In order for #datahike to get durability for ClojureScript in the browser, they had to make the query logic/api asynchronous, so that would be another way to go (that doesn't involve dedicated threads)