16 May 2020 | Andreas Müller
Flow Director natively supports MQTT to connect IoT devices and has the analytics and visualization capabilities required here. It is also ultra-scalable by just connecting more SwiftMQ Routers. These routers can even run directly on the IoT gateways.
Usually, IoT edge solutions have a cloud part to run long-term analytics or use the data in other applications. Most processing takes place at the edge, but with forwarding of aggregated or pre-processed data to the cloud.
Another use case is the direct forwarding of the data from the IoT device to the cloud. An edge solution isn't required here.
An IoT device must be provisioned at a cloud provider before you can send any data; otherwise, it is unknown. For security reasons, it requires to generate certificates and private keys, install them on IoT gateways, test it, and locate the gateway in a network zone with cloud access. A lot of costly and annoying work.
Flow Director solves this problem by providing Automatic Cloud Provisioning for the three major cloud providers Amazon AWS, Micosoft Azure, and Google GCP.
Your IoT gateways connect via plain MQTT to Flow Director, and the client id of that connection is automatically provisioned as device at the cloud during the first message sent. The gateway keeps sending while Flow Director retains the messages until provisioning is finished and then forwards them to the cloud. After that, messages are sent directly to the cloud.
There is only a one-time setup to enter and test the cloud credentials in Flow Director. After that, it works in the background. There is no change in the IoT gateway.
It has never been easier to bring IoT device data into the cloud!
In Realtime Process Mining, the process model is the brain of your application from a business perspective. It knows everything about your processes. We have extended Process Analyzer by configurable events for process stages, links, and alerts. We have also added several new components that provide access to the internal process model.
You can now react to those events and use the data from the process model to create actions. For example, in our Warehouse simulation, we have added a customer notification when his shipment will arrive by predicting an estimated time between two stages.
Apache Pulsar is an open-source distributed pub-sub messaging system originally created at Yahoo. It gets more and more traction and sees itself as alternative to Apache Kafka.
We have added full bi-directional integration by providing Connection, Inbound, and Outbound flow components for Apache Pulsar.
The SwiftMQ Router that is required by Flow Director, and which previously were a separate install, is now integrated into Flow Director and runs in-process. You don't need to touch configuration files anymore but start everything by a single command. The name of the embedded router is flowdirector.
Instead of connecting Flow Director directly to your router network, you now connect the embedded router via routing connection, which is much simpler and can be done from a flow or the Explorer app.
Flow Director requires SwiftMQ release 12.1.0 or later at the connected network. For older versions, you can still use the bridge flow components.
Flow Director now comes with the Flow Director Management app only, which provides access to the new cloud-based App Repository to install and update other apps on demand.
Flow Director 2.0 is a major milestone that proofs it as a universal platform for everything. It is easy to install and use, ultra-scalable without an increase of license costs, and elevates messaging to a new level. Try it out!
Andreas is a well-known messaging expert, creator of SwiftMQ and in his side job CEO of IIT Software GmbH. He leads the overall development of Flow Director and has an eye on any piece to fit into the whole picture.