Flowdapt

Easy orchestration of heterogeneous workflows

Large-scale adaptive modeling in real-time

Flowdapt is an Emergent Methods project for large-scale adaptive modeling challenges. It is highly adaptable to handle customized workflows (plug-ins) in parallelized environments. And the best part, is that Flowdapt is fully open-source

Distributed compute

Flowdapt is for large-scale cluster orchestration, particularly well suited for real-time adaptive modeling. The design principles of Flowdapt include:

🚲 Highly parallelized compute efficiency
🤖 Automatic resource management and sharing
🐞 Rapid (local) prototyping and debuggability
🔌 Intuitive cluster-wide data sharing methods
⏱ Easy scheduling for real-time applications
📝 Intuitive configuration and live configurability
🚚 Deployment cycle efficiency
🔬 Micro-service-first design
🕸 Kubernetes-style schema and behavior
🚀 Vanilla Python, swap effortlessly between Ray, Dask, or Local executors