Evals

Engine comparisons

Compare benchmarked data libraries and language implementations.

How it works

Choose a benchmarked operation and compare the engine or language results shown for that workload. Read the workload notes before applying the result to another dataset.

Choose Bars to compare the engines visually. Choose Table to read the underlying values. Each column shows the measured result for one engine.

Related research

CSV load timing in Zig and n0is C++
See how the study timed first and warmed loads separately, with file reading outside the benchmark clock.
How CSV loading was compared with Swift
See the file, output checks, and dictionary work used to compare the two loaders.
Measuring filters, sorts, and grouped sums
See how the work included in each measurement changes the comparison.

Example

Start with
Grouped totals on a fixture
Try this
Compare engine results
Result
See the measured differences