CUDA Caffe NVIDIA Comparison

CUDA 8.0 + cuDNN Caffe deep learning benchmarks with many different GPUs. Tests by Michael Larabel.

Compare your own system(s) to this result file with the Phoronix Test Suite by running the command: phoronix-test-suite benchmark 1702028-TA-1611066TA46
Jump To Table - Results

View

Do Not Show Noisy Results
Do Not Show Results With Incomplete Data
Do Not Show Results With Little Change/Spread
List Notable Results

Statistics

Show Overall Harmonic Mean(s)
Show Overall Geometric Mean
Show Wins / Losses Counts (Pie Chart)
Normalize Results
Remove Outliers Before Calculating Averages

Graph Settings

Force Line Graphs Where Applicable
Convert To Scalar Where Applicable
Disable Color Branding
Prefer Vertical Bar Graphs
No Box Plots
On Line Graphs With Missing Data, Connect The Line Gaps

Additional Graphs

Show Perf Per Core/Thread Calculation Graphs Where Applicable
Show Perf Per Clock Calculation Graphs Where Applicable

Multi-Way Comparison

Condense Multi-Option Tests Into Single Result Graphs
Condense Test Profiles With Multiple Version Results Into Single Result Graphs

Table

Show Detailed System Result Table

Run Management

Highlight
Result
Hide
Result
Result
Identifier
Performance Per
Dollar
Date
Run
  Test
  Duration
GTX 680
November 05 2016
 
GTX 760
November 05 2016
 
GTX 780 Ti
November 05 2016
 
GTX 950
November 04 2016
 
GTX 960
November 04 2016
 
GTX 970
November 04 2016
 
GTX 980
November 04 2016
 
GTX 980 Ti
November 04 2016
 
GTX 1050
November 04 2016
 
GTX 1050 Ti
November 05 2016
 
GTX 1060
November 04 2016
 
GTX 1070
November 04 2016
 
GTX 1080
November 04 2016
 
ubu_ml2
February 02 2017
 
ubu_ml_375.26
February 02 2017
 
Invert Hiding All Results Option
 

Only show results where is faster than
Only show results matching title/arguments (delimit multiple options with a comma):
Do not show results matching title/arguments (delimit multiple options with a comma):


