ReferenceGeometry

Published configurations

Paired reference-table and pointwise geometry Jacobian measurements.

Measured operation

Measures bulk Jacobians against the original pointwise fallback.

Both paths evaluate \(J_{ji}(\widehat x_q)= \sum_a X_{ja}\partial_i\phi_a(\widehat x_q)\) with the same vertex basis and accumulation order. The pointwise measurement explicitly calls the base transformation's bulk fallback; no second production path is added. Setup, warmup and exact entry comparisons are outside timing. Repeated calls retain output-vector capacity. Physical Jacobians are timed, but no physical-coordinate evaluation, FE assembly or solve is timed. Results are local diagnostics, not CTest or CI gates.

Recorded results

Each case retains its original inputs, iteration policy and correctness checks. Raw reports preserve counters, repetitions and the Google Benchmark context. CPU, wall and manually supplied times are labeled separately; unlike operations and configurations are not averaged together.

Timing history

RodinReferenceGeometryBenchmarks · auto · regular · ranks: 1

Benchmark machine and measurement conditions

Published CI benchmarks run on standard GitHub-hosted ubuntu-24.04 virtual machines for public repositories.

System	CPU	Memory	SSD
Ubuntu 24.04, x64	4 vCPUs	16 GB	14 GB

The benchmark builds use GCC 14 and Release mode. Each job runs one benchmark executable at a time. OpenMP, OpenBLAS and MKL use one thread; MPI ranks share the same runner. The selected configuration identifies the build and rank count.

The CPU model and host load can vary between virtual machines, so small timing differences should be checked across repeated runs. GitHub runner specifications describe the available hardware. The source workflow is BenchmarkGroup.yml.