I’m using Timefold real-time scheduling in this setup:

• Two application nodes

• Requests are distributed via a VIP / load balancer (round-robin)

• Each node runs its own Timefold solver instance

• Nodes do not share the problem facts

here is a problem i have

I have a hard constraint such as:

An employee must not have more than 5 tasks per day

Because requests are handled by different nodes, each solver sees only its own in-memory data. As a result, the same employee can be assigned tasks independently on each node, and the constraint is violated (10 tasks per day)

now Is Redis a good and compatible choice to use with Timefold to share scheduling data or changes between nodes so that constraints (like max tasks per employee per day) are respected in a , multi-node environment?

I’m trying to understand whether using Redis with Timefold is a correct approach, or if Timefold requires a different architecture to handle this scenario.