Software Archive
Read-only legacy content
17061 Discussions

Oversubscription of OpenMP threads for processing small data sets

SergeyKostrov
Valued Contributor II
9,247 Views
*** Oversubscription of OpenMP threads for processing small data sets ***
0 Kudos
87 Replies
SergeyKostrov
Valued Contributor II
1,488 Views
[ Test Case 16 - 32-bit / Cores: 1 / CPUs: 1 - Number of OpenMP threads - 1 ] [ Matrix Size: 1024 x 1024 ] Application - IccTestApp - WIN32_ICC ( 32-bit ) - Release Tests: Start > Test1099 Start < Matrix A, B and C Sizes : 1024 x 1024 Loop Processing Schema ( LPS ): IKJ Loop Blocking Divider : 1 Sub-Test 1.1 - MxMultA1 - Classic 2D LBOT size: N/A Completed: 0.67200 secs Sub-Test 1.2 - MxMultA2 - Classic 2D LBOT LBOT size: 1024x1024 elements Completed: 0.65600 secs Sub-Test 1.3 - MxMultA3 - Classic 2D Fused LBOT size: N/A Completed: 2.79700 secs Sub-Test 1.4 - MxMultA4 - Classic 2D Fused LBOT LBOT size: 1024x1024 elements Completed: 2.82800 secs Sub-Test 2.1 - MxMultB1 - Classic 2D Transposed LBOT size: N/A Completed: 1.51600 secs Sub-Test 2.2 - MxMultB2 - Classic 2D Transposed LBOT LBOT size: 1024x1024 elements Completed: 1.51600 secs Sub-Test 2.3 - MxMultB3 - Classic 2D Fused Transposed LBOT size: N/A Completed: 3.45400 secs Sub-Test 2.4 - MxMultB4 - Classic 2D Fused Transposed LBOT LBOT size: 1024x1024 elements Completed: 3.46900 secs Sub-Test 5.1 - MxMultD1 - Classic 1D LBOT size: N/A Completed: 4.61000 secs Sub-Test 5.2 - MxMultD2 - Classic 1D LBOT LBOT size: 1024x1024 elements Completed: 2.54700 secs > Test1099 End < Tests: Completed
0 Kudos
SergeyKostrov
Valued Contributor II
1,488 Views
[ Test Case 17 - 32-bit / Cores: 1 / CPUs: 1 - Number of OpenMP threads - 2 ] [ Matrix Size: 1024 x 1024 ] Application - IccTestApp - WIN32_ICC ( 32-bit ) - Release Tests: Start > Test1099 Start < Matrix A, B and C Sizes : 1024 x 1024 Loop Processing Schema ( LPS ): IKJ Loop Blocking Divider : 1 Sub-Test 1.1 - MxMultA1 - Classic 2D LBOT size: N/A Completed: 0.70400 secs Sub-Test 1.2 - MxMultA2 - Classic 2D LBOT LBOT size: 1024x1024 elements Completed: 0.70300 secs Sub-Test 1.3 - MxMultA3 - Classic 2D Fused LBOT size: N/A Completed: 2.76500 secs Sub-Test 1.4 - MxMultA4 - Classic 2D Fused LBOT LBOT size: 1024x1024 elements Completed: 2.82800 secs Sub-Test 2.1 - MxMultB1 - Classic 2D Transposed LBOT size: N/A Completed: 1.50000 secs Sub-Test 2.2 - MxMultB2 - Classic 2D Transposed LBOT LBOT size: 1024x1024 elements Completed: 1.51600 secs Sub-Test 2.3 - MxMultB3 - Classic 2D Fused Transposed LBOT size: N/A Completed: 3.45400 secs Sub-Test 2.4 - MxMultB4 - Classic 2D Fused Transposed LBOT LBOT size: 1024x1024 elements Completed: 3.45300 secs Sub-Test 5.1 - MxMultD1 - Classic 1D LBOT size: N/A Completed: 4.61000 secs Sub-Test 5.2 - MxMultD2 - Classic 1D LBOT LBOT size: 1024x1024 elements Completed: 2.53100 secs > Test1099 End < Tests: Completed
0 Kudos
SergeyKostrov
Valued Contributor II
1,488 Views
