Do GPUs Really Need New Tabular File Formats?
For GPU-accelerated data processing systems, this work identifies and resolves a performance bottleneck in Parquet scans without changing the format.
The paper shows that Parquet's poor GPU performance is due to suboptimal CPU-oriented configurations, not the format itself, and achieves up to 125 GB/s effective read bandwidth with GPU-aware configurations.
Parquet is the de facto columnar file format in modern analytical systems, yet its configuration guidelines have largely been shaped by CPU-centric execution models. As GPU-accelerated data processing becomes increasingly prevalent, Parquet files generated with CPU-oriented defaults can severely underutilize GPU parallelism, turning GPU scans into a performance bottleneck. In this work, we systematically study how Parquet configurations affect GPU scan performance. We show that Parquet's poor GPU performance is not inherent to the format itself but rather a consequence of suboptimal configuration choices. By applying GPU-aware configurations, we increase effective read bandwidth up to 125 GB/s without modifying the Parquet specification.