The Durable Task Scheduler is a fully managed backend for Azure Durable Functions that enhances performance, reliability, and ease of monitoring of stateful orchestrations.
Updated Nov 18, 2024
Version 1.0What if I want to do 20,000 or 200,000 orchestrations per second. Is it a matter of scaling workers / instances. Right now, we use Service Fabric, which has been unofficially cancelled, and we're considering shifting the workload to durable entities / functions. Will massive scalability be an option, or should we be looking at a different approach?
I would consider investigating Microsoft Orleans (virtual actor framework) which has some interesting development happening about now for durable work flows.