Skip to content

Commit e058f67

Browse files
committed
fixed graph script
1 parent 6311c6b commit e058f67

File tree

5 files changed

+488
-421
lines changed

5 files changed

+488
-421
lines changed
Lines changed: 30 additions & 30 deletions
Original file line numberDiff line numberDiff line change
@@ -1,31 +1,31 @@
11
m,n,k,lda,ldb,ldc,offa,offb,offc,alpha,beta,order,transa,transb,side,uplo,diag,function,device,library,label,GFLOPS
2-
192,192,192,0,0,0,0,0,0,1.0,1.0,column,none,none,right,upper,unit,dtrsm,gpu,clblas,s9150_dtrsm_14502,8.9202
3-
384,384,384,0,0,0,0,0,0,1.0,1.0,column,none,none,right,upper,unit,dtrsm,gpu,clblas,s9150_dtrsm_14502,46.185
4-
576,576,576,0,0,0,0,0,0,1.0,1.0,column,none,none,right,upper,unit,dtrsm,gpu,clblas,s9150_dtrsm_14502,126.686
5-
768,768,768,0,0,0,0,0,0,1.0,1.0,column,none,none,right,upper,unit,dtrsm,gpu,clblas,s9150_dtrsm_14502,235.366
6-
960,960,960,0,0,0,0,0,0,1.0,1.0,column,none,none,right,upper,unit,dtrsm,gpu,clblas,s9150_dtrsm_14502,375.406
7-
1152,1152,1152,0,0,0,0,0,0,1.0,1.0,column,none,none,right,upper,unit,dtrsm,gpu,clblas,s9150_dtrsm_14502,475.497
8-
1344,1344,1344,0,0,0,0,0,0,1.0,1.0,column,none,none,right,upper,unit,dtrsm,gpu,clblas,s9150_dtrsm_14502,599.527
9-
1536,1536,1536,0,0,0,0,0,0,1.0,1.0,column,none,none,right,upper,unit,dtrsm,gpu,clblas,s9150_dtrsm_14502,437.835
10-
1728,1728,1728,0,0,0,0,0,0,1.0,1.0,column,none,none,right,upper,unit,dtrsm,gpu,clblas,s9150_dtrsm_14502,778.815
11-
1920,1920,1920,0,0,0,0,0,0,1.0,1.0,column,none,none,right,upper,unit,dtrsm,gpu,clblas,s9150_dtrsm_14502,845.844
12-
2112,2112,2112,0,0,0,0,0,0,1.0,1.0,column,none,none,right,upper,unit,dtrsm,gpu,clblas,s9150_dtrsm_14502,969.624
13-
2304,2304,2304,0,0,0,0,0,0,1.0,1.0,column,none,none,right,upper,unit,dtrsm,gpu,clblas,s9150_dtrsm_14502,943.48
14-
2496,2496,2496,0,0,0,0,0,0,1.0,1.0,column,none,none,right,upper,unit,dtrsm,gpu,clblas,s9150_dtrsm_14502,1026.58
15-
2688,2688,2688,0,0,0,0,0,0,1.0,1.0,column,none,none,right,upper,unit,dtrsm,gpu,clblas,s9150_dtrsm_14502,1074.56
16-
2880,2880,2880,0,0,0,0,0,0,1.0,1.0,column,none,none,right,upper,unit,dtrsm,gpu,clblas,s9150_dtrsm_14502,1102.6
17-
3072,3072,3072,0,0,0,0,0,0,1.0,1.0,column,none,none,right,upper,unit,dtrsm,gpu,clblas,s9150_dtrsm_14502,848.076
18-