[ Test Case 18 - 32-bit / Cores: 1 / CPUs: 1 - Number of OpenMP threads - 4 ] [ Matrix Size: 1024 x 1024 ] Application - IccTestApp - WIN32_ICC ( 32-bit ) - Release Tests: Start > Test1099 Start < Matrix A, B and C Sizes : 1024 x 1024 Loop Processing Schema ( LPS ): IKJ Loop Blocking Divider : 1 Sub-Test 1.1 - MxMultA1 - Classic 2D LBOT size: N/A Completed: 1.46900 secs Sub-Test 1.2 - MxMultA2 - Classic 2D LBOT LBOT size: 1024x1024 elements Completed: 1.43800 secs Sub-Test 1.3 - MxMultA3 - Classic 2D Fused LBOT size: N/A Completed: 2.78100 secs Sub-Test 1.4 - MxMultA4 - Classic 2D Fused LBOT LBOT size: 1024x1024 elements Completed: 2.82800 secs Sub-Test 2.1 - MxMultB1 - Classic 2D Transposed LBOT size: N/A Completed: 1.48400 secs Sub-Test 2.2 - MxMultB2 - Classic 2D Transposed LBOT LBOT size: 1024x1024 elements Completed: 1.50000 secs Sub-Test 2.3 - MxMultB3 - Classic 2D Fused Transposed LBOT size: N/A Completed: 3.46900 secs Sub-Test 2.4 - MxMultB4 - Classic 2D Fused Transposed LBOT LBOT size: 1024x1024 elements Completed: 3.45400 secs Sub-Test 5.1 - MxMultD1 - Classic 1D LBOT size: N/A Completed: 4.60900 secs Sub-Test 5.2 - MxMultD2 - Classic 1D LBOT LBOT size: 1024x1024 elements Completed: 2.53100 secs > Test1099 End < Tests: Completed
0 Kudos
SergeyKostrov
Valued Contributor II
1,488 Views
[ Test Case 19 - 32-bit / Cores: 1 / CPUs: 1 - Number of OpenMP threads - 8 ] [ Matrix Size: 1024 x 1024 ] Application - IccTestApp - WIN32_ICC ( 32-bit ) - Release Tests: Start > Test1099 Start < Matrix A, B and C Sizes : 1024 x 1024 Loop Processing Schema ( LPS ): IKJ Loop Blocking Divider : 1 Sub-Test 1.1 - MxMultA1 - Classic 2D LBOT size: N/A Completed: 1.48400 secs Sub-Test 1.2 - MxMultA2 - Classic 2D LBOT LBOT size: 1024x1024 elements Completed: 1.45300 secs Sub-Test 1.3 - MxMultA3 - Classic 2D Fused LBOT size: N/A Completed: 2.78100 secs Sub-Test 1.4 - MxMultA4 - Classic 2D Fused LBOT LBOT size: 1024x1024 elements Completed: 2.82900 secs Sub-Test 2.1 - MxMultB1 - Classic 2D Transposed LBOT size: N/A Completed: 1.50000 secs Sub-Test 2.2 - MxMultB2 - Classic 2D Transposed LBOT LBOT size: 1024x1024 elements Completed: 1.53100 secs Sub-Test 2.3 - MxMultB3 - Classic 2D Fused Transposed LBOT size: N/A Completed: 3.46900 secs Sub-Test 2.4 - MxMultB4 - Classic 2D Fused Transposed LBOT LBOT size: 1024x1024 elements Completed: 3.46900 secs Sub-Test 5.1 - MxMultD1 - Classic 1D LBOT size: N/A Completed: 4.62500 secs Sub-Test 5.2 - MxMultD2 - Classic 1D LBOT LBOT size: 1024x1024 elements Completed: 2.53100 secs > Test1099 End < Tests: Completed
0 Kudos
SergeyKostrov
Valued Contributor II
1,488 Views
[ Test Case 20 - 32-bit / Cores: 1 / CPUs: 1 - Number of OpenMP threads - 16 ] [ Matrix Size: 1024 x 1024 ] Application - IccTestApp - WIN32_ICC ( 32-bit ) - Release Tests: Start > Test1099 Start < Matrix A, B and C Sizes : 1024 x 1024 Loop Processing Schema ( LPS ): IKJ Loop Blocking Divider : 1 Sub-Test 1.1 - MxMultA1 - Classic 2D LBOT size: N/A Completed: 1.46800 secs Sub-Test 1.2 - MxMultA2 - Classic 2D LBOT LBOT size: 1024x1024 elements Completed: 1.42200 secs Sub-Test 1.3 - MxMultA3 - Classic 2D Fused LBOT size: N/A Completed: 2.76500 secs Sub-Test 1.4 - MxMultA4 - Classic 2D Fused LBOT LBOT size: 1024x1024 elements Completed: 2.82800 secs Sub-Test 2.1 - MxMultB1 - Classic 2D Transposed LBOT size: N/A Completed: 1.46900 secs Sub-Test 2.2 - MxMultB2 - Classic 2D Transposed LBOT LBOT size: 1024x1024 elements Completed: 1.48500 secs Sub-Test 2.3 - MxMultB3 - Classic 2D Fused Transposed LBOT size: N/A Completed: 3.45300 secs Sub-Test 2.4 - MxMultB4 - Classic 2D Fused Transposed LBOT LBOT size: 1024x1024 elements Completed: 3.45300 secs Sub-Test 5.1 - MxMultD1 - Classic 1D LBOT size: N/A Completed: 4.60900 secs Sub-Test 5.2 - MxMultD2 - Classic 1D LBOT LBOT size: 1024x1024 elements Completed: 2.53100 secs > Test1099 End < Tests: Completed
