Okay, so I converted a regular collection to a time-series collection and the compression abilities definitely worked (50%) of original size! I went to query the new collection and my normal queries took forever to return and my app was timing out. I’m using mongo v6 on high end AWS class machines (256GB RAM, 32 (V) CPUs). I chose what I thought would be a good shard key which was the obTimeUTC that I created + and id (which just looks like a UUID). I dumped 1/2 a TB into this collection and looked at the balance on shard 1, and it was 50% and shard 2 was 49.99%, so I thought I did alright. I of course created the index for that obTimeUTC so that my app could query using date ranges and so on. I ran “explain” on these queries (manually minus the app) and it is using the index, it’s just taking forever. What is my best bet? Start setting the profiling level-> db.setProfilingLevel(1, { slowms: 50 }) and hunt through logfiles?