3264,3264,3264,0,0,0,0,0,0,1.0,1.0,column,none,none,right,upper,unit,dtrsm,gpu,clblas,s9150_dtrsm_14502,1010.06
19-
3456,3456,3456,0,0,0,0,0,0,1.0,1.0,column,none,none,right,upper,unit,dtrsm,gpu,clblas,s9150_dtrsm_14502,1034.51
20-
3648,3648,3648,0,0,0,0,0,0,1.0,1.0,column,none,none,right,upper,unit,dtrsm,gpu,clblas,s9150_dtrsm_14502,1059.02
21-
3840,3840,3840,0,0,0,0,0,0,1.0,1.0,column,none,none,right,upper,unit,dtrsm,gpu,clblas,s9150_dtrsm_14502,1037.95
22-
4032,4032,4032,0,0,0,0,0,0,1.0,1.0,column,none,none,right,upper,unit,dtrsm,gpu,clblas,s9150_dtrsm_14502,1103.8
23-
4224,4224,4224,0,0,0,0,0,0,1.0,1.0,column,none,none,right,upper,unit,dtrsm,gpu,clblas,s9150_dtrsm_14502,1109.83
24-
4416,4416,4416,0,0,0,0,0,0,1.0,1.0,column,none,none,right,upper,unit,dtrsm,gpu,clblas,s9150_dtrsm_14502,1096.15
25-
4608,4608,4608,0,0,0,0,0,0,1.0,1.0,column,none,none,right,upper,unit,dtrsm,gpu,clblas,s9150_dtrsm_14502,1055.28
26-
4800,4800,4800,0,0,0,0,0,0,1.0,1.0,column,none,none,right,upper,unit,dtrsm,gpu,clblas,s9150_dtrsm_14502,1140.07
27-
4992,4992,4992,0,0,0,0,0,0,1.0,1.0,column,none,none,right,upper,unit,dtrsm,gpu,clblas,s9150_dtrsm_14502,1152.31
28-
5184,5184,5184,0,0,0,0,0,0,1.0,1.0,column,none,none,right,upper,unit,dtrsm,gpu,clblas,s9150_dtrsm_14502,1165.47
29-
5376,5376,5376,0,0,0,0,0,0,1.0,1.0,column,none,none,right,upper,unit,dtrsm,gpu,clblas,s9150_dtrsm_14502,1152.36
30-
5568,5568,5568,0,0,0,0,0,0,1.0,1.0,column,none,none,right,upper,unit,dtrsm,gpu,clblas,s9150_dtrsm_14502,1193.66
31-
5760,5760,5760,0,0,0,0,0,0,1.0,1.0,column,none,none,right,upper,unit,dtrsm,gpu,clblas,s9150_dtrsm_14502,1199.05
2+
192,192,192,0,0,0,0,0,0,1.0,1.0,column,none,none,right,upper,unit,dtrsm,gpu,clblas,S9150,8.9202
3+
384,384,384,0,0,0,0,0,0,1.0,1.0,column,none,none,right,upper,unit,dtrsm,gpu,clblas,S9150,46.185
4+
576,576,576,0,0,0,0,0,0,1.0,1.0,column,none,none,right,upper,unit,dtrsm,gpu,clblas,S9150,126.686
5+
768,768,768,0,0,0,0,0,0,1.0,1.0,column,none,none,right,upper,unit,dtrsm,gpu,clblas,S9150,235.366
6+
960,960,960,0,0,0,0,0,0,1.0,1.0,column,none,none,right,upper,unit,dtrsm,gpu,clblas,S9150,375.406
7+
1152,1152,1152,0,0,0,0,0,0,1.0,1.0,column,none,none,right,upper,unit,dtrsm,gpu,clblas,S9150,475.497
8+
1344,1344,1344,0,0,0,0,0,0,1.0,1.0,column,none,none,right,upper,unit,dtrsm,gpu,clblas,S9150,599.527
9+
1536,1536,1536,0,0,0,0,0,0,1.0,1.0,column,none,none,right,upper,unit,dtrsm,gpu,clblas,S9150,437.835
