Hello All! Could someone explain why (or provide material to why) presto is the recommended FileSystem to use for checkpointing? In addition, if using both hadoop and presto, is there no way to force a specific default for the base

s3://

schema?

Might be related w.r.t retrieval speed.