0 Kudos
SergeyKostrov
Valued Contributor II
1,488 Views
[ Test Case 21 - 64-bit / Cores: 1 / CPUs: 2 - Number of OpenMP threads - 1 ] [ Matrix Size: 256 x 256 ] Application - IccTestApp - WIN64_ICC ( 64-bit ) - Release Tests: Start > Test1099 Start < Matrix A, B and C Sizes : 256 x 256 Loop Processing Schema ( LPS ): IKJ Loop Blocking Divider : 1 Sub-Test 1.1 - MxMultA1 - Classic 2D LBOT size: N/A Completed: 0.00194 secs Sub-Test 1.2 - MxMultA2 - Classic 2D LBOT LBOT size: 256x256 elements Completed: 0.00194 secs Sub-Test 1.3 - MxMultA3 - Classic 2D Fused LBOT size: N/A Completed: 0.01075 secs Sub-Test 1.4 - MxMultA4 - Classic 2D Fused LBOT LBOT size: 256x256 elements Completed: 0.00975 secs Sub-Test 2.1 - MxMultB1 - Classic 2D Transposed LBOT size: N/A Completed: 0.00294 secs Sub-Test 2.2 - MxMultB2 - Classic 2D Transposed LBOT LBOT size: 256x256 elements Completed: 0.00387 secs Sub-Test 2.3 - MxMultB3 - Classic 2D Fused Transposed LBOT size: N/A Completed: 0.00587 secs Sub-Test 2.4 - MxMultB4 - Classic 2D Fused Transposed LBOT LBOT size: 256x256 elements Completed: 0.00581 secs Sub-Test 5.1 - MxMultD1 - Classic 1D LBOT size: N/A Completed: 0.03119 secs Sub-Test 5.2 - MxMultD2 - Classic 1D LBOT LBOT size: 256x256 elements Completed: 0.00587 secs > Test1099 End < Tests: Completed
0 Kudos
SergeyKostrov
Valued Contributor II
1,488 Views
Here is a set of tests for 64-bit Windows running on a multi-core system ( Dell Precision Mobile )
0 Kudos
SergeyKostrov
Valued Contributor II
1,488 Views
[ Test Case 22 - 64-bit / Cores: 1 / CPUs: 2 - Number of OpenMP threads - 2 ] [ Matrix Size: 256 x 256 ] Application - IccTestApp - WIN64_ICC ( 64-bit ) - Release Tests: Start > Test1099 Start < Matrix A, B and C Sizes : 256 x 256 Loop Processing Schema ( LPS ): IKJ Loop Blocking Divider : 1 Sub-Test 1.1 - MxMultA1 - Classic 2D LBOT size: N/A Completed: 0.00194 secs Sub-Test 1.2 - MxMultA2 - Classic 2D LBOT LBOT size: 256x256 elements Completed: 0.00100 secs Sub-Test 1.3 - MxMultA3 - Classic 2D Fused LBOT size: N/A Completed: 0.00581 secs Sub-Test 1.4 - MxMultA4 - Classic 2D Fused LBOT LBOT size: 256x256 elements Completed: 0.00681 secs Sub-Test 2.1 - MxMultB1 - Classic 2D Transposed LBOT size: N/A Completed: 0.00294 secs Sub-Test 2.2 - MxMultB2 - Classic 2D Transposed LBOT LBOT size: 256x256 elements Completed: 0.00294 secs Sub-Test 2.3 - MxMultB3 - Classic 2D Fused Transposed LBOT size: N/A Completed: 0.00387 secs Sub-Test 2.4 - MxMultB4 - Classic 2D Fused Transposed LBOT LBOT size: 256x256 elements Completed: 0.00394 secs Sub-Test 5.1 - MxMultD1 - Classic 1D LBOT size: N/A Completed: 0.01950 secs Sub-Test 5.2 - MxMultD2 - Classic 1D LBOT LBOT size: 256x256 elements Completed: 0.00387 secs > Test1099 End < Tests: Completed
0 Kudos
SergeyKostrov
Valued Contributor II
1,488 Views
