Slow Rancher Redeploy

We have around 50 stacks of 10 microservices each which all need to be updated during a redeploy. We do a rolling deploy of 3 stacks at a time in order to somewhat parallelize this process. However, during the rolling deploy the service startup and the Rancher UI get supppper slow. But the CPU and RAM usage statistics on the Rancher server aren’t near capacity. Anyone have an idea what’s up?