fdurastante
/
amg4psblas
mirror of https://github.com/sfilippone/amg4psblas.git
1
0
Fork 0
Code Issues Packages Projects Releases Wiki Activity
You cannot select more than 25 topics Topics must start with a letter or number, can include dashes ('-') and can be up to 35 characters long.
repackage
repack-newsolve
l1-and-0-aggr
development
l1aggregation
PolySmooth
Poly-novrl
maint-1.1
jac_solver
master
dev-openmp
TestFerdous
omp-walther
openmp-match
maint-1.1.0
fixmatch
tspmm
maint-1.0
maint-1.0.1
fix-sludist7
remap-coarse
savebcmatch
configry
mergeparmatch
documentation
new-context
implement-ainv
merge-amgext
unify_aggr_bld
unify_aggr_bld_richardson
pizdaint-runs
richardson
stopcriterion
V1.0.0
v1.0-rc2
v1.0.1
v1.1.0
v1.1.1
v1.1.2
Branches Tags
${ item.name }
Create tag ${ searchTerm }
Create branch ${ searchTerm }
from '9c85e54740'
${ noResults }
amg4psblas/test/fileread/runs/drt.sh

85 lines
4.9 KiB
Bash
Raw Normal View History Unescape Escape

Added test directory, populated with benchmark code.
17 years ago
#!/bin/sh
#
#
# NUmber of attempts for each configuration
ntry=1
Merged minor typos from Daniela.
17 years ago
export GFORTRAN_UNBUFFERED_ALL=y
Added test directory, populated with benchmark code.
17 years ago
date=`date +%Y%m%d%H%M%S`
for np in $*
do
# 3rd batch: 14bis,64bis 4 sweeps
echo "mpirun -np $np -machinefile locm df_bench >>log.part$part.ren$renum.${np}p"
#/usr/local/mpich-gcc42/bin/mpirun -np $np -machinefile locm df_bench <<EOF
mld2p4: mlprec/mld_caggrmap_bld.f90 mlprec/mld_cmlprec_bld.f90 mlprec/mld_cprecinit.f90 mlprec/mld_cprecset.f90 mlprec/mld_daggrmap_bld.f90 mlprec/mld_dmlprec_bld.f90 mlprec/mld_dprecinit.f90 mlprec/mld_dprecset.f90 mlprec/mld_inner_mod.f90 mlprec/mld_prec_type.f90 mlprec/mld_saggrmap_bld.f90 mlprec/mld_smlprec_bld.f90 mlprec/mld_sprecinit.f90 mlprec/mld_sprecset.f90 mlprec/mld_zaggrmap_bld.f90 mlprec/mld_zmlprec_bld.f90 mlprec/mld_zprecinit.f90 mlprec/mld_zprecset.f90 test/fileread/cf_sample.f90 test/fileread/df_bench.f90 test/fileread/df_sample.f90 test/fileread/runs/cfs.inp test/fileread/runs/dfs.inp test/fileread/runs/drt.sh test/fileread/runs/sfs.inp test/fileread/runs/zfs.inp test/fileread/sf_sample.f90 test/fileread/zf_sample.f90 Added an aggregation threshold to account for anisotropies.
17 years ago
mpirun -np $np -machinefile locm df_bench >>run.gfc.kiva.log.${np}p.$date 2>err.gfc.kiva.log.${np}p.$date <<EOF
Added test directory, populated with benchmark code.
17 years ago
out.${np}p.$date Out file 1: summary
stat.${np}p.$date Out file 2: detailed for statistics
BICGSTAB iterative method to use
1.D-6 EPS
CSR Matrix format
0 IPART: Partition method: 0: BLOCK 1: BLK2 2:GRAPH
00250 ITMAX
-1 ITRACE
2 ISTOPC 1: NBE Infty 2: |r|2/|b|2
30 IRST Restart parameter for GMRES and BiCGSTAB(L)
0 RENUM: 0: none 1: global indices (2: GPS band reduction)
$ntry NTRY for each comb. print out best timings
Merged ilu(k) branch. Branch still open to play with ILU(P,T).
17 years ago
30 NPRCS nov rst prl fc1 fl1 mlt agg smt cm smp ft2 fl2 jsw nl omg th1 th2 name
none none 0 0 0 0 0 0 0 0 0 0 0 0 0 1 -1.0 1e-4 1e-4 NOPREC
diag none 0 0 0 1 0 2 0 1 0 2 1 0 4 1 -1.0 1e-4 1e-4 DIAG
Reorganized AS build and application routines.
17 years ago
bjac none 0 0 0 1 0 2 0 1 0 2 1 0 4 1 -1.0 1e-4 1e-4 BJAC
as none 0 1 0 1 0 2 0 1 0 2 1 0 4 1 -1.0 1e-4 1e-4 RAS
as none 1 1 0 1 0 2 0 1 0 2 1 0 4 1 -1.0 1e-4 1e-4 RAS
as none 2 1 0 1 0 2 0 1 0 2 1 0 4 1 -1.0 1e-4 1e-4 RAS
Merged ilu(k) branch. Branch still open to play with ILU(P,T).
