Papers lose me at the equations.
Read the abstract, the conclusion, and the experimental setup first, and skip the proofs entirely on the first pass. Most of the transferable content of a systems paper is in what they measured and what surprised them.
Finished Designing Data-Intensive Applications. What next?
Papers, in small doses: the Dynamo paper for the tradeoffs, Raft for consensus you can actually follow, and the Google SRE book for how it fails in practice. Another textbook at this point adds vocabulary rather than understanding.
How long does this usually take in practice?
An afternoon to do it properly the first time, then minutes when it recurs. Most of the first afternoon goes into discovering which of the defaults do not match your situation.