Notes · Figures checked September 2026
Notes on Spark ETL acceleration
In brief
Notes are dated, sourced articles on questions that come up when teams evaluate Spark accelerators and that do not fit a single review. The newest cover a buyer's checklist, what each product publishes about price, how to run a proof of concept, six common claims checked against documentation and Iceberg compaction as an accelerator test.
A buyer's checklist for Spark ETL accelerators
Twenty questions, in five groups, to answer before the first trial run.
- Buying
- How-to
What Spark accelerators cost: what each vendor publishes about price
Some products have a cost you can model from public numbers, and some need a quote. Here is what is published, and how to compare the rest.
- Buying
- Pricing
How to run a Spark accelerator proof of concept on your own ETL pipeline
A trial plan that produces a decision, not a demo: one real pipeline, a logged baseline, three cluster sizes and a threshold set in advance.
- How-to
- Buying
Six common claims about Spark accelerators, checked against public documentation
What vendor and project pages actually say about code changes, platforms, hardware, coverage, benchmarks and cost.
- Buying
- Benchmarks
Iceberg compaction on Spark: why it is expensive and what to measure
Compaction reads, sorts, shuffles and rewrites table data on a schedule, which makes it a good test of an accelerator's shuffle and spill claims.
- Workloads
- Benchmarks
Apache DataFusion under Spark: how Comet, Auron and Flarion share one engine
Three of the seven accelerators we track run on the same Rust engine. What that means for features, releases and buyers.
- Engines
- Open source
Spark 4.x support in Spark accelerators: where each one stands in September 2026
Accelerators replace internals that change between Spark releases, so each one lags the Spark release calendar by a different amount.
- Releases
- Compatibility
Spark accelerator news, June to September 2026: what changed
Four months of releases, published results and technical writing, grouped by theme, with the source for every item.
- News
- Releases