| | | | | | | 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 | 1x128 | 1x128 | 1x128 | 1x128 | 1x128 | 1x128 | 1x128 | 1x128 |
|---|
| libgromacs.so.8:0x4dff60 | nonbonded_verlet_t::dispatchNonbondedKernel(gmx::Interaction... | kerneldispatch.cpp:266 | 0 | 0 | runtime | parallel | 128 | 16.434 | 10.0 E3 | 3.585 | 3.584 | 21.8 | 1.279 | 1.058 |
| libgromacs.so.8:0xa59ff2 | gmx_pme_do(gmx_pme_t*, gmx::ArrayRef<gmx::BasicVector<float>... | pme.cpp:1292 | 0 | 0 | runtime | parallel | 128 | 16.217 | 10.0 E3 | 4.846 | 4.829 | 29.9 | 1.426 | 1.081 |
| libgromacs.so.8:0x4d1069 | reduceForces(nbnxn_atomdata_t*, gmx::AtomLocality, Nbnxm::Gr... | atomdata.cpp:1157 | 0 | 0 | runtime | parallel | 128 | 3.784 | 10.0 E3 | 1.415 | 1.413 | 37.4 | 1.597 | 1.022 |
| libgromacs.so.8:0x98842b | gmx::constrain_lincs(bool, t_inputrec const&, long, gmx::Lin... | lincs.cpp:2534 | 0 | 0 | runtime | parallel | 128 | 3.168 | 10.0 E3 | 1.550 | 1.482 | 48.9 | 1.957 | 1.024 |
| libgromacs.so.8:0xa70bd9 | spread_on_grid(gmx_pme_t const*, PmeAtomComm*, pmegrids_t co... | pme_spread.cpp:936 | 0 | 0 | runtime | parallel | 128 | 2.449 | 10.0 E3 | 0.728 | 0.727 | 29.7 | 1.423 | 1.011 |
| libgromacs.so.8:0xa5a0cc | gmx_pme_do(gmx_pme_t*, gmx::ArrayRef<gmx::BasicVector<float>... | pme.cpp:1397 | 0 | 0 | runtime | parallel | 128 | 1.724 | 10.0 E3 | 0.526 | 0.525 | 30.5 | 1.439 | 1.008 |
| libgromacs.so.8:0xa70c4a | spread_on_grid(gmx_pme_t const*, PmeAtomComm*, pmegrids_t co... | pme_spread.cpp:1010 | 0 | 0 | runtime | parallel | 128 | 1.584 | 10.0 E3 | 0.535 | 0.533 | 33.8 | 1.510 | 1.008 |
| libgromacs.so.8:0x4d0f33 | nbnxn_atomdata_copy_x_to_nbat_x(Nbnxm::GridSet const&, gmx::... | atomdata.cpp:964 | 0 | 0 | runtime | parallel | 128 | 1.456 | 9.80 E3 | 0.766 | 0.764 | 52.6 | 2.111 | 1.012 |
| libgromacs.so.8:0xa46b14 | ListedForces::calculate(gmx_wallcycle*, float const (*) [3],... | listed_forces.cpp:387 | 0 | 0 | runtime | parallel | 128 | 1.117 | 10.0 E3 | 0.251 | 0.250 | 22.5 | 1.290 | 1.004 |
| libgromacs.so.8:0xa70b4d | spread_on_grid(gmx_pme_t const*, PmeAtomComm*, pmegrids_t co... | pme_spread.cpp:910 | 0 | 0 | runtime | parallel | 128 | 1.043 | 10.0 E3 | 0.373 | 0.371 | 35.7 | 1.556 | 1.006 |
| libgromacs.so.8:0x95c9ab | gmx::Constraints::Impl::apply(bool, bool, long, int, float, ... | constr.cpp:590 | 0 | 0 | runtime | parallel | 128 | 0.923 | 10.0 E3 | 0.312 | 0.310 | 33.8 | 1.510 | 1.005 |