10+
1728,1728,1728,0,0,0,0,0,0,1.0,1.0,column,none,none,right,upper,unit,dtrsm,gpu,clblas,S9150,778.815
11+
1920,1920,1920,0,0,0,0,0,0,1.0,1.0,column,none,none,right,upper,unit,dtrsm,gpu,clblas,S9150,845.844
12+
2112,2112,2112,0,0,0,0,0,0,1.0,1.0,column,none,none,right,upper,unit,dtrsm,gpu,clblas,S9150,969.624
13+
2304,2304,2304,0,0,0,0,0,0,1.0,1.0,column,none,none,right,upper,unit,dtrsm,gpu,clblas,S9150,943.48
14+
2496,2496,2496,0,0,0,0,0,0,1.0,1.0,column,none,none,right,upper,unit,dtrsm,gpu,clblas,S9150,1026.58
15+
2688,2688,2688,0,0,0,0,0,0,1.0,1.0,column,none,none,right,upper,unit,dtrsm,gpu,clblas,S9150,1074.56
16+
2880,2880,2880,0,0,0,0,0,0,1.0,1.0,column,none,none,right,upper,unit,dtrsm,gpu,clblas,S9150,1102.6
17+
3072,3072,3072,0,0,0,0,0,0,1.0,1.0,column,none,none,right,upper,unit,dtrsm,gpu,clblas,S9150,848.076
18+
3264,3264,3264,0,0,0,0,0,0,1.0,1.0,column,none,none,right,upper,unit,dtrsm,gpu,clblas,S9150,1010.06
19+
3456,3456,3456,0,0,0,0,0,0,1.0,1.0,column,none,none,right,upper,unit,dtrsm,gpu,clblas,S9150,1034.51
20+
3648,3648,3648,0,0,0,0,0,0,1.0,1.0,column,none,none,right,upper,unit,dtrsm,gpu,clblas,S9150,1059.02
21+
3840,3840,3840,0,0,0,0,0,0,1.0,1.0,column,none,none,right,upper,unit,dtrsm,gpu,clblas,S9150,1037.95
22+
4032,4032,4032,0,0,0,0,0,0,1.0,1.0,column,none,none,right,upper,unit,dtrsm,gpu,clblas,S9150,1103.8
23+
4224,4224,4224,0,0,0,0,0,0,1.0,1.0,column,none,none,right,upper,unit,dtrsm,gpu,clblas,S9150,1109.83
24+
4416,4416,4416,0,0,0,0,0,0,1.0,1.0,column,none,none,right,upper,unit,dtrsm,gpu,clblas,S9150,1096.15
25+
4608,4608,4608,0,0,0,0,0,0,1.0,1.0,column,none,none,right,upper,unit,dtrsm,gpu,clblas,S9150,1055.28
26+
4800,4800,4800,0,0,0,0,0,0,1.0,1.0,column,none,none,right,upper,unit,dtrsm,gpu,clblas,S9150,1140.07
27+
4992,4992,4992,0,0,0,0,0,0,1.0,1.0,column,none,none,right,upper,unit,dtrsm,gpu,clblas,S9150,1152.31
28+
5184,5184,5184,0,0,0,0,0,0,1.0,1.0,column,none,none,right,upper,unit,dtrsm,gpu,clblas,S9150,1165.47
29+
5376,5376,5376,0,0,0,0,0,0,1.0,1.0,column,none,none,right,upper,unit,dtrsm,gpu,clblas,S9150,1152.36
30+
5568,5568,5568,0,0,0,0,0,0,1.0,1.0,column,none,none,right,upper,unit,dtrsm,gpu,clblas,S9150,1193.66
31+
5760,5760,5760,0,0,0,0,0,0,1.0,1.0,column,none,none,right,upper,unit,dtrsm,gpu,clblas,S9150,1199.05

