In an environment where your DB may be combined with an Elasticsearch cluster, caches, etc... how do you realistically use branching features in a DB? I love the idea, but idk how I could test if my ES cluster doesn't have a similar branching feature.