You are right, this was one of the first problems I have noticed with SqlPools. The elastic architecture we have used to rebuild
#synapse #sql allows us to work well at any scale, but tuning the operating parameters takes time. And smaller to larger seems easier than the opposite