Qbox build on the BNL CFN nano cluster

Qbox installation issues
Forum rules
You must be a registered user to post in this forum. Registered users may also post new topics if they consider that their subject do not correspond to any topic already present on the forum.
Post Reply
fgygi
Site Admin
Posts: 131
Joined: Tue Jun 17, 2008 7:03 pm

Qbox build on the BNL CFN nano cluster

Post by fgygi » Sun Feb 17, 2013 2:17 am

Qbox was built on the nano cluster at the Center for Functional Nanomaterials (CFN) of Brookhaven National Laboratory.

The modules used are:
# 1) /intel/11.1.069 3) /impi/4.1.0.024
# 2) /intel_mkl/10.2.4.032

The following Makefile include was used to build Qbox:

Code: Select all

#-------------------------------------------------------------------------------
#
#  nano.bnl.gov
#
#  required modules: intel intel_mkl impi
#  1) /intel/11.1.069         3) /impi/4.1.0.024
#  2) /intel_mkl/10.2.4.032
#-------------------------------------------------------------------------------
#
 PLT=x86_64
#-------------------------------------------------------------------------------
 FFTWDIR=$(HOME)/software/fftw-2.1.5/fftw
 XERCESCDIR=$(HOME)/software/xerces/xerces-c-src_2_8_0
 PLTOBJECTS = readTSC.o

 CXX=icc
 LD=$(CXX)
 LD=icpc

 PLTFLAGS += -DIA32 -DUSE_FFTW -D_LARGEFILE_SOURCE \
             -D_FILE_OFFSET_BITS=64 -DUSE_MPI -DSCALAPACK -DADD_ \
             -DAPP_NO_THREADS -DXML_USE_NO_THREADS -DUSE_XERCES
 PLTFLAGS += -DMPICH_IGNORE_CXX_SEEK

 INCLUDE = -I$(FFTWDIR) -I$(XERCESCDIR)/include \
           -I$(MKLROOT)/include -I$(I_MPI_ROOT)/include64

 CXXFLAGS= -O3 -vec-report1 -D$(PLT) $(INCLUDE) $(PLTFLAGS) $(DFLAGS)

 LIBPATH = -L$(FFTWDIR)/.libs \
           -L$(XERCESCDIR)/lib \
           -L$(I_MPI_ROOT)/lib64

 LIBS =  $(PLIBS) \
         -lmkl_intel_lp64 \
         -lmkl_lapack95_lp64 -lmkl_sequential -lmkl_core \
         -lirc -lifcore -lsvml \
         -lmpi \
         -lfftw -luuid $(XERCESCDIR)/lib/libxerces-c.a -lpthread

# Parallel libraries
 PLIBS = -lmkl_scalapack_lp64 -lmkl_blacs_intelmpi_lp64

 LDFLAGS = $(LIBPATH) $(LIBS)

#-------------------------------------------------------------------------------
This build is typical of an Intel environment in which the MKL ScaLAPACK/BLACS libraries are used.
It uses the Intel impi MPI library rather than the mvapich2 library.

Post Reply