Basic Architecture

Cloud Deployment

This is the recommended way to use Heisenware. In this scenario your apps will be hosted by us and run in the cloud. However, we have a multitude of possibilities to bring you local code or your local machines into the game.

Simplified architecture of the Heisenware ecosystem when used as a cloud deployment.

Platform

All essential components of our software are running in the cloud. Once you built and deployed an app, it is accessible from any place on the planet that provides you an internet connection. The apps themselves are so called PWAs (Progressive Web Apps) and run on any operating system and on any device without feeling different to "native" apps.

With this you can already connect to other entities also placed in the cloud, but you could not see how your machines worked over night during your breakfast coffee at home. Fortunately, we invested a lot of effort to make exactly this possible and also very easy.

In the following lets look at the different options of bringing either connectivity or even custom code that executes only available at your local premises into the cloud.

Native Agent

The native agent allows you to utilize various protocols (like S7, Modbus, OPC UA, MQTT) to connect your devices from within your local network (in industrial context often called OT network). It hence must also run on some local hardware that - from a network perspective - can reach the device required to be connected.

The native agent in the end is just a piece of software that can be started in one click and directly runs on your operating system - it simply is a single Linux/MacOS binary or a Windows .exe. Security is baked in, as you download the agent from your platform (to which you are currently logged in) and it is indeed freshly compiled (!) only for you, with your very personal set of credentials. Other Heisenware users will have other agents that are inherently incompatible with your account on the platform. Please read here for more details on this topic.

Docker Agent

This is a new feature shipping first time with v88 - Almost Forever!

This is much like the native agent packed into a docker container. There are times where a docker environment instead of a native executable is preferrable to connect your edge. Docker technology renders especially useful for all those vendors that offer edge-connectivity hardware (such as Siemens, WAGO, Hilscher, Welotec, Weidmüller, etc.). We offer our docker-based agent for all relevant architectures (amd64, arm64, arm/v7) and you get started safely and quickly by providing the necessary credentials as environmental variables to the container. See here for all the details.

Native Custom Code Adapter

While the agents discussed above, already ship pre-made code by Heisenware - typically to utilize industrial protocols - not so for the code adapters. Think of the code adapter like a Heisenware specific envelope that fits any of your custom source code and ships it to be directly usable as low-code blocks within your cloud platform. Like the on-prem agent, the code adapter comes in a native and a dockerized fashion (which we also call Custom Extension).

The native code adapter relies on programming language specific versions of our powerful VRPC library.

Docker Custom Code Adapter

We are providing a starter-project which you can download to build a docker image that will seamlessly talk to the platform but will be equipped with your custom code and algorithms. Once built, you can decide where you want to run the corresponding container. You essentially have two options:

1. Inside the platform This is accomplished by loading your image as a so-called Extension. In this case the platform will take care about it's lifetime and automatically persist any files into the central shared folder. Your code now runs as part of the Heisenware cloud.

2. Outside the platform This comes in very handy, whenever you need to bridge a private/local network. In this case your code could for example interact with a device only accessible locally (on-premises) but your are still enabled to control everything as usual from within the platform, seamlessly. Read here for details.

On-Premises Deployment

In the on-premises deployment, you get the entire platform to run on your local servers (or private cloud). It is like every component moves one level down. Now the on-board connectivity of our platform is ready (without the need for agents) to directly connect your local devices. In case you are having a large shop-floor setup with segmented networks, you are still on the safe site as now the former agents can act to bridge lower-level networks.

Please contact us the get started with an on-premise installation. But rest assured that this is a simple process and won't take neither much time nor hassle. See here for more details.