Splicebench
Benchmark space100 versions · lower is better ↙
DRAG TO ORBIT · SCROLL TO ZOOMLOGARITHMIC AXES
The 3D library could not load. Check your internet connection and reopen this file.
About the benchmark
- Each point represents the same 100-version history. All three axes are logarithmic; less time and storage is closer to the shared origin.
- Distance measures how far a point is from (1 ms, 1 ms, 1 KiB), using equally weighted, normalized logarithmic axes.
- For version stores, rebuild retrieves the latest version from the archive. CLI startup and normal storage work are included.
- For delta utilities, rebuild sums 99 sequential patch applications from the initial base; storage excludes an index and checkpoints.
- Git is shown unpacked.
- Synthetic sparse edits, warm cache, CLI startup included; median of three trials.
- Machine: MacBook Air, Apple M4 (10 cores: 4 performance and 6 efficiency), 16 GiB memory, macOS 26.0.1.
- Hardware, operating system, and background load affect timing. Hardware alone should not change archive sizes for the same inputs, tool versions, settings, and metadata; different metadata or software versions can change sizes.
- DNAB's patch and rebuild operations do not use application-level multicore workers. Some scanning uses ARM NEON SIMD on a single core; this is based on source inspection, not CPU profiling.
- The histories were generated for this benchmark, not taken from a Git repository. Read the benchmark scripts: version stores and delta utilities.
- The delta chain utilities store a chain of files. Their much higher time comes from process startup time for each version and from reading/writing all the files files to/from disk. Mitigating this is done by putting it in a single-file storage utility such as
nab