| libgromacs.so.8:0x9b5143 | gmx::Update::Impl::finish_update(t_inputrec const&, bool, in... | update.cpp:1660 | 0 | 0 | runtime | parallel | 128 | 0.588 | 10.0 E3 | 0.141 | 0.139 | 24.0 | 1.316 | 1.002 |
| libgromacs.so.8:0x9b4e9e | gmx::Update::Impl::update_coords(t_inputrec const&, long, in... | update.cpp:1710 | 0 | 0 | runtime | parallel | 128 | 0.510 | 10.0 E3 | 0.149 | 0.147 | 29.2 | 1.412 | 1.002 |
| libgromacs.so.8:0x591fea | dd_make_local_constraints(gmx_domdec_t*, int, gmx_mtop_t con... | domdec_constraints.cpp:430 | 0 | 0 | runtime | parallel | 128 | 0.467 | 202 | 0.456 | 0.456 | 97.5 | 39.904 | 1.007 |
| libgromacs.so.8:0x557407 | PairlistSet::constructPairlists(gmx::InteractionLocality, Nb... | pairlist.cpp:4059 | 0 | 0 | runtime | parallel | 128 | 0.389 | 201 | 58.7 E-3 | 58.7 E-3 | 15.1 | 1.178 | 1.001 |
| libgromacs.so.8:0x5e329c | gmx::ThreadedForceBuffer<float [4]>::reduce(gmx::ForceWithSh... | threaded_force_buffer.cpp:170 | 0 | 0 | runtime | parallel | 128 | 0.389 | 10.0 E3 | 0.162 | 0.160 | 41.6 | 1.713 | 1.002 |
| libgromacs.so.8:0xa60403 | unwrap_periodic_pmegrid(gmx_pme_t*, float*) | pme_grid.cpp:529 | 0 | 0 | runtime | parallel | 128 | 0.329 | 10.0 E3 | 0.174 | 0.172 | 52.8 | 2.118 | 1.003 |
| libgromacs.so.8:0x566b38 | nonbonded_verlet_t::dispatchPruneKernelCpu(gmx::InteractionL... | prunekerneldispatch.cpp:69 | 0 | 0 | runtime | parallel | 128 | 0.241 | 1.00 E3 | 65.9 E-3 | 65.8 E-3 | 27.3 | 1.376 | 1.001 |
| libgromacs.so.8:0xa603a3 | unwrap_periodic_pmegrid(gmx_pme_t*, float*) | pme_grid.cpp:512 | 0 | 0 | runtime | parallel | 128 | 0.192 | 10.0 E3 | 93.9 E-3 | 91.9 E-3 | 49.0 | 1.959 | 1.001 |
| libgromacs.so.8:0x9aed5e | setupForceOutputs(ForceHelperBuffers*, gmx::ArrayRefWithPadd... | sim_util.cpp:473 | 0 | 0 | runtime | parallel | 128 | 0.128 | 10.0 E3 | 44.5 E-3 | 43.2 E-3 | 34.8 | 1.534 | 1.001 |
| libgromacs.so.8:0x4d10cd | reduceForces(nbnxn_atomdata_t*, gmx::AtomLocality, Nbnxm::Gr... | atomdata.cpp:1231 | 0 | 0 | runtime | parallel | 128 | 93.9 E-3 | 10.0 E3 | 34.2 E-3 | 32.8 E-3 | 36.4 | 1.573 | 1.001 |
| libgromacs.so.8:0xa48b95 | setup_bonded_threading(bonded_threading_t*, int, bool, Inter... | manage_threading.cpp:396 | 0 | 0 | runtime | parallel | 128 | 72.0 E-3 | 202 | 17.4 E-3 | 17.4 E-3 | 24.2 | 1.319 | 1.000 |
| libgromacs.so.8:0x59dbaf | make_local_bondeds_excls(gmx_domdec_t const&, gmx_domdec_zon... | localtopology.cpp:846 | 0 | 0 | runtime | parallel | 128 | 67.1 E-3 | 202 | 23.6 E-3 | 23.6 E-3 | 35.2 | 1.543 | 1.000 |
