Note:
Web Vision distributed architecture is available from N3uron version 1.22.4 onwards.
Web Vision supports distributed architectures using Dynamic Data Retrieval (DDR): a central node provides a single access point and dynamically retrieves data from multiple distributed Web Vision instances. This distributes the workload and delivers significant scalability for large for large deployments.
A distributed architecture requires no separate or dedicated module; it is built on the same Web Vision module. A single Web Vision project resides on the central node, referred to here as the Central Web Vision node, which retrieves the tag models and data of the connected distributed nodes over HTTPS. These nodes do not need their own Web Vision project, just an instance of the Web Vision module to enable communication in the distributed architecture.
A standalone Web Vision project interacts solely with the local tag model of the node on which it is installed. Within a distributed architecture, by contrast, the Central Web Vision node can additionally access the tag models and both real-time and historical data of the connected distributed nodes.
.webp)
How It Works
A distributed architecture is configured in a simple and intuitive way for the Central Web Vision node to access the tag models of distributed nodes. Starting from the Central Web Vision node, the accessible data expands outward across the network.
The Central Web Vision Node
The Central Web Vision node can visualize:
Its own local tag model.
The tag model of any distributed node declared in its configuration.
Distributed Nodes and Beyond
Each distributed node can expose its own local tag model, or a tag model that also consolidates data from further downstream nodes — via N3uron Links . All of this data is accessible seamlessly and can be visualized from the Central Web Vision node.
This cascades outwards without limit:
Central Web Vision Node
│
├──► Distributed Node A
│
├──► Distributed Node B
│
│
└──► Distributed Node C ...A single Central Web Vision node can therefore visualize an entire distributed network of nodes, regardless of how large or complex the underlying architecture is.