======================================= Linkcell neighbours and carried updates ======================================= A path under ``benchmarks/`` on this page names a record of the d-SEAMS 2.0 benchmark campaign. The records and the Snakemake rules that write them are in the private paper repository and the Hugging Face dataset ``d-SEAMS/dseams2-benchmark-records`` (``docs/records/README.md`` names the revision); the scripts under ``scripts/`` that time a run stay here. Build the cutoff neighbour list, then keep rings when only part of the next frame moved. The scaling argument is `Linkcell strong scaling <../explanation/linkcell-scaling.rst>`_. The list itself is `Cutoff neighbour lists `_. The cutoff list --------------- ``linkcell::pairs_within`` owns the cutoff list when the build links linkcell. vesin is the fallback. The orthorhombic list is one shared CSR. Each pair is written once, from the lower slot, on the z layer that owns that atom. Even layers run together. An odd last layer wraps onto layer 0, so it runs only after the even layers have stopped, and the odd layers run after that. One thread walks the same half stencil. The pair set does not depend on the thread count. Published CHILL+ and cutoff q6 read this list, as do q4 and the ring census. The half list keeps each pair on the lower index. The test that locks this is the Catch2 case ``half cutoff list is the lower index of the full list``: .. code:: cpp // excerpt: tests/test_neighbours.cpp REQUIRE(nneigh::orthoFullCsr(cloud, nullptr, 3.5, fullOff, fullCols, false)); REQUIRE(nneigh::orthoFullCsr(cloud, nullptr, 3.5, halfOff, halfCols, true)); A skin reuses the wide list while every atom stays inside the Verlet shell. Bonds are the pairs still inside the cutoff. Carried rings ------------- A frame walk on one worker visits frames in order through one ``FrameCarry``. ``RingUpdater`` redoes only the sources near a changed bond. The result equals ``ringNetwork`` on that frame. ``--jobs`` above 1, a single frame, and ``--rebuild`` use an instance with carry off, which recomputes every call. The hop-ball test moves one atom across a periodic face and requires the carried rings to match a full search, with fewer sources recomputed than the atom count: .. code:: cpp // excerpt: tests/test_ring_hop.cpp const auto &carried = updater.update(moved); const auto full = primitive::ringNetwork(moved, 6); REQUIRE(carried == full); What was timed -------------- Both tables are 4096 molecules, 200 transitions, eight threads. Every transition stayed inside the hop ball, and the carried rings matched a full search. Speedup is rebuild divided by carry. The mW window (10 fs step) at 0.1, 1, and 10 ps is in `Benchmarks `_. The strict seeded graph there is the row ``seeded-strict``. The TIP4P/2005 ice/water slab (2 fs step) is finer: 0.01, 0.1, and 1 ps. An oxygen moves farther in one picosecond than an mW site. The ``seeded-strict`` speedups are 8.31, 3.28, and 2.59. The logs are ``benchmarks/hop-stride-tip4p-5.txt``, ``benchmarks/hop-stride-tip4p-50.txt``, and ``benchmarks/hop-stride-tip4p-500.txt``. .. code:: python # excerpt: tests/cli_surface.py ["--frame", "1", "--last", "2", "--jobs", "1", "--rebuild", "read", cubic], That argv is the walk with carry forced off. Drop ``--rebuild`` to keep the updater.