Skip to content

GitLab

Commit 1b16108d authored by one's avatar one
Browse files

Update rocHPL inputs for BW150

parent 6d1f7610
Hide whitespace changes
Inline Side-by-side
Showing with 138 additions and 12 deletions
projects/rocHPL/bw1000/run-all.sh
View file @ 1b16108d
#!/bin/bash
export HSA_FORCE_FINE_GRAIN_PCIE=1
BIN_PATH=./build/mpirun_rochpl
# rochpl-bw1000-16r.patch is required
# $BIN_PATH -P 8 -Q 2 --it 6 -i HPL-16.dat
# rochpl-bw.patch is required
$BIN_PATH -P 4 -Q 2 --it 6 -i HPL-8.dat # N=512*498
$BIN_PATH -P 4 -Q 1 --it 6 -i HPL-4.dat
$BIN_PATH -P 2 -Q 1 --it 6 -i HPL-2.dat
./build/mpirun_rochpl -P 8 -Q 2 --it 6 -i HPL-16.dat
./build/mpirun_rochpl -P 4 -Q 2 --it 6 -i HPL-8.dat # N=512*498
./build/mpirun_rochpl -P 4 -Q 1 --it 6 -i HPL-4.dat
./build/mpirun_rochpl -P 2 -Q 1 --it 6 -i HPL-2.dat
./build/mpirun_rochpl -P 1 -Q 1 --it 6 -i HPL-1.dat
projects/rocHPL/bw150/HPL-1.dat 0 → 100644
View file @ 1b16108d
HPLinpack benchmark input file
Innovative Computing Laboratory, University of Tennessee
HPL.out output file name (if any)
0 device out (6=stdout,7=stderr,file)
1 # of problems sizes (N)
90624 Ns
1 # of NBs
512 NBs
1 PMAP process mapping (0=Row-,1=Column-major)
1 # of process grids (P x Q)
1 Ps
1 Qs
16.0 threshold
1 # of panel fact
2 PFACTs (0=left, 1=Crout, 2=Right)
1 # of recursive stopping criterium
16 NBMINs (>= 1)
1 # of panels in recursion
2 NDIVs
1 # of recursive panel fact.
2 RFACTs (0=left, 1=Crout, 2=Right)
1 # of broadcast
0 BCASTs (0=1rg,1=1rM,2=2rg,3=2rM,4=Lng,5=LnM)
1 # of lookahead depth
1 DEPTHs (>=0)
1 SWAP (0=bin-exch,1=long,2=mix)
64 swapping threshold
0 L1 in (0=transposed,1=no-transposed) form
0 U in (0=transposed,1=no-transposed) form
0 Equilibration (0=no,1=yes)
8 memory alignment in double (> 0)
projects/rocHPL/bw150/HPL-2.dat 0 → 100644
View file @ 1b16108d
HPLinpack benchmark input file
Innovative Computing Laboratory, University of Tennessee
HPL.out output file name (if any)
0 device out (6=stdout,7=stderr,file)
1 # of problems sizes (N)
128000 Ns
1 # of NBs
512 NBs
1 PMAP process mapping (0=Row-,1=Column-major)
1 # of process grids (P x Q)
2 Ps
1 Qs
16.0 threshold
1 # of panel fact
2 PFACTs (0=left, 1=Crout, 2=Right)
1 # of recursive stopping criterium
32 NBMINs (>= 1)
1 # of panels in recursion
2 NDIVs
1 # of recursive panel fact.
2 RFACTs (0=left, 1=Crout, 2=Right)
1 # of broadcast
3 BCASTs (0=1rg,1=1rM,2=2rg,3=2rM,4=Lng,5=LnM)
1 # of lookahead depth
1 DEPTHs (>=0)
1 SWAP (0=bin-exch,1=long,2=mix)
64 swapping threshold
0 L1 in (0=transposed,1=no-transposed) form
0 U in (0=transposed,1=no-transposed) form
0 Equilibration (0=no,1=yes)
8 memory alignment in double (> 0)
projects/rocHPL/bw150/HPL-4.dat 0 → 100644
View file @ 1b16108d
HPLinpack benchmark input file
Innovative Computing Laboratory, University of Tennessee