| libgromacs.so.8:0x4dc93b | Nbnxm::Grid::setCellIndices(int, int, Nbnxm::GridSetData*, g... | grid.cpp:1483 | 0 | 0 | runtime | parallel | 128 | 60.6 E-3 | 202 | 28.7 E-3 | 28.7 E-3 | 47.4 | 1.900 | 1.000 |
| libgromacs.so.8:0x557def | PairlistSet::constructPairlists(gmx::InteractionLocality, Nb... | pairlist.cpp:3782 | 0 | 0 | runtime | parallel | 128 | 59.5 E-3 | 201 | 38.4 E-3 | 38.4 E-3 | 64.6 | 2.827 | 1.001 |
| libgromacs.so.8:0x98cef2 | compute_globals(gmx_global_stat*, t_commrec*, t_inputrec con... | md_support.cpp:113 | 0 | 0 | runtime | parallel | 128 | 57.7 E-3 | 2.00 E3 | 22.1 E-3 | 21.8 E-3 | 38.4 | 1.623 | 1.000 |
| libgromacs.so.8:0xa792b5 | fft5d_plan_3d(int, int, int, tmpi_comm_**, int, t_complex**,... | fft5d.cpp:630 | 0 | 0 | runtime | parallel | 128 | 41.4 E-3 | 2.00 | 1.52 E-3 | 1.52 E-3 | 3.67 | 1.038 | 1.000 |
| libgromacs.so.8:0x9af61f | postProcessForces(t_commrec const*, long, t_nrnb*, gmx_wallc... | vec.h:538 | 0 | 0 | runtime | parallel | 128 | 21.1 E-3 | 1.00 E3 | 6.74 E-3 | 6.57 E-3 | 32.0 | 1.470 | 1.000 |
| libgromacs.so.8:0x98f7a5 | atoms2md(gmx_mtop_t const&, t_inputrec const&, int, gmx::Arr... | mdatoms.cpp:266 | 0 | 0 | runtime | parallel | 128 | 19.6 E-3 | 202 | 4.24 E-3 | 4.21 E-3 | 21.6 | 1.275 | 1.000 |
| libgromacs.so.8:0x986a61 | gmx::set_lincs(InteractionDefinitions const&, int, gmx::Arra... | lincs.cpp:2246 | 0 | 0 | runtime | parallel | 128 | 17.5 E-3 | 202 | 4.66 E-3 | 4.63 E-3 | 26.7 | 1.364 | 1.000 |
| libgromacs.so.8:0x9aef51 | setupForceOutputs(ForceHelperBuffers*, gmx::ArrayRefWithPadd... | sim_util.cpp:473 | 0 | 0 | runtime | parallel | 128 | 13.7 E-3 | 1.00 E3 | 5.37 E-3 | 5.23 E-3 | 39.1 | 1.641 | 1.000 |
| libgromacs.so.8:0x95b47d | calc_vir(int, float const (*) [3], float const (*) [3], floa... | calcvir.cpp:109 | 0 | 0 | runtime | parallel | 128 | 13.3 E-3 | 1.00 E3 | 5.23 E-3 | 5.10 E-3 | 39.2 | 1.645 | 1.000 |
| libgromacs.so.8:0x987bcd | gmx::set_lincs_matrix(gmx::Lincs*, gmx::ArrayRef<float const... | lincs.cpp:1397 | 0 | 0 | runtime | parallel | 128 | 12.7 E-3 | 202 | 5.20 E-3 | 4.98 E-3 | 40.9 | 1.693 | 1.000 |
| libgromacs.so.8:0x987338 | gmx::set_lincs(InteractionDefinitions const&, int, gmx::Arra... | lincs.cpp:1686 | 0 | 0 | runtime | parallel | 128 | 11.8 E-3 | 202 | 3.26 E-3 | 3.23 E-3 | 27.6 | 1.381 | 1.000 |