[ Test Case 23 - 64-bit / Cores: 1 / CPUs: 2 - Number of OpenMP threads - 4 ] [ Matrix Size: 256 x 256 ] Application - IccTestApp - WIN64_ICC ( 64-bit ) - Release Tests: Start > Test1099 Start < Matrix A, B and C Sizes : 256 x 256 Loop Processing Schema ( LPS ): IKJ Loop Blocking Divider : 1 Sub-Test 1.1 - MxMultA1 - Classic 2D LBOT size: N/A Completed: 0.00194 secs Sub-Test 1.2 - MxMultA2 - Classic 2D LBOT LBOT size: 256x256 elements Completed: 0.00100 secs Sub-Test 1.3 - MxMultA3 - Classic 2D Fused LBOT size: N/A Completed: 0.00581 secs Sub-Test 1.4 - MxMultA4 - Classic 2D Fused LBOT LBOT size: 256x256 elements Completed: 0.00781 secs Sub-Test 2.1 - MxMultB1 - Classic 2D Transposed LBOT size: N/A Completed: 0.00294 secs Sub-Test 2.2 - MxMultB2 - Classic 2D Transposed LBOT LBOT size: 256x256 elements Completed: 0.00294 secs Sub-Test 2.3 - MxMultB3 - Classic 2D Fused Transposed LBOT size: N/A Completed: 0.00488 secs Sub-Test 2.4 - MxMultB4 - Classic 2D Fused Transposed LBOT LBOT size: 256x256 elements Completed: 0.00488 secs Sub-Test 5.1 - MxMultD1 - Classic 1D LBOT size: N/A Completed: 0.02044 secs Sub-Test 5.2 - MxMultD2 - Classic 1D LBOT LBOT size: 256x256 elements Completed: 0.00294 secs > Test1099 End < Tests: Completed
0 Kudos
SergeyKostrov
Valued Contributor II
1,488 Views
[ Test Case 24 - 64-bit / Cores: 1 / CPUs: 2 - Number of OpenMP threads - 8 ] [ Matrix Size: 256 x 256 ] Application - IccTestApp - WIN64_ICC ( 64-bit ) - Release Tests: Start > Test1099 Start < Matrix A, B and C Sizes : 256 x 256 Loop Processing Schema ( LPS ): IKJ Loop Blocking Divider : 1 Sub-Test 1.1 - MxMultA1 - Classic 2D LBOT size: N/A Completed: 0.00194 secs Sub-Test 1.2 - MxMultA2 - Classic 2D LBOT LBOT size: 256x256 elements Completed: 0.00200 secs Sub-Test 1.3 - MxMultA3 - Classic 2D Fused LBOT size: N/A Completed: 0.00681 secs Sub-Test 1.4 - MxMultA4 - Classic 2D Fused LBOT LBOT size: 256x256 elements Completed: 0.00681 secs Sub-Test 2.1 - MxMultB1 - Classic 2D Transposed LBOT size: N/A Completed: 0.00294 secs Sub-Test 2.2 - MxMultB2 - Classic 2D Transposed LBOT LBOT size: 256x256 elements Completed: 0.00294 secs Sub-Test 2.3 - MxMultB3 - Classic 2D Fused Transposed LBOT size: N/A Completed: 0.00387 secs Sub-Test 2.4 - MxMultB4 - Classic 2D Fused Transposed LBOT LBOT size: 256x256 elements Completed: 0.00387 secs Sub-Test 5.1 - MxMultD1 - Classic 1D LBOT size: N/A Completed: 0.02244 secs Sub-Test 5.2 - MxMultD2 - Classic 1D LBOT LBOT size: 256x256 elements Completed: 0.00394 secs > Test1099 End < Tests: Completed
0 Kudos
SergeyKostrov
Valued Contributor II
1,488 Views
[ Test Case 25 - 64-bit / Cores: 1 / CPUs: 2 - Number of OpenMP threads - 16 ] [ Matrix Size: 256 x 256 ] Application - IccTestApp - WIN64_ICC ( 64-bit ) - Release Tests: Start > Test1099 Start < Matrix A, B and C Sizes : 256 x 256 Loop Processing Schema ( LPS ): IKJ Loop Blocking Divider : 1 Sub-Test 1.1 - MxMultA1 - Classic 2D LBOT size: N/A Completed: 0.00100 secs Sub-Test 1.2 - MxMultA2 - Classic 2D LBOT LBOT size: 256x256 elements Completed: 0.00094 secs Sub-Test 1.3 - MxMultA3 - Classic 2D Fused LBOT size: N/A Completed: 0.00581 secs Sub-Test 1.4 - MxMultA4 - Classic 2D Fused LBOT LBOT size: 256x256 elements Completed: 0.00687 secs Sub-Test 2.1 - MxMultB1 - Classic 2D Transposed LBOT size: N/A Completed: 0.00294 secs Sub-Test 2.2 - MxMultB2 - Classic 2D Transposed LBOT LBOT size: 256x256 elements Completed: 0.00294 secs Sub-Test 2.3 - MxMultB3 - Classic 2D Fused Transposed LBOT size: N/A Completed: 0.00387 secs Sub-Test 2.4 - MxMultB4 - Classic 2D Fused Transposed LBOT LBOT size: 256x256 elements Completed: 0.00387 secs Sub-Test 5.1 - MxMultD1 - Classic 1D LBOT size: N/A Completed: 0.02050 secs Sub-Test 5.2 - MxMultD2 - Classic 1D LBOT LBOT size: 256x256 elements Completed: 0.00387 secs > Test1099 End < Tests: Completed
