Distributed Polars beats a single node only on I/O‑bound queries
On a 1TB TPC‑H benchmark, distributed Polars across 32 m8i.xlarge nodes edged out a single m8i.32xlarge only on I/O‑heavy queries, hitting 400 Gbps burst network bandwidth. For compute‑intensive joins the single node still wins, showing distributed Polars is not a universal speedup.