doc/performance/clBLAS_2.6.0/S9150/generate_graphs.sh

100644100755
Lines changed: 68 additions & 1 deletion
Original file line numberDiff line numberDiff line change
@@ -3,14 +3,25 @@
33
python ../../../../src/scripts/perf/plotPerformance.py \
44
-d peak_sp.csv \
55
-d ../../cuBLAS_7.0/Tesla_K40/peak_sp.csv \
6-
-d sgemm_96.csv \
6+
-d sgemm_32.csv \
77
-d ../../cuBLAS_7.0/Tesla_K40/sgemm.csv \
88
-x sizem --x_axis_label "m,n,k" \
99
-y gflops --y_axis_label "GFlop/s" \
1010
--x_axis_scale linear \
1111
--plot label \
1212
--title "sgemm S9150 vs K40" --outputfile sgemm_S9150_K40.png
1313

14+
# sgemm AMD only
15+
python ../../../../src/scripts/perf/plotPerformance.py \
16+
-d peak_sp.csv \
17+
-d sgemm_32.csv \
18+
-x sizem --x_axis_label "m,n,k" \
19+
-y gflops --y_axis_label "GFlop/s" \
20+
--x_axis_scale linear \
21+
--plot label \
22+
--title "sgemm S9150" --outputfile sgemm_S9150.png
23+
24+
1425
# dgemm AMD vs NVIDIA
1526
python ../../../../src/scripts/perf/plotPerformance.py \
1627
-d peak_dp.csv \
@@ -23,3 +34,59 @@ python ../../../../src/scripts/perf/plotPerformance.py \
2334
--plot label \
2435
--title "dgemm S9150 vs K40" --outputfile dgemm_S9150_K40.png
2536

37+
# dgemm AMD only
38+
python ../../../../src/scripts/perf/plotPerformance.py \
39+
-d peak_dp.csv \
40+
-d dgemm_96.csv \
41+
-x sizem --x_axis_label "m,n,k" \
42+
-y gflops --y_axis_label "GFlop/s" \
43+
--x_axis_scale linear \
44+
--plot label \
45+
--title "dgemm S9150" --outputfile dgemm_S9150.png
46+
47+
48+
# zgemm AMD vs NVIDIA
49+
python ../../../../src/scripts/perf/plotPerformance.py \
50+
-d peak_dp.csv \
51+
-d ../../cuBLAS_7.0/Tesla_K40/peak_dp.csv \
52+
-d zgemm_64.csv \
53+
-d ../../cuBLAS_7.0/Tesla_K40/zgemm.csv \
54+
-x sizem --x_axis_label "m,n,k" \
55+
-y gflops --y_axis_label "GFlop/s" \
56+
--x_axis_scale linear \
57+
--plot label \
58+
--title "zgemm S9150 vs K40" --outputfile zgemm_S9150_K40.png
59+
60+
# zgemm AMD only
61+
python ../../../../src/scripts/perf/plotPerformance.py \
62+
-d peak_dp.csv \
63+
-d zgemm_64.csv \
64+
-x sizem --x_axis_label "m,n,k" \
65+
-y gflops --y_axis_label "GFlop/s" \
66+
--x_axis_scale linear \
67+
--plot label \
68+
--title "zgemm S9150" --outputfile zgemm_S9150.png
69+
70+
71+
# dtrsm AMD vs NVIDIA
72+
python ../../../../src/scripts/perf/plotPerformance.py \
73+
-d peak_dp.csv \
74+
-d ../../cuBLAS_7.0/Tesla_K40/peak_dp.csv \
75+
-d dtrsm_192.csv \
76+
-d ../../cuBLAS_7.0/Tesla_K40/dtrsm.csv \
77+
-x sizem --x_axis_label "m,n,k" \
78+
-y gflops --y_axis_label "GFlop/s" \
79+
--x_axis_scale linear \
80+
--plot label \
81+
--title "dtrsm S9150 vs K40" --outputfile dtrsm_S9150_K40.png
82+
83+
# dtrsm AMD only
84+
python ../../../../src/scripts/perf/plotPerformance.py \
85+
-d peak_dp.csv \
86+
-d dtrsm_192.csv \
87+
-x sizem --x_axis_label "m,n,k" \
88+
-y gflops --y_axis_label "GFlop/s" \
89+
--x_axis_scale linear \
90+
--plot label \
91+
--title "dtrsm S9150" --outputfile dtrsm_S9150.png
92+

0 commit comments

Comments
 (0)