Lets say I have 2 spicedb pods I send
# spicedbt
tilting_irons
09/29/2023, 3:08 PMLets say I have 2 spicedb pods, I send one request, and it lands on pod A and caches there
Then I spam send the same request, assuming the initial landing distributes evenly to pod A and B, but every request that B got is now going to A from B because it is cached on A. Is this a correct assumption?
v
vroldanbet
09/29/2023, 3:15 PMmostly. dispatches across pods happens based on subproblems. So PodB will dispatch just like Pod A, and it is the subproblems something is decomposed into that get dispatched to other nodes
vroldanbet
09/29/2023, 3:16 PMso if you hit the 2 pods with the same request, eventually what you will see is that both pods will be returned the cached response within the default quantization window of 5 seconds