Small benchmarks for various parts of the model codebase. Should be used intermittently to ensure against regression in performance.
Benchmark | Description |
---|---|
Iterations | Tests the effects of alignment and struct sizes on raw iterative loops. Utilised to motivate the decision behind using Option<NonMaxU64> for smaller sizes. |
Routing | Rudimentary insights into direct and transit routing times, as well as cost of selecting nearby transit nodes on the GranularGrid |
Event Loop | Runs the full event loop and checks the time to simulate a day on various sizes of test Synthetic Environments |
cargo bench --bench $BENCH_NAME
where$BENCH_NAME
is any of: [iterations
,routing
,event_loop
]