alt Hacker NewsWe ran some benchmarks (TPC-H, designed for OLAP) with ~10M records https://github.com/BemiHQ/BemiDB#benchmark
The BemiDB storage layer produced ~300MB columnar Parquet files (with ZSTD compression) vs 1.6GB of data in Postgres.
We ran some benchmarks (TPC-H, designed for OLAP) with ~10M records https://github.com/BemiHQ/BemiDB#benchmark
The BemiDB storage layer produced ~300MB columnar Parquet files (with ZSTD compression) vs 1.6GB of data in Postgres.
Does TPC-H SF1 really take _one and a half hours_ for you on regular Postgres? Last time I tried (in the form of DBT-3), it was 22 queries and most of them ran in a couple seconds.