Glossary
What is Velox?
In brief
Velox is an open-source C++ execution engine created at Meta, described by its project as “a high-performance, open-source execution engine designed for flexibility and reuse”. It is not a Spark accelerator on its own: Apache Gluten is the layer that plugs Velox into Spark, so that Spark plans run on Velox's native, vectorized operators.
How does Velox relate to Spark?
Spark plans the query; Gluten translates the physical plan for Velox; Velox executes supported operators natively and Gluten falls back to vanilla Spark for the rest. Gluten can also use ClickHouse as a backend.
How is Velox different from DataFusion?
Both are native execution engines used under Spark. Velox is written in C++ and reaches Spark through Gluten. DataFusion is written in Rust on Apache Arrow and reaches Spark through Comet, Auron and Flarion.
Related
Sources
- Velox · Fetched 27 Sep 2026
- Apache Gluten · Fetched 27 Sep 2026
- Apache DataFusion · Fetched 27 Sep 2026