CUDA Caffe NVIDIA Comparison CUDA 8.0 + cuDNN Caffe deep learning benchmarks with many different GPUs. Tests by Michael Larabel. GTX 680: Processor: Intel Xeon E3-1280 v5 @ 4.00GHz (8 Cores), Motherboard: MSI C236A WORKSTATION (MS-7998) v1.0, Chipset: Intel Sky Lake, Memory: 16384MB, Disk: 256GB INTEL SSDPEKKW256G7, Graphics: NVIDIA GeForce GTX 680 2048MB (1006/3004MHz), Audio: Realtek ALC1150, Network: Intel Connection OS: Ubuntu 16.04, Kernel: 4.8.4-040804-generic (x86_64), Desktop: Unity 7.4.0, Display Server: X Server 1.18.4, Display Driver: NVIDIA 375.10, OpenGL: 4.5.0, Vulkan: 1.0.8, Compiler: GCC 5.4.0 20160609 + LLVM 3.8.0 + CUDA 8.0, File-System: ext4, Screen Resolution: 3840x2160 GTX 760: Processor: Intel Xeon E3-1280 v5 @ 4.00GHz (8 Cores), Motherboard: MSI C236A WORKSTATION (MS-7998) v1.0, Chipset: Intel Sky Lake, Memory: 16384MB, Disk: 256GB INTEL SSDPEKKW256G7, Graphics: NVIDIA GeForce GTX 760 2048MB (980/3004MHz), Audio: Realtek ALC1150, Network: Intel Connection OS: Ubuntu 16.04, Kernel: 4.8.4-040804-generic (x86_64), Desktop: Unity 7.4.0, Display Server: X Server 1.18.4, Display Driver: NVIDIA 375.10, OpenGL: 4.5.0, Vulkan: 1.0.8, Compiler: GCC 5.4.0 20160609 + LLVM 3.8.0 + CUDA 8.0, File-System: ext4, Screen Resolution: 3840x2160 GTX 780 Ti: Processor: Intel Xeon E3-1280 v5 @ 4.00GHz (8 Cores), Motherboard: MSI C236A WORKSTATION (MS-7998) v1.0, Chipset: Intel Sky Lake, Memory: 16384MB, Disk: 256GB INTEL SSDPEKKW256G7, Graphics: NVIDIA GeForce GTX 780 Ti 3072MB (875/3500MHz), Audio: Realtek ALC1150, Network: Intel Connection OS: Ubuntu 16.04, Kernel: 4.8.4-040804-generic (x86_64), Desktop: Unity 7.4.0, Display Server: X Server 1.18.4, Display Driver: NVIDIA 375.10, OpenGL: 4.5.0, Vulkan: 1.0.8, Compiler: GCC 5.4.0 20160609 + LLVM 3.8.0 + CUDA 8.0, File-System: ext4, Screen Resolution: 3840x2160 GTX 950: Processor: Intel Xeon E3-1280 v5 @ 4.00GHz (8 Cores), Motherboard: MSI C236A WORKSTATION (MS-7998) v1.0, Chipset: Intel Sky Lake, Memory: 16384MB, Disk: 256GB INTEL SSDPEKKW256G7, Graphics: eVGA NVIDIA GeForce GTX 950 2048MB (1201/3304MHz), Audio: Realtek ALC1150, Network: Intel Connection OS: Ubuntu 16.04, Kernel: 4.8.4-040804-generic (x86_64), Desktop: Unity 7.4.0, Display Server: X Server 1.18.4, Display Driver: NVIDIA 375.10, OpenGL: 4.5.0, Vulkan: 1.0.8, Compiler: GCC 5.4.0 20160609 + LLVM 3.8.0 + CUDA 8.0, File-System: ext4, Screen Resolution: 3840x2160 GTX 960: Processor: Intel Xeon E3-1280 v5 @ 4.00GHz (8 Cores), Motherboard: MSI C236A WORKSTATION (MS-7998) v1.0, Chipset: Intel Sky Lake, Memory: 16384MB, Disk: 256GB INTEL SSDPEKKW256G7, Graphics: eVGA NVIDIA GeForce GTX 960 2048MB (1277/3505MHz), Audio: Realtek ALC1150, Network: Intel Connection OS: Ubuntu 16.04, Kernel: 4.8.4-040804-generic (x86_64), Desktop: Unity 7.4.0, Display Server: X Server 1.18.4, Display Driver: NVIDIA 375.10, OpenGL: 4.5.0, Vulkan: 1.0.8, Compiler: GCC 5.4.0 20160609 + LLVM 3.8.0 + CUDA 8.0, File-System: ext4, Screen Resolution: 3840x2160 GTX 970: Processor: Intel Xeon E3-1280 v5 @ 4.00GHz (8 Cores), Motherboard: MSI C236A WORKSTATION (MS-7998) v1.0, Chipset: Intel Sky Lake, Memory: 16384MB, Disk: 256GB INTEL SSDPEKKW256G7, Graphics: eVGA NVIDIA GeForce GTX 970 4096MB (1163/3505MHz), Audio: Realtek ALC1150, Network: Intel Connection OS: Ubuntu 16.04, Kernel: 4.8.4-040804-generic (x86_64), Desktop: Unity 7.4.0, Display Server: X Server 1.18.4, Display Driver: NVIDIA 375.10, OpenGL: 4.5.0, Vulkan: 1.0.8, Compiler: GCC 5.4.0 20160609 + LLVM 3.8.0 + CUDA 8.0, File-System: ext4, Screen Resolution: 3840x2160 GTX 980: Processor: Intel Xeon E3-1280 v5 @ 4.00GHz (8 Cores), Motherboard: MSI C236A WORKSTATION (MS-7998) v1.0, Chipset: Intel Sky Lake, Memory: 16384MB, Disk: 256GB INTEL SSDPEKKW256G7, Graphics: NVIDIA GeForce GTX 980 4096MB (135/324MHz), Audio: Realtek ALC1150, Network: Intel Connection OS: Ubuntu 16.04, Kernel: 4.8.4-040804-generic (x86_64), Desktop: Unity 7.4.0, Display Server: X Server 1.18.4, Display Driver: NVIDIA 375.10, OpenGL: 4.5.0, Vulkan: 1.0.8, Compiler: GCC 5.4.0 20160609 + LLVM 3.8.0 + CUDA 8.0, File-System: ext4, Screen Resolution: 3840x2160 GTX 980 Ti: Processor: Intel Xeon E3-1280 v5 @ 4.00GHz (8 Cores), Motherboard: MSI C236A WORKSTATION (MS-7998) v1.0, Chipset: Intel Sky Lake, Memory: 16384MB, Disk: 256GB INTEL SSDPEKKW256G7, Graphics: NVIDIA GeForce GTX 980 Ti 6144MB (999/3505MHz), Audio: Realtek ALC1150, Network: Intel Connection OS: Ubuntu 16.04, Kernel: 4.8.4-040804-generic (x86_64), Desktop: Unity 7.4.0, Display Server: X Server 1.18.4, Display Driver: NVIDIA 375.10, OpenGL: 4.5.0, Vulkan: 1.0.8, Compiler: GCC 5.4.0 20160609 + LLVM 3.8.0 + CUDA 8.0, File-System: ext4, Screen Resolution: 3840x2160 GTX 1050: Processor: Intel Xeon E3-1280 v5 @ 4.00GHz (8 Cores), Motherboard: MSI C236A WORKSTATION (MS-7998) v1.0, Chipset: Intel Sky Lake, Memory: 16384MB, Disk: 256GB INTEL SSDPEKKW256G7, Graphics: Zotac NVIDIA GeForce GTX 1050 2048MB (1316/3504MHz), Audio: Realtek ALC1150, Network: Intel Connection OS: Ubuntu 16.04, Kernel: 4.8.4-040804-generic (x86_64), Desktop: Unity 7.4.0, Display Server: X Server 1.18.4, Display Driver: NVIDIA 375.10, OpenGL: 4.5.0, Vulkan: 1.0.8, Compiler: GCC 5.4.0 20160609 + LLVM 3.8.0 + CUDA 8.0, File-System: ext4, Screen Resolution: 3840x2160 GTX 1050 Ti: Processor: Intel Xeon E3-1280 v5 @ 4.00GHz (8 Cores), Motherboard: MSI C236A WORKSTATION (MS-7998) v1.0, Chipset: Intel Sky Lake, Memory: 16384MB, Disk: 256GB INTEL SSDPEKKW256G7, Graphics: eVGA NVIDIA GeForce GTX 1050 Ti 4096MB (1341/3504MHz), Audio: Realtek ALC1150, Network: Intel Connection OS: Ubuntu 16.04, Kernel: 4.8.4-040804-generic (x86_64), Desktop: Unity 7.4.0, Display Server: X Server 1.18.4, Display Driver: NVIDIA 375.10, OpenGL: 4.5.0, Vulkan: 1.0.8, Compiler: GCC 5.4.0 20160609 + LLVM 3.8.0 + CUDA 8.0, File-System: ext4, Screen Resolution: 3840x2160 GTX 1060: Processor: Intel Xeon E3-1280 v5 @ 4.00GHz (8 Cores), Motherboard: MSI C236A WORKSTATION (MS-7998) v1.0, Chipset: Intel Sky Lake, Memory: 16384MB, Disk: 256GB INTEL SSDPEKKW256G7, Graphics: NVIDIA GeForce GTX 1060 6GB 6144MB (1506/4006MHz), Audio: Realtek ALC1150, Network: Intel Connection OS: Ubuntu 16.04, Kernel: 4.8.4-040804-generic (x86_64), Desktop: Unity 7.4.0, Display Server: X Server 1.18.4, Display Driver: NVIDIA 375.10, OpenGL: 4.5.0, Vulkan: 1.0.8, Compiler: GCC 5.4.0 20160609 + LLVM 3.8.0 + CUDA 8.0, File-System: ext4, Screen Resolution: 3840x2160 GTX 1070: Processor: Intel Xeon E3-1280 v5 @ 4.00GHz (8 Cores), Motherboard: MSI C236A WORKSTATION (MS-7998) v1.0, Chipset: Intel Sky Lake, Memory: 16384MB, Disk: 256GB INTEL SSDPEKKW256G7, Graphics: NVIDIA GeForce GTX 1070 8192MB (1505/4006MHz), Audio: Realtek ALC1150, Network: Intel Connection OS: Ubuntu 16.04, Kernel: 4.8.4-040804-generic (x86_64), Desktop: Unity 7.4.0, Display Server: X Server 1.18.4, Display Driver: NVIDIA 375.10, OpenGL: 4.5.0, Vulkan: 1.0.8, Compiler: GCC 5.4.0 20160609 + LLVM 3.8.0 + CUDA 8.0, File-System: ext4, Screen Resolution: 3840x2160 GTX 1080: Processor: Intel Xeon E3-1280 v5 @ 4.00GHz (8 Cores), Motherboard: MSI C236A WORKSTATION (MS-7998) v1.0, Chipset: Intel Sky Lake, Memory: 16384MB, Disk: 256GB INTEL SSDPEKKW256G7, Graphics: NVIDIA GeForce GTX 1080 8192MB (1615/5005MHz), Audio: Realtek ALC1150, Network: Intel Connection OS: Ubuntu 16.04, Kernel: 4.8.4-040804-generic (x86_64), Desktop: Unity 7.4.0, Display Server: X Server 1.18.4, Display Driver: NVIDIA 375.10, OpenGL: 4.5.0, Vulkan: 1.0.8, Compiler: GCC 5.4.0 20160609 + LLVM 3.8.0 + CUDA 8.0, File-System: ext4, Screen Resolution: 3840x2160 ubu_ml2: Processor: 2 x Intel 0000 @ 3.00GHz (48 Cores), Motherboard: Supermicro X10DRi-LN4+ v1.01, Chipset: Intel Xeon E7 v4/Xeon, Memory: 64512MB, Disk: 1000GB My Passport 0820, Audio: NVIDIA Device 10f0, Network: Intel I350 Gigabit Connection OS: Ubuntu 16.04, Kernel: 4.4.0-59-generic (x86_64), Vulkan: 1.0.8, Compiler: GCC 5.4.0 20160609 + CUDA 8.0, File-System: ext4, Screen Resolution: 1024x768 ubu_ml_375.26: Processor: 2 x Intel 0000 @ 3.00GHz (48 Cores), Motherboard: Supermicro X10DRi-LN4+ v1.01, Chipset: