I'm currently working on a project that requires me to make a large distance matrix using houses from Indiana. I've previously used datasets with ~2000 locations scaled up to 13,000 to simulate weighing, but when I tried using a dataset of 12,500 actual houses, my osrm backend constantly uses too much memory and kills osrm-routed. I've allocated 10gb to my docker's WSL2 program, and am only using the indiana-latest osrm set. I was wondering if there was something specifically wrong with my backend server, as the original indiana file was only around 150mb.