HPL.out output file name (if any)
0 device out (6=stdout,7=stderr,file)
1 # of problems sizes (N)
180224 Ns
1 # of NBs
512 NBs
1 PMAP process mapping (0=Row-,1=Column-major)
1 # of process grids (P x Q)
4 Ps
1 Qs
16.0 threshold
1 # of panel fact
2 PFACTs (0=left, 1=Crout, 2=Right)
1 # of recursive stopping criterium
32 NBMINs (>= 1)
1 # of panels in recursion
2 NDIVs
1 # of recursive panel fact.
2 RFACTs (0=left, 1=Crout, 2=Right)
1 # of broadcast
3 BCASTs (0=1rg,1=1rM,2=2rg,3=2rM,4=Lng,5=LnM)
1 # of lookahead depth
1 DEPTHs (>=0)
1 SWAP (0=bin-exch,1=long,2=mix)
64 swapping threshold
0 L1 in (0=transposed,1=no-transposed) form
0 U in (0=transposed,1=no-transposed) form
0 Equilibration (0=no,1=yes)
8 memory alignment in double (> 0)
projects/rocHPL/bw150/HPL-8.dat 0 → 100644
View file @ 1b16108d
HPLinpack benchmark input file
Innovative Computing Laboratory, University of Tennessee
HPL.out output file name (if any)
0 device out (6=stdout,7=stderr,file)
1 # of problems sizes (N)
256000 Ns
1 # of NBs
512 NBs
1 PMAP process mapping (0=Row-,1=Column-major)
1 # of process grids (P x Q)
4 Ps
2 Qs
16.0 threshold
1 # of panel fact
2 PFACTs (0=left, 1=Crout, 2=Right)
1 # of recursive stopping criterium
32 NBMINs (>= 1)
1 # of panels in recursion
2 NDIVs
1 # of recursive panel fact.
2 RFACTs (0=left, 1=Crout, 2=Right)
1 # of broadcast
5 BCASTs (0=1rg,1=1rM,2=2rg,3=2rM,4=Lng,5=LnM)
1 # of lookahead depth
1 DEPTHs (>=0)
1 SWAP (0=bin-exch,1=long,2=mix)
64 swapping threshold
0 L1 in (0=transposed,1=no-transposed) form
0 U in (0=transposed,1=no-transposed) form
0 Equilibration (0=no,1=yes)
8 memory alignment in double (> 0)
projects/rocHPL/bw150/run-all.sh 0 → 100644
View file @ 1b16108d
#!/bin/bash
./build/mpirun_rochpl -P 4 -Q 2 --it 6 -i HPL-8.dat
./build/mpirun_rochpl -P 4 -Q 1 --it 6 -i HPL-4.dat
./build/mpirun_rochpl -P 2 -Q 1 --it 6 -i HPL-2.dat
./build/mpirun_rochpl -P 1 -Q 1 --it 6 -i HPL-1.dat
projects/rocHPL/rochpl-scripts-bw.patch
View file @ 1b16108d
diff --git a/scripts/mpirun_rochpl.in b/scripts/mpirun_rochpl.in
index 155f502..2344410 100755
index 155f502..932218b 100755
--- a/scripts/mpirun_rochpl.in
+++ b/scripts/mpirun_rochpl.in
@@ -46,7 +46,7 @@ supported_distro( )
......@@ -56,7 +56,7 @@ index 155f502..2344410 100755
exit 0
;;
-P)
@@ -219,4 +211,11 @@ if [ ! -z "${devices}" ]; then
@@ -219,4 +211,12 @@ if [ ! -z "${devices}" ]; then
fi
#run
......@@ -67,6 +67,7 @@ index 155f502..2344410 100755
+ --mca btl ^vader,tcp,openib,uct \
+ -x UCX_TLS=self,sm,rocm,rc \
+ -x UCX_MEMTYPE_CACHE=n \
+ -x HSA_FORCE_FINE_GRAIN_PCIE=1 \
+ -np ${np} \
+ ${rochpl_runscript} ${rochpl_args}
diff --git a/scripts/run_rochpl.in b/scripts/run_rochpl.in
......
Markdown is supported
0% or .
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment