| | | | | | | requested parallelism | walltime sum (s) | nb instances | any sync average per thread time (s) | any wait average per thread time (s) | parallelism overhead (%) | local speedup if perfectly balanced | global speedup if perfectly balanced |
|---|
| start addr | function name | source location | level | ancestor thread num | invoker | parallel or teams | 1x52 | 1x52 | 1x52 | 1x52 | 1x52 | 1x52 | 1x52 | 1x52 |
|---|
| libgromacs_mpi.so.7:0x51fe1d | nonbonded_verlet_t::dispatchNonbondedKernel(gmx::Interaction... | kerneldispatch.cpp:265 | 0 | 0 | runtime | parallel | 52 | 19.948 | 10.0 E3 | 1.138 | 1.137 | 5.71 | 1.061 | 1.021 |
| libgromacs_mpi.so.7:0xa5f05f | gmx_pme_do(gmx_pme_t*, gmx::ArrayRef<gmx::BasicVector<float>... | pme.cpp:1296 | 0 | 0 | runtime | parallel | 52 | 10.122 | 10.0 E3 | 1.787 | 1.775 | 17.7 | 1.214 | 1.034 |
| libgromacs_mpi.so.7:0xa74dd0 | spread_on_grid(gmx_pme_t const*, PmeAtomComm*, pmegrids_t co... | pme_spread.cpp:936 | 0 | 0 | runtime | parallel | 52 | 3.340 | 10.0 E3 | 0.947 | 0.946 | 28.4 | 1.396 | 1.018 |
| libgromacs_mpi.so.7:0x991d76 | gmx::constrain_lincs(bool, t_inputrec const&, long, gmx::Lin... | lincs.cpp:2483 | 0 | 0 | runtime | parallel | 52 | 2.442 | 10.0 E3 | 0.988 | 0.940 | 40.5 | 1.679 | 1.019 |
| libgromacs_mpi.so.7:0xa5f12f | gmx_pme_do(gmx_pme_t*, gmx::ArrayRef<gmx::BasicVector<float>... | pme.cpp:1401 | 0 | 0 | runtime | parallel | 52 | 2.440 | 10.0 E3 | 0.801 | 0.800 | 32.8 | 1.489 | 1.015 |
| libgromacs_mpi.so.7:0xa74e40 | spread_on_grid(gmx_pme_t const*, PmeAtomComm*, pmegrids_t co... | pme_spread.cpp:1010 | 0 | 0 | runtime | parallel | 52 | 2.280 | 10.0 E3 | 0.608 | 0.606 | 26.6 | 1.363 | 1.011 |
| libgromacs_mpi.so.7:0xa4be47 | ListedForces::calculate(gmx_wallcycle*, float const (*) [3],... | listed_forces.cpp:385 | 0 | 0 | runtime | parallel | 52 | 1.608 | 10.0 E3 | 0.292 | 0.291 | 18.1 | 1.222 | 1.005 |
| libgromacs_mpi.so.7:0x512018 | reduceForces(nbnxn_atomdata_t*, gmx::AtomLocality, Nbnxm::Gr... | atomdata.cpp:1188 | 0 | 0 | runtime | parallel | 52 | 1.172 | 10.0 E3 | 0.291 | 0.289 | 24.8 | 1.330 | 1.005 |
| libgromacs_mpi.so.7:0xa74d3b | spread_on_grid(gmx_pme_t const*, PmeAtomComm*, pmegrids_t co... | pme_spread.cpp:910 | 0 | 0 | runtime | parallel | 52 | 1.038 | 10.0 E3 | 0.166 | 0.165 | 16.0 | 1.191 | 1.003 |
| libgromacs_mpi.so.7:0x511ee6 | nbnxn_atomdata_copy_x_to_nbat_x(Nbnxm::GridSet const&, gmx::... | atomdata.cpp:1023 | 0 | 0 | runtime | parallel | 52 | 0.724 | 9.88 E3 | 0.339 | 0.338 | 46.9 | 1.882 | 1.006 |
| libgromacs_mpi.so.7:0x967b1d | gmx::Constraints::Impl::apply(bool, bool, long, int, float, ... | constr.cpp:587 | 0 | 0 | runtime | parallel | 52 | 0.642 | 10.0 E3 | 0.178 | 0.177 | 27.8 | 1.384 | 1.003 |
| libgromacs_mpi.so.7:0x59627a | PairlistSet::constructPairlists(gmx::InteractionLocality, Nb... | pairlist.cpp:3958 | 0 | 0 | runtime | parallel | 52 | 0.462 | 126 | 44.0 E-3 | 44.0 E-3 | 9.52 | 1.105 | 1.001 |
| libgromacs_mpi.so.7:0x61c4d2 | gmx::ThreadedForceBuffer<float [4]>::reduce(gmx::ForceWithSh... | threaded_force_buffer.cpp:170 | 0 | 0 | runtime | parallel | 52 | 0.456 | 10.0 E3 | 0.208 | 0.207 | 45.6 | 1.838 | 1.004 |
| libgromacs_mpi.so.7:0x9bc4ee | gmx::Update::Impl::finish_update(t_inputrec const&, bool, in... | update.cpp:1637 | 0 | 0 | runtime | parallel | 52 | 0.444 | 10.0 E3 | 0.100 | 99.1 E-3 | 22.6 | 1.292 | 1.002 |
| libgromacs_mpi.so.7:0x9bc28a | gmx::Update::Impl::update_coords(t_inputrec const&, long, in... | update.cpp:1687 | 0 | 0 | runtime | parallel | 52 | 0.344 | 10.0 E3 | 89.1 E-3 | 87.7 E-3 | 25.9 | 1.350 | 1.002 |
| libgromacs_mpi.so.7:0x5a59a9 | nonbonded_verlet_t::dispatchPruneKernelCpu(gmx::InteractionL... | prunekerneldispatch.cpp:68 | 0 | 0 | runtime | parallel | 52 | 0.297 | 876 | 25.9 E-3 | 25.8 E-3 | 8.72 | 1.096 | 1.000 |
| libgromacs_mpi.so.7:0x5ceb5a | dd_make_local_constraints(gmx_domdec_t*, int, gmx_mtop_t con... | domdec_constraints.cpp:458 | 0 | 0 | runtime | parallel | 52 | 0.282 | 127 | 0.265 | 0.265 | 94.2 | 17.303 | 1.005 |
| libgromacs_mpi.so.7:0xa64f8b | unwrap_periodic_pmegrid(gmx_pme_t*, float*) | pme_grid.cpp:543 | 0 | 0 | runtime | parallel | 52 | 0.137 | 10.0 E3 | 71.8 E-3 | 70.5 E-3 | 52.4 | 2.099 | 1.001 |
| libgromacs_mpi.so.7:0x512078 | reduceForces(nbnxn_atomdata_t*, gmx::AtomLocality, Nbnxm::Gr... | atomdata.cpp:1272 | 0 | 0 | runtime | parallel | 52 | 0.122 | 10.0 E3 | 31.0 E-3 | 29.8 E-3 | 25.5 | 1.342 | 1.001 |
| libgromacs_mpi.so.7:0x9b6cae | setupForceOutputs(ForceHelperBuffers*, gmx::ArrayRefWithPadd... | sim_util.cpp:473 | 0 | 0 | runtime | parallel | 52 | 0.119 | 10.0 E3 | 36.7 E-3 | 35.5 E-3 | 30.9 | 1.448 | 1.001 |
| libgromacs_mpi.so.7:0xa64f2d | unwrap_periodic_pmegrid(gmx_pme_t*, float*) | pme_grid.cpp:512 | 0 | 0 | runtime | parallel | 52 | 95.0 E-3 | 10.0 E3 | 32.6 E-3 | 31.2 E-3 | 34.3 | 1.522 | 1.001 |
| libgromacs_mpi.so.7:0x5d9631 | dd_make_local_top(gmx_domdec_t const&, gmx_domdec_zones_t co... | localtopology.cpp:845 | 0 | 0 | runtime | parallel | 52 | 85.6 E-3 | 127 | 29.1 E-3 | 29.1 E-3 | 34.0 | 1.515 | 1.001 |
| libgromacs_mpi.so.7:0xa4dc11 | setup_bonded_threading(bonded_threading_t*, int, bool, Inter... | manage_threading.cpp:396 | 0 | 0 | runtime | parallel | 52 | 71.8 E-3 | 127 | 6.09 E-3 | 6.06 E-3 | 8.49 | 1.093 | 1.000 |
| libgromacs_mpi.so.7:0x51cae2 | Nbnxm::Grid::setCellIndices(int, int, Nbnxm::GridSetData*, g... | grid.cpp:1481 | 0 | 0 | runtime | parallel | 52 | 37.9 E-3 | 127 | 16.3 E-3 | 16.3 E-3 | 43.1 | 1.758 | 1.000 |
| libgromacs_mpi.so.7:0x996e28 | compute_globals(gmx_global_stat*, t_commrec*, t_inputrec con... | md_support.cpp:113 | 0 | 0 | runtime | parallel | 52 | 36.2 E-3 | 2.00 E3 | 5.79 E-3 | 5.54 E-3 | 16.0 | 1.190 | 1.000 |
| libgromacs_mpi.so.7:0xa7d10b | fft5d_plan_3d(int, int, int, ompi_communicator_t**, int, t_c... | fft5d.cpp:643 | 0 | 0 | runtime | parallel | 52 | 36.1 E-3 | 2.00 | 728 E-6 | 727 E-6 | 2.01 | 1.021 | 1.000 |
| libgromacs_mpi.so.7:0x596cc6 | PairlistSet::constructPairlists(gmx::InteractionLocality, Nb... | pairlist.cpp:3681 | 0 | 0 | runtime | parallel | 52 | 31.8 E-3 | 126 | 9.59 E-3 | 9.57 E-3 | 30.1 | 1.431 | 1.000 |
| libgromacs_mpi.so.7:0xa7d1d5 | fft5d_plan_3d(int, int, int, ompi_communicator_t**, int, t_c... | fft5d.cpp:643 | 0 | 0 | runtime | parallel | 52 | 27.0 E-3 | 2.00 | 578 E-6 | 578 E-6 | 2.14 | 1.022 | 1.000 |
| libgromacs_mpi.so.7:0x9b7485 | postProcessForces(t_commrec const*, long, t_nrnb*, gmx_wallc... | vec.h:538 | 0 | 0 | runtime | parallel | 52 | 20.9 E-3 | 1.00 E3 | 5.27 E-3 | 5.14 E-3 | 25.2 | 1.337 | 1.000 |
| libgromacs_mpi.so.7:0x9b6eaa | setupForceOutputs(ForceHelperBuffers*, gmx::ArrayRefWithPadd... | sim_util.cpp:473 | 0 | 0 | runtime | parallel | 52 | 19.0 E-3 | 1.00 E3 | 6.05 E-3 | 5.92 E-3 | 31.8 | 1.467 | 1.000 |
| libgromacs_mpi.so.7:0x999627 | atoms2md(gmx_mtop_t const&, t_inputrec const&, int, gmx::Arr... | mdatoms.cpp:503 | 0 | 0 | runtime | parallel | 52 | 17.7 E-3 | 127 | 4.58 E-3 | 4.56 E-3 | 25.9 | 1.350 | 1.000 |
| libgromacs_mpi.so.7:0x966719 | calc_vir(int, float const (*) [3], float const (*) [3], floa... | calcvir.cpp:109 | 0 | 0 | runtime | parallel | 52 | 16.8 E-3 | 1.00 E3 | 4.41 E-3 | 4.28 E-3 | 26.3 | 1.357 | 1.000 |
| libgromacs_mpi.so.7:0x990bb2 | gmx::lincs_thread_setup(gmx::Lincs*, int) | lincs.cpp:1636 | 0 | 0 | runtime | parallel | 52 | 13.8 E-3 | 127 | 1.59 E-3 | 1.56 E-3 | 11.5 | 1.130 | 1.000 |
| libgromacs_mpi.so.7:0x9902ce | gmx::set_lincs(InteractionDefinitions const&, int, gmx::Arra... | lincs.cpp:2196 | 0 | 0 | runtime | parallel | 52 | 13.5 E-3 | 127 | 2.25 E-3 | 2.23 E-3 | 16.7 | 1.200 | 1.000 |
