Standalone · Tooling

cudf-spark: GPU-accelerated Spark

Spark is the workhorse of data engineering, and it's CPU-bound. NVIDIA's cudf-spark swaps the CPU executors for GPU ones, and the speedup is not subtle.

Spark is the standard for big data processing, and for a decade the standard has been CPU-bound. NVIDIA's cudf-spark changes the executor, not the API: your Spark jobs run on GPUs, and the speedup on the data-heavy parts is often an order of magnitude.

The idea

Spark's executors do the work: filter, join, aggregate, all in JVM-land on CPUs. cudf-spark replaces the CPU executor with a GPU one backed by cuDF, RAPIDS' GPU DataFrame library. Your Spark code stays the same; the hardware underneath changes.

What you get

Mental model

cudf-spark is like swapping the engine in your car without changing the steering wheel. Same driving experience, much faster.

Why it matters

Inference teams don't usually think about Spark. But the data that feeds your models, the logs you analyze, the metrics you aggregate, all of it flows through a data pipeline. When that pipeline is the bottleneck, GPU acceleration is the lever.

The takeaway

You don't need a new data platform, you need the same one running on the hardware you already own.

Sources

Back to the blog