0 Kudos
SergeyKostrov
Valued Contributor II
1,488 Views
[ Test Case 26 - 64-bit / Cores: 1 / CPUs: 2 - Number of OpenMP threads - 1 ] [ Matrix Size: 512 x 512 ] Application - IccTestApp - WIN64_ICC ( 64-bit ) - Release Tests: Start > Test1099 Start < Matrix A, B and C Sizes : 512 x 512 Loop Processing Schema ( LPS ): IKJ Loop Blocking Divider : 1 Sub-Test 1.1 - MxMultA1 - Classic 2D LBOT size: N/A Completed: 0.01850 secs Sub-Test 1.2 - MxMultA2 - Classic 2D LBOT LBOT size: 512x512 elements Completed: 0.01756 secs Sub-Test 1.3 - MxMultA3 - Classic 2D Fused LBOT size: N/A Completed: 0.05069 secs Sub-Test 1.4 - MxMultA4 - Classic 2D Fused LBOT LBOT size: 512x512 elements Completed: 0.04775 secs Sub-Test 2.1 - MxMultB1 - Classic 2D Transposed LBOT size: N/A Completed: 0.01950 secs Sub-Test 2.2 - MxMultB2 - Classic 2D Transposed LBOT LBOT size: 512x512 elements Completed: 0.01563 secs Sub-Test 2.3 - MxMultB3 - Classic 2D Fused Transposed LBOT size: N/A Completed: 0.02731 secs Sub-Test 2.4 - MxMultB4 - Classic 2D Fused Transposed LBOT LBOT size: 512x512 elements Completed: 0.03119 secs Sub-Test 5.1 - MxMultD1 - Classic 1D LBOT size: N/A Completed: 0.15500 secs Sub-Test 5.2 - MxMultD2 - Classic 1D LBOT LBOT size: 512x512 elements Completed: 0.03025 secs > Test1099 End < Tests: Completed
0 Kudos
SergeyKostrov
Valued Contributor II
1,488 Views
[ Test Case 27 - 64-bit / Cores: 1 / CPUs: 2 - Number of OpenMP threads - 2 ] [ Matrix Size: 512 x 512 ] Application - IccTestApp - WIN64_ICC ( 64-bit ) - Release Tests: Start > Test1099 Start < Matrix A, B and C Sizes : 512 x 512 Loop Processing Schema ( LPS ): IKJ Loop Blocking Divider : 1 Sub-Test 1.1 - MxMultA1 - Classic 2D LBOT size: N/A Completed: 0.01169 secs Sub-Test 1.2 - MxMultA2 - Classic 2D LBOT LBOT size: 512x512 elements Completed: 0.01075 secs Sub-Test 1.3 - MxMultA3 - Classic 2D Fused LBOT size: N/A Completed: 0.04775 secs Sub-Test 1.4 - MxMultA4 - Classic 2D Fused LBOT LBOT size: 512x512 elements Completed: 0.04781 secs Sub-Test 2.1 - MxMultB1 - Classic 2D Transposed LBOT size: N/A Completed: 0.01656 secs Sub-Test 2.2 - MxMultB2 - Classic 2D Transposed LBOT LBOT size: 512x512 elements Completed: 0.01656 secs Sub-Test 2.3 - MxMultB3 - Classic 2D Fused Transposed LBOT size: N/A Completed: 0.03119 secs Sub-Test 2.4 - MxMultB4 - Classic 2D Fused Transposed LBOT LBOT size: 512x512 elements Completed: 0.03119 secs Sub-Test 5.1 - MxMultD1 - Classic 1D LBOT size: N/A Completed: 0.14531 secs Sub-Test 5.2 - MxMultD2 - Classic 1D LBOT LBOT size: 512x512 elements Completed: 0.02244 secs > Test1099 End < Tests: Completed
0 Kudos
SergeyKostrov
Valued Contributor II
1,488 Views