17 years ago
as ml 0 1 0 1 0 2 0 1 0 2 1 0 4 2 -1.0 1e-4 1e-4 2L-M-RAS-I-D4
as ml 1 1 0 1 0 2 0 1 0 2 1 0 4 2 -1.0 1e-4 1e-4 2L-M-RAS-I-D4
as ml 2 1 0 1 0 2 0 1 0 2 1 0 4 2 -1.0 1e-4 1e-4 2L-M-RAS-I-D4
as ml 0 1 0 1 0 2 0 1 0 2 5 0 4 2 -1.0 1e-4 1e-4 2L-M-RAS-U-D4
as ml 1 1 0 1 0 2 0 1 0 2 5 0 4 2 -1.0 1e-4 1e-4 2L-M-RAS-U-D4
as ml 2 1 0 1 0 2 0 1 0 2 5 0 4 2 -1.0 1e-4 1e-4 2L-M-RAS-U-D4
as ml 0 1 0 1 0 2 0 1 0 2 1 0 4 3 -1.0 1e-4 1e-4 3L-M-RAS-I-D4
as ml 1 1 0 1 0 2 0 1 0 2 1 0 4 3 -1.0 1e-4 1e-4 3L-M-RAS-I-D4
as ml 2 1 0 1 0 2 0 1 0 2 1 0 4 3 -1.0 1e-4 1e-4 3L-M-RAS-I-D4
as ml 0 1 0 1 0 2 0 1 0 2 5 0 4 3 -1.0 1e-4 1e-4 3L-M-RAS-U-D4
as ml 1 1 0 1 0 2 0 1 0 2 5 0 4 3 -1.0 1e-4 1e-4 3L-M-RAS-U-D4
as ml 2 1 0 1 0 2 0 1 0 2 5 0 4 3 -1.0 1e-4 1e-4 3L-M-RAS-U-D4
as ml 0 1 0 1 0 2 0 1 1 2 1 0 1 2 -1.0 1e-4 1e-4 2L-M-RAS-I-R
as ml 1 1 0 1 0 2 0 1 1 2 1 0 1 2 -1.0 1e-4 1e-4 2L-M-RAS-I-R
as ml 2 1 0 1 0 2 0 1 1 2 1 0 1 2 -1.0 1e-4 1e-4 2L-M-RAS-I-R
as ml 0 1 0 1 0 2 0 1 1 2 5 0 1 2 -1.0 1e-4 1e-4 2L-M-RAS-U-R
as ml 1 1 0 1 0 2 0 1 1 2 5 0 1 2 -1.0 1e-4 1e-4 2L-M-RAS-U-R
as ml 2 1 0 1 0 2 0 1 1 2 5 0 1 2 -1.0 1e-4 1e-4 2L-M-RAS-U-R
as ml 0 1 0 1 0 2 0 1 1 2 1 0 1 3 -1.0 1e-4 1e-4 3L-M-RAS-I-R
as ml 1 1 0 1 0 2 0 1 1 2 1 0 1 3 -1.0 1e-4 1e-4 3L-M-RAS-I-R
as ml 2 1 0 1 0 2 0 1 1 2 1 0 1 3 -1.0 1e-4 1e-4 3L-M-RAS-I-R
as ml 0 1 0 1 0 2 0 1 1 2 5 0 1 3 -1.0 1e-4 1e-4 3L-M-RAS-U-R
as ml 1 1 0 1 0 2 0 1 1 2 5 0 1 3 -1.0 1e-4 1e-4 3L-M-RAS-U-R
as ml 2 1 0 1 0 2 0 1 1 2 5 0 1 3 -1.0 1e-4 1e-4 3L-M-RAS-U-R
Added test directory, populated with benchmark code.
17 years ago
2 Number of matrices
Fixed mld_ name glitch.
17 years ago
kivap004.mtx none
kivap001.mtx none
Added test directory, populated with benchmark code.
17 years ago
thm50x30.mtx none
thm200x120.mtx none
thm1000x600.mtx none
a400x400.mtx b400.mtx
kivap007.mtx none
!!! preconditioner templates
bja none 0 0 0 0 0 0 0 0 0 0 0 0.0 Block Jacobi
none none 0 0 0 0 0 0 0 0 0 0 0 0.1 No preconditioner
diagsc none 0 0 0 0 0 0 0 0 0 0 0 0.0 Diagonal scaling
as none 1 4 1 0 0 0 0 0 0 0 0 0.0 Additive Schwarz 1 overlap
as none 1 4 0 0 0 0 0 0 0 0 0 0.0 Restricted Additive Schwarz 1 overlap
EOF
cat out.${np}p.$date >>dat.out.kiva.$date
cat stat.${np}p.$date >>dat.stat.kiva.$date
done