Skip to content

README.md

Latest commit

 

History

History
70 lines (54 loc) · 3.23 KB

README.md

File metadata and controls

70 lines (54 loc) · 3.23 KB
Raw

Ahtola performance benchmarks

This is the canonical BenchmarkDotNet suite for Ahtola. It compares equivalent managed Ahtola workloads with Microsoft.Data.Sqlite in the same run and emits machine-readable results for Ahtola-to-Ahtola historical comparisons.

The workload inventory and upstream provenance are documented in docs/turso-benchmark-portability.md. Rust-only benchmarks are not built or executed.

Run profiles

Run from the repository root with PowerShell 7:

./build.ps1 benchmark -BenchmarkProfile smoke
./build.ps1 benchmark -BenchmarkProfile write-short
./build.ps1 benchmark -BenchmarkProfile coverage
./build.ps1 benchmark -BenchmarkProfile full
./build.ps1 benchmark -BenchmarkProfile large
./build.ps1 benchmark -BenchmarkProfile diagnostic -BenchmarkFilter '*Write*'

smoke executes every case once and produces no useful timing. write-short is the normal write-path development loop. coverage uses BenchmarkDotNet's Short job across the suite. full uses the default statistically adaptive job. large selects benchmarks explicitly categorized as Large. diagnostic collects an EventPipe CPU trace for a narrow filter.

Results are written below artifacts/benchmarks/<run-id>/, including raw BDN reports, normalized JSON, environment metadata, and a historical comparison when -BenchmarkBaseline names a prior normalized result.

.NET hotspots

DotNetHotspotBenchmarks measures the access patterns typical ADO.NET, Dapper and EF Core applications issue, through the Microsoft.Data.Sqlite-compatible facade against Microsoft.Data.Sqlite on the same SQLite-written WAL file (synchronous=NORMAL). Each case is parameterized by Engine (Sqlite/Ahtola):

./build.ps1 benchmark -BenchmarkProfile coverage -BenchmarkFilter '*DotNetHotspot*'

For the optimize-measure loop, an in-process stopwatch runner prints the Ahtola/SQLite ratio of every case in about a minute (optionally filtered by case name, with a per-case time budget in milliseconds and a single engine):

dotnet run -c Release -f net10.0 --project src/Benchmarks/Ahtola.Benchmarks.csproj -- --hotspots-quick
dotnet run -c Release -f net10.0 --project src/Benchmarks/Ahtola.Benchmarks.csproj -- --hotspots-quick Lookup 2000 Ahtola

It is a development aid: BenchmarkDotNet results remain the source of record.

Interpretation

  • Native SQLite ratios provide context; they are not a regression gate.
  • Compare Ahtola history only on equivalent hardware, OS, runtime, journal mode, durability mode, and fixture version.
  • Statement-only cases exclude fixture creation, connection open, and reset. End-to-end and recovery cases include those costs intentionally.
  • Allocated is managed allocation per operation. File, WAL, SHM, and logical log sizes are returned or recorded by macrobenchmarks where they matter.
  • Default parameter sets are suitable for development and scheduled coverage. Expensive upstream scales are isolated in the Large category.

Before trusting a new benchmark, run smoke, confirm SQLite/Ahtola result equivalence, inspect its generated report, and ensure repeated invocations do not accumulate mutable state unless growth is the workload under test.