| libgromacs_mpi.so.7:0x991376 | gmx::set_lincs_matrix(gmx::Lincs*, gmx::ArrayRef<float const... | lincs.cpp:1347 | 0 | 0 | runtime | parallel | 52 | 7.04 E-3 | 127 | 2.47 E-3 | 2.38 E-3 | 35.0 | 1.539 | 1.000 |
| libgromacs_mpi.so.7:0x51e511 | Nbnxm::GridSet::putOnGrid(float const (*) [3], int, float co... | gridset.cpp:212 | 0 | 0 | runtime | parallel | 52 | 5.36 E-3 | 127 | 2.04 E-3 | 2.02 E-3 | 38.0 | 1.613 | 1.000 |
| libgromacs_mpi.so.7:0xac90e6 | gmx_set_thread_affinity(gmx::MDLogger const&, t_commrec cons... | threadaffinity.cpp:314 | 0 | 0 | runtime | parallel | 52 | 4.85 E-3 | 1.00 | 435 E-6 | 137 E-6 | 8.98 | 1.099 | 1.000 |
| libgromacs_mpi.so.7:0x5ed01d | dd_redistribute_cg(_IO_FILE*, long, gmx_domdec_t*, int*, t_s... | redistribute.cpp:624 | 0 | 0 | runtime | parallel | 52 | 4.20 E-3 | 126 | 605 E-6 | 590 E-6 | 14.4 | 1.168 | 1.000 |
| libgromacs_mpi.so.7:0x990e97 | gmx::lincs_thread_setup(gmx::Lincs*, int) | lincs.cpp:1695 | 0 | 0 | runtime | parallel | 52 | 3.97 E-3 | 127 | 1.04 E-3 | 1.02 E-3 | 26.2 | 1.354 | 1.000 |
| libgromacs_mpi.so.7:0x9cbfa4 | calc_vcm_grp(t_mdatoms const&, gmx::ArrayRef<gmx::BasicVecto... | vcm.cpp:222 | 0 | 0 | runtime | parallel | 52 | 2.47 E-3 | 102 | 726 E-6 | 667 E-6 | 29.4 | 1.417 | 1.000 |
| libgromacs_mpi.so.7:0x9ccbb3 | process_and_stopcm_grp(_IO_FILE*, t_vcm*, t_mdatoms const&, ... | vcm.cpp:374 | 0 | 0 | runtime | parallel | 52 | 1.87 E-3 | 102 | 884 E-6 | 831 E-6 | 47.3 | 1.896 | 1.000 |
| libgromacs_mpi.so.7:0xa7d29f | fft5d_plan_3d(int, int, int, ompi_communicator_t**, int, t_c... | fft5d.cpp:643 | 0 | 0 | runtime | parallel | 52 | 756 E-6 | 2.00 | 370 E-6 | 370 E-6 | 49.0 | 1.960 | 1.000 |
| libgromacs_mpi.so.7:0x61bbaa | gmx::ThreadedForceBuffer<float [4]>::ThreadedForceBuffer(int... | threaded_force_buffer.cpp:243 | 0 | 0 | runtime | parallel | 52 | 190 E-6 | 1.00 | 97.1 E-6 | 97.0 E-6 | 51.1 | 2.043 | 1.000 |
| libgromacs_mpi.so.7:0xa6eaae | pme_init_all_work(pme_solve_work_t**, int, int) | pme_solve.cpp:153 | 0 | 0 | runtime | parallel | 52 | 39.9 E-6 | 1.00 | 19.1 E-6 | 19.0 E-6 | 47.9 | 1.920 | 1.000 |
| libgromacs_mpi.so.7:0xa61d1d | PmeAtomComm& std::vector<PmeAtomComm, std::allocator<PmeAtom... | new_allocator.h:136 | 0 | 0 | runtime | parallel | 52 | 9.18 E-6 | 1.00 | 3.25 E-6 | 3.10 E-6 | 35.4 | 1.547 | 1.000 |
| libgromacs_mpi.so.7:0x606b2a | gmx_ekindata_t::gmx_ekindata_t(int, float, int) | group.cpp:66 | 0 | 0 | runtime | parallel | 52 | 7.70 E-6 | 1.00 | 2.84 E-6 | 2.69 E-6 | 36.9 | 1.585 | 1.000 |
