments the large imaging task into smaller sub-tasks distributed across multiple master devices. Each master device handles a portion of the managed devices, allowing parallel processing of image deployments without overwhelming any single server.
Solution Approach 2:
The system adds a spatial dimension to the update distribution architecture by introducing multiple master devices across different locations or network segments. This multi-dimensional distribution allows concurrent downloads from multiple sources, increasing overall throughput without increasing individual server load beyond acceptable limits.
3Productivity
If no limit is placed on concurrent downloads from the management server, then deployment speed increases, but the deployment process can still take substantial time in large environments
Solution Approach 1:
The system performs preliminary actions by having a subset of managed devices download and store update content in advance, becoming master devices before the actual deployment begins. This preliminary caching enables subsequent devices to download from local masters rather than waiting for remote server availability.
Solution Approach 2:
Managed devices that have received updates serve themselves and others by acting as master devices for peer-to-peer distribution. This self-service mechanism reduces dependency on the central on-premise server and enables autonomous update propagation throughout the managed device population.
4Productivity
If a managed device acts as a master device for multiple update commands, then update distribution efficiency increases, but the master device may become overloaded
Solution Approach 1:
The system segments the workload of master devices by distributing update commands across multiple master devices rather than concentrating all requests on a single device. This segmentation prevents any one master device from becoming overloaded while maintaining efficient peer-to-peer distribution.
Solution Approach 2:
The management server monitors the performance and load of master devices, using feedback information to dynamically adjust which devices serve as masters and how many commands each handles. This feedback mechanism ensures master devices operate within reliable performance thresholds while maintaining overall distribution efficiency.
Data Source
AI summary
Peer assisted updates can be provided in a device management environment. When it is desired to deploy an update to a group of managed devices, an update command can be sent to a first set of managed devices in the group instructing these managed devices to download the update. Once these managed devices have downloaded and installed the update, they will notify the management server. In response, the management server can instruct these managed devices to remain awake so that they may each function as a master device or repository for deploying the update to other managed devices in the group. The management server can then send update commands to the other managed devices instructing them to download the update from one of the managed devices that is now acting as a master device. This process can be repeated until all managed devices in the group have been updated.