hkaiser changed the topic of #ste||ar to: STE||AR: Systems Technology, Emergent Parallelism, and Algorithm Research | stellar-group.org | HPX: A cure for performance impaired parallel applications | github.com/STEllAR-GROUP/hpx | This channel is logged: irclog.cct.lsu.edu
hkaiser has joined #ste||ar
hkaiser has quit [Quit: Bye!]
Yorlik_ has joined #ste||ar
Yorlik__ has quit [Ping timeout: 264 seconds]
K-ballo1 has joined #ste||ar
K-ballo has quit [Ping timeout: 260 seconds]
K-ballo1 is now known as K-ballo
hkaiser has joined #ste||ar
K-ballo1 has joined #ste||ar
K-ballo has quit [Ping timeout: 252 seconds]
K-ballo1 is now known as K-ballo
hkaiser has quit [Quit: Bye!]
weilewei has joined #ste||ar
weilewei has quit [Quit: Ping timeout (120 seconds)]
weilewei has joined #ste||ar
hkaiser has joined #ste||ar
<weilewei> continuing our conversation on Telegram, std par with tbb enabled seems the fastest across all, is it something expected as well?
weilewei has quit [Quit: Ping timeout (120 seconds)]
weilewei has joined #ste||ar
weilewei has quit [Quit: Ping timeout (120 seconds)]
weilewei has joined #ste||ar
<weilewei> how to setup par_simd? any special compilation flags needed?
<weilewei> aha, let me ask Srinivas directly
<hkaiser> weilewei: I'd expect hpx to be on par with the std implementation
<weilewei> hkaiser that's what I would expect as well, let me try master branch
<weilewei> Perlmutter is degraded and on maintenance today lol, could not do much
<weilewei>     cmake \
<weilewei>         -DCMAKE_BUILD_TYPE=$BUILD_TYPE \
<weilewei>      -DBOOST_ROOT=$BOOST_PATH\
<weilewei>         -DHPX_WITH_CXX_STANDARD=20 \
<weilewei>         -DHPX_WITH_MALLOC=JEMALLOC \
<weilewei>         -DJEMALLOC_ROOT=${JEMALLOC_ROOT} \
<weilewei>         -DHPX_WITH_CUDA=OFF \
<weilewei>         -DHPX_WITH_FETCH_ASIO=On \
<weilewei>         -DHPX_WITH_APEX=OFF \
<weilewei>      -DHPX_WITH_NETWORKING=OFF \
<weilewei>      -DHPX_WITH_PARCELPORT_MPI=OFF \
<weilewei>      -DHPX_WITH_PARCELPORT_TCP=OFF\
<weilewei>      -DDHPX_WITH_GENERIC_CONTEXT_COROUTINES=ON \
<weilewei>         -DHPX_WITH_MAX_CPU_COUNT=256 \
<weilewei>      -DHPX_WITH_MALLOC=system\
<weilewei>         -DCMAKE_INSTALL_PREFIX=$INSTALL_PATH \
<weilewei>         -DHPX_WITH_EXAMPLES=OFF \
<weilewei>      -DHPX_WITH_TOOLS=OFF\
weilewei has quit [Quit: Ping timeout (120 seconds)]
weilewei has joined #ste||ar
weilewei has quit [Quit: Ping timeout (120 seconds)]
weilewei has joined #ste||ar
<hkaiser> weilewei: why HPX_WITH_GENERIC_CONTEXT_COROUTINES=On? also, you have a type there
<hkaiser> typo
weilewei has quit [Quit: Ping timeout (120 seconds)]
weilewei has joined #ste||ar
<weilewei> ok let me remove it
<hkaiser> weilewei: is it an x64 architecture?
<hkaiser> if yes, then you don't need that
<weilewei> it is  AMD EPYC 7763
<hkaiser> for benchmarking, you might want to add HPX_WITH_LOGGING=OFF
<weilewei> got it
weilewei has quit [Quit: Ping timeout (120 seconds)]
weilewei has joined #ste||ar
weilewei has quit [Quit: Ping timeout (120 seconds)]
weilewei has joined #ste||ar
<weilewei> with hpx master and all other configurations, hpx one improves a bit, for par execution, std 0.22, hpx 0.32, kokkos 0.32
tufei has quit [Remote host closed the connection]
tufei has joined #ste||ar
weilewei has quit [Quit: Ping timeout (120 seconds)]
weilewei has joined #ste||ar
K-ballo has quit [Ping timeout: 252 seconds]
K-ballo1 has joined #ste||ar
K-ballo1 is now known as K-ballo
weilewei has quit [Quit: Ping timeout (120 seconds)]
tufei has quit [Remote host closed the connection]
tufei has joined #ste||ar
diehlpk_work has joined #ste||ar
<gonidelis[m]> hkaiser: yt?
<hkaiser> gonidelis[m]: here
<gonidelis[m]> pm please