Building NAMD version 2.7b1 from source. See this benchmarks page for results.
First attempt was with UDP, icc, plus smp. Charm++ was built using
./build charm++ net-linux-x86_64 icc smp ifort -O -DCMK_OPTIMIZE
NAMD was build with
./config Linux-x86_64-icc --charm-arch net-linux-x86_64-ifort-smp-icc
after removing the static cxa stuff from the corresponding 'arch' file.
Note that a warning was issued by charm++ : Charm warning> Randomization of stack pointer is turned on in Kernel, run 'echo 0 > /proc/sys/kernel/randomize_va_space' as root to disable it. Thread migration may not work!
Second attempt was with TCP, icc plus smp. Charm++ didn't pass the megatest:
./build charm++ net-linux-x86_64 icc ifort smp tcp -O2 -DCMK_OPTIMIZE ... ... ./charmrun ++local +p4 ./pgm ------------- Processor 2 Exiting: Called CmiAbort ------------ Reason: Charm++ Error> +isomalloc_sync requires CmiBarrier() implemented.
Third attempt was the MPI version, again with icc & smp. Charm++ was built using:
./build charm++ mpi-linux-x86_64 smp mpicxx -O2 -DCMK_OPTIMIZE
NAMD was built with
./config Linux-x86_64-icc --charm-arch mpi-linux-x86_64-smp-mpicxx