| libgromacs.so.8:0x4de56e | Nbnxm::GridSet::putOnGrid(float const (*) [3], int, float co... | gridset.cpp:212 | 0 | 0 | runtime | parallel | 128 | 10.1 E-3 | 202 | 3.54 E-3 | 3.51 E-3 | 35.0 | 1.539 | 1.000 |
| libgromacs.so.8:0x5e2919 | gmx::ThreadedForceBuffer<float [4]>::ThreadedForceBuffer(int... | threaded_force_buffer.cpp:243 | 0 | 0 | runtime | parallel | 128 | 8.31 E-3 | 1.00 | 98.4 E-6 | 92.8 E-6 | 1.18 | 1.012 | 1.000 |
| libgromacs.so.8:0xa791d9 | fft5d_plan_3d(int, int, int, tmpi_comm_**, int, t_complex**,... | fft5d.cpp:630 | 0 | 0 | runtime | parallel | 128 | 7.33 E-3 | 2.00 | 2.02 E-3 | 2.02 E-3 | 27.5 | 1.380 | 1.000 |
| libgromacs.so.8:0x5b2332 | dd_redistribute_cg(_IO_FILE*, long, gmx_domdec_t*, int*, t_s... | redistribute.cpp:624 | 0 | 0 | runtime | parallel | 128 | 6.83 E-3 | 201 | 1.87 E-3 | 1.85 E-3 | 27.5 | 1.378 | 1.000 |
| libgromacs.so.8:0x987625 | gmx::set_lincs(InteractionDefinitions const&, int, gmx::Arra... | lincs.cpp:1745 | 0 | 0 | runtime | parallel | 128 | 6.74 E-3 | 202 | 2.44 E-3 | 2.40 E-3 | 36.2 | 1.566 | 1.000 |
| libgromacs.so.8:0x9c39e4 | process_and_stopcm_grp(_IO_FILE*, t_vcm*, t_mdatoms const&, ... | vcm.cpp:372 | 0 | 0 | runtime | parallel | 128 | 2.34 E-3 | 102 | 1.22 E-3 | 1.11 E-3 | 52.2 | 2.094 | 1.000 |
| libgromacs.so.8:0xa79381 | fft5d_plan_3d(int, int, int, tmpi_comm_**, int, t_complex**,... | fft5d.cpp:630 | 0 | 0 | runtime | parallel | 128 | 2.10 E-3 | 2.00 | 1.04 E-3 | 1.04 E-3 | 49.5 | 1.980 | 1.000 |
| libgromacs.so.8:0x9c2dd9 | calc_vcm_grp(t_mdatoms const&, gmx::ArrayRef<gmx::BasicVecto... | vcm.cpp:222 | 0 | 0 | runtime | parallel | 128 | 2.02 E-3 | 102 | 985 E-6 | 863 E-6 | 48.7 | 1.949 | 1.000 |
| libgromacs.so.8:0xac9498 | gmx_set_thread_affinity(gmx::MDLogger const&, t_commrec cons... | threadaffinity.cpp:315 | 0 | 0 | runtime | parallel | 128 | 214 E-6 | 1.00 | 57.3 E-6 | 56.4 E-6 | 26.7 | 1.365 | 1.000 |
| libgromacs.so.8:0xa6a5dc | pme_init_all_work(pme_solve_work_t**, int, int) | pme_solve.cpp:152 | 0 | 0 | runtime | parallel | 128 | 65.0 E-6 | 1.00 | 33.3 E-6 | 33.2 E-6 | 51.2 | 2.051 | 1.000 |
| libgromacs.so.8:0x5cdf66 | gmx_ekindata_t::gmx_ekindata_t(gmx::ArrayRef<float const>, E... | group.cpp:76 | 0 | 0 | runtime | parallel | 128 | 13.1 E-6 | 1.00 | 4.65 E-6 | 4.49 E-6 | 35.4 | 1.548 | 1.000 |
| libgromacs.so.8:0xa5d17d | PmeAtomComm& std::vector<PmeAtomComm, std::allocator<PmeAtom... | new_allocator.h:162 | 0 | 0 | runtime | parallel | 128 | 12.0 E-6 | 1.00 | 4.38 E-6 | 4.21 E-6 | 36.4 | 1.572 | 1.000 |