[ Test Case 28 - 64-bit / Cores: 1 / CPUs: 2 - Number of OpenMP threads - 4 ] [ Matrix Size: 512 x 512 ] Application - IccTestApp - WIN64_ICC ( 64-bit ) - Release Tests: Start > Test1099 Start < Matrix A, B and C Sizes : 512 x 512 Loop Processing Schema ( LPS ): IKJ Loop Blocking Divider : 1 Sub-Test 1.1 - MxMultA1 - Classic 2D LBOT size: N/A Completed: 0.01075 secs Sub-Test 1.2 - MxMultA2 - Classic 2D LBOT LBOT size: 512x512 elements Completed: 0.01169 secs Sub-Test 1.3 - MxMultA3 - Classic 2D Fused LBOT size: N/A Completed: 0.04975 secs Sub-Test 1.4 - MxMultA4 - Classic 2D Fused LBOT LBOT size: 512x512 elements Completed: 0.05750 secs Sub-Test 2.1 - MxMultB1 - Classic 2D Transposed LBOT size: N/A Completed: 0.01856 secs Sub-Test 2.2 - MxMultB2 - Classic 2D Transposed LBOT LBOT size: 512x512 elements Completed: 0.01750 secs Sub-Test 2.3 - MxMultB3 - Classic 2D Fused Transposed LBOT size: N/A Completed: 0.03512 secs Sub-Test 2.4 - MxMultB4 - Classic 2D Fused Transposed LBOT LBOT size: 512x512 elements Completed: 0.03025 secs Sub-Test 5.1 - MxMultD1 - Classic 1D LBOT size: N/A Completed: 0.17156 secs Sub-Test 5.2 - MxMultD2 - Classic 1D LBOT LBOT size: 512x512 elements Completed: 0.02344 secs > Test1099 End < Tests: Completed
0 Kudos
SergeyKostrov
Valued Contributor II
1,488 Views
[ Test Case 29 - 64-bit / Cores: 1 / CPUs: 2 - Number of OpenMP threads - 8 ] [ Matrix Size: 512 x 512 ] Application - IccTestApp - WIN64_ICC ( 64-bit ) - Release Tests: Start > Test1099 Start < Matrix A, B and C Sizes : 512 x 512 Loop Processing Schema ( LPS ): IKJ Loop Blocking Divider : 1 Sub-Test 1.1 - MxMultA1 - Classic 2D LBOT size: N/A Completed: 0.01169 secs Sub-Test 1.2 - MxMultA2 - Classic 2D LBOT LBOT size: 512x512 elements Completed: 0.01069 secs Sub-Test 1.3 - MxMultA3 - Classic 2D Fused LBOT size: N/A Completed: 0.04681 secs Sub-Test 1.4 - MxMultA4 - Classic 2D Fused LBOT LBOT size: 512x512 elements Completed: 0.04775 secs Sub-Test 2.1 - MxMultB1 - Classic 2D Transposed LBOT size: N/A Completed: 0.01656 secs Sub-Test 2.2 - MxMultB2 - Classic 2D Transposed LBOT LBOT size: 512x512 elements Completed: 0.01663 secs Sub-Test 2.3 - MxMultB3 - Classic 2D Fused Transposed LBOT size: N/A Completed: 0.03019 secs Sub-Test 2.4 - MxMultB4 - Classic 2D Fused Transposed LBOT LBOT size: 512x512 elements Completed: 0.03025 secs Sub-Test 5.1 - MxMultD1 - Classic 1D LBOT size: N/A Completed: 0.14525 secs Sub-Test 5.2 - MxMultD2 - Classic 1D LBOT LBOT size: 512x512 elements Completed: 0.02344 secs > Test1099 End < Tests: Completed
0 Kudos
SergeyKostrov
Valued Contributor II
1,488 Views
[ Test Case 30 - 64-bit / Cores: 1 / CPUs: 2 - Number of OpenMP threads - 16 ] [ Matrix Size: 512 x 512 ] Application - IccTestApp - WIN64_ICC ( 64-bit ) - Release Tests: Start > Test1099 Start < Matrix A, B and C Sizes : 512 x 512 Loop Processing Schema ( LPS ): IKJ Loop Blocking Divider : 1 Sub-Test 1.1 - MxMultA1 - Classic 2D LBOT size: N/A Completed: 0.01169 secs Sub-Test 1.2 - MxMultA2 - Classic 2D LBOT LBOT size: 512x512 elements Completed: 0.01169 secs Sub-Test 1.3 - MxMultA3 - Classic 2D Fused LBOT size: N/A Completed: 0.04875 secs Sub-Test 1.4 - MxMultA4 - Classic 2D Fused LBOT LBOT size: 512x512 elements Completed: 0.04875 secs Sub-Test 2.1 - MxMultB1 - Classic 2D Transposed LBOT size: N/A Completed: 0.01656 secs Sub-Test 2.2 - MxMultB2 - Classic 2D Transposed LBOT LBOT size: 512x512 elements Completed: 0.01756 secs Sub-Test 2.3 - MxMultB3 - Classic 2D Fused Transposed LBOT size: N/A Completed: 0.03212 secs Sub-Test 2.4 - MxMultB4 - Classic 2D Fused Transposed LBOT LBOT size: 512x512 elements Completed: 0.03219 secs Sub-Test 5.1 - MxMultD1 - Classic 1D LBOT size: N/A Completed: 0.15894 secs Sub-Test 5.2 - MxMultD2 - Classic 1D LBOT LBOT size: 512x512 elements Completed: 0.02631 secs > Test1099 End < Tests: Completed