Intel Xeon E7 v4/Xeon, Memory: 64512MB, Disk: 1000GB My Passport 0820, Audio: NVIDIA Device 10f0, Network: Intel I350 Gigabit Connection OS: Ubuntu 16.04, Kernel: 4.4.0-59-generic (x86_64), Vulkan: 1.0.24, Compiler: GCC 5.4.0 20160609 + CUDA 8.0, File-System: ext4, Screen Resolution: 1024x768 System Power Consumption Monitor Phoronix Test Suite System Monitoring Watts GTX 680 ..... MIN: 51 AVG: 193 MAX: 205 GTX 760 ..... MIN: 50 AVG: 197 MAX: 211 GTX 780 Ti .. MIN: 49 AVG: 242 MAX: 285 GTX 950 ..... MIN: 42 AVG: 127 MAX: 141 GTX 960 ..... MIN: 43 AVG: 136 MAX: 153 GTX 970 ..... MIN: 44 AVG: 162 MAX: 191 GTX 980 ..... MIN: 46 AVG: 178 MAX: 212 GTX 980 Ti .. MIN: 49 AVG: 194 MAX: 243 GTX 1050 .... MIN: 37 AVG: 97 MAX: 106 GTX 1050 Ti . MIN: 36 AVG: 90 MAX: 99 GTX 1060 .... MIN: 38 AVG: 127 MAX: 149 GTX 1070 .... MIN: 40 AVG: 147 MAX: 186 GTX 1080 .... MIN: 41 AVG: 165 MAX: 203 GPU Temperature Monitor Phoronix Test Suite System Monitoring Celsius GTX 680 ..... MIN: 40.0 AVG: 73.6 MAX: 79.0 GTX 760 ..... MIN: 38.0 AVG: 76.1 MAX: 81.0 GTX 780 Ti .. MIN: 37.0 AVG: 70.4 MAX: 81.0 GTX 950 ..... MIN: 33.0 AVG: 63.2 MAX: 70.0 GTX 960 ..... MIN: 31.0 AVG: 62.5 MAX: 71.0 GTX 970 ..... MIN: 30.0 AVG: 50.3 MAX: 59.0 GTX 980 ..... MIN: 36.0 AVG: 61.1 MAX: 75.0 GTX 980 Ti .. MIN: 38.0 AVG: 64.3 MAX: 79.0 GTX 1050 .... MIN: 28.0 AVG: 46.2 MAX: 52.0 GTX 1050 Ti . MIN: 30.0 AVG: 51.1 MAX: 59.0 GTX 1060 .... MIN: 30.0 AVG: 47.8 MAX: 58.0 GTX 1070 .... MIN: 32.0 AVG: 52.1 MAX: 65.0 GTX 1080 .... MIN: 31.0 AVG: 51.0 MAX: 64.0 Caffe AlexNet 2016-06-11 System Power Consumption Monitor Watts < Lower Is Better GTX 680 ..... MIN: 51 AVG: 201 MAX: 205 GTX 760 ..... MIN: 50 AVG: 205 MAX: 211 GTX 780 Ti .. MIN: 58 AVG: 263 MAX: 285 GTX 950 ..... MIN: 42 AVG: 137 MAX: 141 GTX 960 ..... MIN: 114 AVG: 150 MAX: 153 GTX 970 ..... MIN: 44 AVG: 181 MAX: 191 GTX 980 ..... MIN: 46 AVG: 198 MAX: 212 GTX 980 Ti .. MIN: 49 AVG: 226 MAX: 243 GTX 1050 .... MIN: 37 AVG: 102 MAX: 106 GTX 1050 Ti . MIN: 36 AVG: 95 MAX: 99 GTX 1060 .... MIN: 101 AVG: 145 MAX: 149 GTX 1070 .... MIN: 40 AVG: 165 MAX: 186 GTX 1080 .... MIN: 41 AVG: 182 MAX: 203 Caffe AlexNet 2016-06-11 GPU Temperature Monitor Celsius < Lower Is Better GTX 680 ..... MIN: 69 AVG: 77 MAX: 79 GTX 760 ..... MIN: 63 AVG: 79 MAX: 81 GTX 780 Ti .. MIN: 59 AVG: 75 MAX: 81 GTX 950 ..... MIN: 64 AVG: 68 MAX: 70 GTX 960 ..... MIN: 51 AVG: 69 MAX: 71 GTX 970 ..... MIN: 46 AVG: 55 MAX: 59 GTX 980 ..... MIN: 57 AVG: 68 MAX: 75 GTX 980 Ti .. MIN: 59 AVG: 72 MAX: 79 GTX 1050 .... MIN: 43 AVG: 50 MAX: 52 GTX 1050 Ti . MIN: 48 AVG: 56 MAX: 59 GTX 1060 .... MIN: 37 AVG: 52 MAX: 58 GTX 1070 .... MIN: 42 AVG: 58 MAX: 65 GTX 1080 .... MIN: 43 AVG: 56 MAX: 64 Caffe AlexNet 2016-06-11 Build: CUDA Googlenet Milli-Seconds < Lower Is Better GTX 680 ....... 