# V. Weber and U. Borstnik                                #
#                                                         #
# The purpose of this benchmark is to analyse the         #
# time spent in the following linear algebra routines     #
# as displayed at the end of the cp2k output.             #
#                                                         #
# main routines
make_preconditioner
cp_dbcsr_multiply_d
apply_preconditioner
#                                                         #
# secondary routines                                      #
# (separate timings for different multiplication options) #
dbcsr_mult_NSS_NRN
dbcsr_mult_NRN_TRN
dbcsr_mult_NSN_NRN
dbcsr_mult_TRN_NRN
dbcsr_mult_NRN_NSN
#                                                         #
# On rosa the total run times are approximately:          #
#                                                         #
# H2O-256.inp    5 min on  576 cores                      #
# H2O-1024.inp  40 min on  576 cores                      #
# H2O-4096.inp  15 min on 2304 cores                      #
# H2O-65536.inp doesnt run so far...                      #
#                                                         #
# Runs performed with cp2k from 30.01.09 (no libdbcsr     #
# routines used), compiled with gfortran and linked to    #
# the default libraries.                                  #