0 Kudos
SergeyKostrov
Valued Contributor II
1,488 Views
[ Test Case 31 - 64-bit / Cores: 1 / CPUs: 2 - Number of OpenMP threads - 1 ] [ Matrix Size: 1024 x 1024 ] Application - IccTestApp - WIN64_ICC ( 64-bit ) - Release Tests: Start > Test1099 Start < Matrix A, B and C Sizes : 1024 x 1024 Loop Processing Schema ( LPS ): IKJ Loop Blocking Divider : 1 Sub-Test 1.1 - MxMultA1 - Classic 2D LBOT size: N/A Completed: 0.14425 secs Sub-Test 1.2 - MxMultA2 - Classic 2D LBOT LBOT size: 1024x1024 elements Completed: 0.11900 secs Sub-Test 1.3 - MxMultA3 - Classic 2D Fused LBOT size: N/A Completed: 0.41337 secs Sub-Test 1.4 - MxMultA4 - Classic 2D Fused LBOT LBOT size: 1024x1024 elements Completed: 0.40950 secs Sub-Test 2.1 - MxMultB1 - Classic 2D Transposed LBOT size: N/A Completed: 0.13075 secs Sub-Test 2.2 - MxMultB2 - Classic 2D Transposed LBOT LBOT size: 1024x1024 elements Completed: 0.13062 secs Sub-Test 2.3 - MxMultB3 - Classic 2D Fused Transposed LBOT size: N/A Completed: 0.23200 secs Sub-Test 2.4 - MxMultB4 - Classic 2D Fused Transposed LBOT LBOT size: 1024x1024 elements Completed: 0.23600 secs Sub-Test 5.1 - MxMultD1 - Classic 1D LBOT size: N/A Completed: 1.27525 secs Sub-Test 5.2 - MxMultD2 - Classic 1D LBOT LBOT size: 1024x1024 elements Completed: 0.22425 secs > Test1099 End < Tests: Completed
0 Kudos
SergeyKostrov
Valued Contributor II
1,488 Views
[ Test Case 32 - 64-bit / Cores: 1 / CPUs: 2 - Number of OpenMP threads - 2 ] [ Matrix Size: 1024 x 1024 ] Application - IccTestApp - WIN64_ICC ( 64-bit ) - Release Tests: Start > Test1099 Start < Matrix A, B and C Sizes : 1024 x 1024 Loop Processing Schema ( LPS ): IKJ Loop Blocking Divider : 1 Sub-Test 1.1 - MxMultA1 - Classic 2D LBOT size: N/A Completed: 0.10925 secs Sub-Test 1.2 - MxMultA2 - Classic 2D LBOT LBOT size: 1024x1024 elements Completed: 0.11112 secs Sub-Test 1.3 - MxMultA3 - Classic 2D Fused LBOT size: N/A Completed: 0.39788 secs Sub-Test 1.4 - MxMultA4 - Classic 2D Fused LBOT LBOT size: 1024x1024 elements Completed: 0.39775 secs Sub-Test 2.1 - MxMultB1 - Classic 2D Transposed LBOT size: N/A Completed: 0.12875 secs Sub-Test 2.2 - MxMultB2 - Classic 2D Transposed LBOT LBOT size: 1024x1024 elements Completed: 0.12863 secs Sub-Test 2.3 - MxMultB3 - Classic 2D Fused Transposed LBOT size: N/A Completed: 0.23988 secs Sub-Test 2.4 - MxMultB4 - Classic 2D Fused Transposed LBOT LBOT size: 1024x1024 elements Completed: 0.24175 secs Sub-Test 5.1 - MxMultD1 - Classic 1D LBOT size: N/A Completed: 1.16025 secs Sub-Test 5.2 - MxMultD2 - Classic 1D LBOT LBOT size: 1024x1024 elements Completed: 0.18725 secs > Test1099 End < Tests: Completed
0 Kudos
SergeyKostrov
Valued Contributor II
1,488 Views
[ Test Case 33 - 64-bit / Cores: 1 / CPUs: 2 - Number of OpenMP threads - 4 ] [ Matrix Size: 1024 x 1024 ] Application - IccTestApp - WIN64_ICC ( 64-bit ) - Release Tests: Start > Test1099 Start < Matrix A, B and C Sizes : 1024 x 1024 Loop Processing Schema ( LPS ): IKJ Loop Blocking Divider : 1 Sub-Test 1.1 - MxMultA1 - Classic 2D LBOT size: N/A Completed: 0.09550 secs Sub-Test 1.2 - MxMultA2 - Classic 2D LBOT LBOT size: 1024x1024 elements Completed: 0.09750 secs Sub-Test 1.3 - MxMultA3 - Classic 2D Fused LBOT size: N/A Completed: 0.40375 secs Sub-Test 1.4 - MxMultA4 - Classic 2D Fused LBOT LBOT size: 1024x1024 elements Completed: 0.39975 secs Sub-Test 2.1 - MxMultB1 - Classic 2D Transposed LBOT size: N/A Completed: 0.13450 secs Sub-Test 2.2 - MxMultB2 - Classic 2D Transposed LBOT LBOT size: 1024x1024 elements Completed: 0.13250 secs Sub-Test 2.3 - MxMultB3 - Classic 2D Fused Transposed LBOT size: N/A Completed: 0.25163 secs Sub-Test 2.4 - MxMultB4 - Classic 2D Fused Transposed LBOT LBOT size: 1024x1024 elements Completed: 0.24375 secs Sub-Test 5.1 - MxMultD1 - Classic 1D LBOT size: N/A Completed: 1.17587 secs Sub-Test 5.2 - MxMultD2 - Classic 1D LBOT LBOT size: 1024x1024 elements Completed: 0.18712 secs > Test1099 End < Tests: Completed
0 Kudos
SergeyKostrov
Valued Contributor II
1,492 Views
[ Test Case 34 - 64-bit / Cores: 1 / CPUs: 2 - Number of OpenMP threads - 8 ] [ Matrix Size: 1024 x 1024 ] Application - IccTestApp - WIN64_ICC ( 64-bit ) - Release Tests: Start > Test1099 Start < Matrix A, B and C Sizes : 1024 x 1024 Loop Processing Schema ( LPS ): IKJ Loop Blocking Divider : 1 Sub-Test 1.1 - MxMultA1 - Classic 2D LBOT size: N/A Completed: 0.10725 secs Sub-Test 1.2 - MxMultA2 - Classic 2D LBOT LBOT size: 1024x1024 elements Completed: 0.10537 secs Sub-Test 1.3 - MxMultA3 - Classic 2D Fused LBOT size: N/A Completed: 0.43088 secs Sub-Test 1.4 - MxMultA4 - Classic 2D Fused LBOT LBOT size: 1024x1024 elements Completed: 0.40762 secs Sub-Test 2.1 - MxMultB1 - Classic 2D Transposed LBOT size: N/A Completed: 0.13075 secs Sub-Test 2.2 - MxMultB2 - Classic 2D Transposed LBOT LBOT size: 1024x1024 elements Completed: 0.13450 secs Sub-Test 2.3 - MxMultB3 - Classic 2D Fused Transposed LBOT size: N/A Completed: 0.25350 secs Sub-Test 2.4 - MxMultB4 - Classic 2D Fused Transposed LBOT LBOT size: 1024x1024 elements Completed: 0.25938 secs Sub-Test 5.1 - MxMultD1 - Classic 1D LBOT size: N/A Completed: 1.18562 secs Sub-Test 5.2 - MxMultD2 - Classic 1D LBOT LBOT size: 1024x1024 elements Completed: 0.21262 secs > Test1099 End < Tests: Completed
0 Kudos
SergeyKostrov
Valued Contributor II
1,492 Views
[ Test Case 35 - 64-bit / Cores: 1 / CPUs: 2 - Number of OpenMP threads - 16 ] [ Matrix Size: 1024 x 1024 ] Application - IccTestApp - WIN64_ICC ( 64-bit ) - Release Tests: Start > Test1099 Start < Matrix A, B and C Sizes : 1024 x 1024 Loop Processing Schema ( LPS ): IKJ Loop Blocking Divider : 1 Sub-Test 1.1 - MxMultA1 - Classic 2D LBOT size: N/A Completed: 0.09750 secs Sub-Test 1.2 - MxMultA2 - Classic 2D LBOT LBOT size: 1024x1024 elements Completed: 0.09950 secs Sub-Test 1.3 - MxMultA3 - Classic 2D Fused LBOT size: N/A Completed: 0.41725 secs Sub-Test 1.4 - MxMultA4 - Classic 2D Fused LBOT LBOT size: 1024x1024 elements Completed: 0.41537 secs Sub-Test 2.1 - MxMultB1 - Classic 2D Transposed LBOT size: N/A Completed: 0.12863 secs Sub-Test 2.2 - MxMultB2 - Classic 2D Transposed LBOT LBOT size: 1024x1024 elements Completed: 0.13262 secs Sub-Test 2.3 - MxMultB3 - Classic 2D Fused Transposed LBOT size: N/A Completed: 0.24762 secs Sub-Test 2.4 - MxMultB4 - Classic 2D Fused Transposed LBOT LBOT size: 1024x1024 elements Completed: 0.25750 secs Sub-Test 5.1 - MxMultD1 - Classic 1D LBOT size: N/A Completed: 1.23038 secs Sub-Test 5.2 - MxMultD2 - Classic 1D LBOT LBOT size: 1024x1024 elements Completed: 0.18925 secs > Test1099 End < Tests: Completed
0 Kudos
Reply