138342.00 |============================================ GTX 760 ....... 164643.00 |==================================================== GTX 780 Ti .... 65152.00 |===================== GTX 950 ....... 69528.20 |====================== GTX 960 ....... 60318.23 |=================== GTX 970 ....... 40193.77 |============= GTX 980 ....... 36217.23 |=========== GTX 980 Ti .... 31349.83 |========== GTX 1050 ...... 70347.70 |====================== GTX 1050 Ti ... 61541.57 |=================== GTX 1060 ...... 37604.73 |============ GTX 1070 ...... 27661.90 |========= GTX 1080 ...... 24019.97 |======== ubu_ml2 ....... 25975.07 |======== ubu_ml_375.26 . 25979.53 |======== Caffe AlexNet 2016-06-11 System Power Consumption Monitor Watts < Lower Is Better GTX 680 ..... MIN: 183 AVG: 190 MAX: 195 GTX 760 ..... MIN: 173 AVG: 191 MAX: 199 GTX 780 Ti .. MIN: 196 AVG: 256 MAX: 267 GTX 950 ..... MIN: 125 AVG: 130 MAX: 133 GTX 960 ..... MIN: 80 AVG: 130 MAX: 137 GTX 970 ..... MIN: 91 AVG: 163 MAX: 175 GTX 980 ..... MIN: 161 AVG: 186 MAX: 192 GTX 980 Ti .. MIN: 122 AVG: 211 MAX: 235 GTX 1050 .... MIN: 94 AVG: 99 MAX: 100 GTX 1050 Ti . MIN: 59 AVG: 94 MAX: 97 GTX 1060 .... MIN: 133 AVG: 134 MAX: 136 GTX 1070 .... MIN: 81 AVG: 149 MAX: 171 GTX 1080 .... MIN: 186 AVG: 189 MAX: 193 Caffe AlexNet 2016-06-11 GPU Temperature Monitor Celsius < Lower Is Better GTX 680 ..... MIN: 52 AVG: 68 MAX: 75 GTX 760 ..... MIN: 48 AVG: 72 MAX: 81 GTX 780 Ti .. MIN: 45 AVG: 62 MAX: 72 GTX 950 ..... MIN: 46 AVG: 57 MAX: 67 GTX 960 ..... MIN: 43 AVG: 54 MAX: 63 GTX 970 ..... MIN: 41 AVG: 46 MAX: 50 GTX 980 ..... MIN: 45 AVG: 51 MAX: 58 GTX 980 Ti .. MIN: 46 AVG: 53 MAX: 60 GTX 1050 .... MIN: 37 AVG: 42 MAX: 46 GTX 1050 Ti . MIN: 39 AVG: 45 MAX: 50 GTX 1060 .... MIN: 36 AVG: 41 MAX: 46 GTX 1070 .... MIN: 39 AVG: 43 MAX: 48 GTX 1080 .... MIN: 38 AVG: 43 MAX: 47 Caffe AlexNet 2016-06-11 Build: CUDA AlexNet Milli-Seconds < Lower Is Better GTX 680 ....... 54520.37 |=========================================== GTX 760 ....... 66711.23 |===================================================== GTX 780 Ti .... 28177.90 |====================== GTX 950 ....... 30783.13 |======================== GTX 960 ....... 27512.80 |====================== GTX 970 ....... 16987.30 |============= GTX 980 ....... 15013.43 |============ GTX 980 Ti .... 11652.17 |========= GTX 1050 ...... 30970.00 |========================= GTX 1050 Ti ... 27452.53 |====================== GTX 1060 ...... 16184.70 |============= GTX 1070 ...... 11438.00 |========= GTX 1080 ...... 9630.36 |======== ubu_ml2 ....... 9452.12 |======== ubu_ml_375.26 . 9781.05 |========