@Volker Hi, you can simply run it on your local computer where you define the number of threads Spark can use (e.g. -Dspark.master=local[8]). Running it on a cluster or cloud requires more setup. But I'll also try to add these optimizations to BigStitcher directly!