GPU运算能力对比
文章目录
- 计算能力换算
- 信息
- 1080TI
- 1080
- Jteson
- 对比
计算能力换算
理论峰值 ＝ GPU芯片数量*GPU Boost主频*核心数量*单个时钟周期内能处理的浮点计算次数
只不过在GPU里单精度和双精度的浮点计算能力需要分开计算，以最新的Tesla P100为例：
双精度理论峰值 ＝ FP64 Cores ＊ GPU Boost Clock ＊ 2 ＝ 1792 ＊1.48GHz＊2 = 5.3 TFlops
单精度理论峰值 ＝ FP32 cores ＊ GPU Boost Clock ＊ 2 ＝ 3584 ＊ 1.58GHz ＊ 2 ＝ 10.6 TFlop
信息
# 1080TI
Total amount of global memory: 11172 MBytes (11715084288 bytes)
(28) Multiprocessors, (128) CUDA Cores/MP: 3584 CUDA Cores
GPU Max Clock rate: 1582 MHz (1.58 GHz)
Memory Clock rate: 5505 Mhz
Memory Bus Width: 352-bit
L2 Cache Size: 2883584 bytes
# 1080
Total amount of global memory: 8111 MBytes (8504868864 bytes)
(20) Multiprocessors, (128) CUDA Cores/MP: 2560 CUDA Cores
GPU Max Clock rate: 1734 MHz (1.73 GHz)
Memory Clock rate: 5005 Mhz
Memory Bus Width: 256-bit
L2 Cache Size: 2097152 bytes
1080TI
~/NVIDIA_CUDA-8.0_Samples/7_CUDALibraries/batchCUBLAS$ export CUDA_VISIBLE_DEVICES=0
~/NVIDIA_CUDA-8.0_Samples/7_CUDALibraries/batchCUBLAS$ ./batchCUBLAS -m1024 -n1024 -k1024
batchCUBLAS Starting...
GPU Device 0: "GeForce GTX 1080 Ti" with compute capability 6.1
==== Running single kernels ====
Testing sgemm
#### args: ta=0 tb=0 m=1024 n=1024 k=1024 alpha = (0xbf800000, -1) beta= (0x40000000, 2)
#### args: lda=1024 ldb=1024 ldc=1024
^^^^ elapsed = 0.00037980 sec GFLOPS=5654.24
@@@@ sgemm test OK
Testing dgemm
#### args: ta=0 tb=0 m=1024 n=1024 k=1024 alpha = (0x0000000000000000, 0) beta= (0x0000000000000000, 0)
#### args: lda=1024 ldb=1024 ldc=1024
^^^^ elapsed = 0.00894690 sec GFLOPS=240.026
@@@@ dgemm test OK
==== Running N=10 without streams ====
Testing sgemm
#### args: ta=0 tb=0 m=1024 n=1024 k=1024 alpha = (0xbf800000, -1) beta= (0x00000000, 0)
#### args: lda=1024 ldb=1024 ldc=1024
^^^^ elapsed = 0.00294209 sec GFLOPS=7299.19
@@@@ sgemm test OK
Testing dgemm
#### args: ta=0 tb=0 m=1024 n=1024 k=1024 alpha = (0xbff0000000000000, -1) beta= (0x0000000000000000, 0)
#### args: lda=1024 ldb=1024 ldc=1024
^^^^ elapsed = 0.07993412 sec GFLOPS=268.657
@@@@ dgemm test OK
==== Running N=10 with streams ====
Testing sgemm
#### args: ta=0 tb=0 m=1024 n=1024 k=1024 alpha = (0x40000000, 2) beta= (0x40000000, 2)
#### args: lda=1024 ldb=1024 ldc=1024
^^^^ elapsed = 0.00224590 sec GFLOPS=9561.78
@@@@ sgemm test OK
Testing dgemm
#### args: ta=0 tb=0 m=1024 n=1024 k=1024 alpha = (0xbff0000000000000, -1) beta= (0x0000000000000000, 0)
#### args: lda=1024 ldb=1024 ldc=1024
^^^^ elapsed = 0.05540895 sec GFLOPS=387.57
@@@@ dgemm test OK
==== Running N=10 batched ====
Testing sgemm
#### args: ta=0 tb=0 m=1024 n=1024 k=1024 alpha = (0x3f800000, 1) beta= (0xbf800000, -1)
#### args: lda=1024 ldb=1024 ldc=1024
^^^^ elapsed = 0.00197387 sec GFLOPS=10879.6
@@@@ sgemm test OK
Testing dgemm
#### args: ta=0 tb=0 m=1024 n=1024 k=1024 alpha = (0xbff0000000000000, -1) beta= (0x4000000000000000, 2)
#### args: lda=1024 ldb=1024 ldc=1024
^^^^ elapsed = 0.05372214 sec GFLOPS=399.739
@@@@ dgemm test OK
Test Summary
0 error(s)
1080
liu@iridescent:~/NVIDIA_CUDA-8.0_Samples/7_CUDALibraries/batchCUBLAS$ export CUDA_VISIBLE_DEVICES=1
liu@iridescent:~/NVIDIA_CUDA-8.0_Samples/7_CUDALibraries/batchCUBLAS$ ./batchCUBLAS -m1024 -n1024 -k1024
batchCUBLAS Starting...
GPU Device 0: "GeForce GTX 1080" with compute capability 6.1
==== Running single kernels ====
Testing sgemm
#### args: ta=0 tb=0 m=1024 n=1024 k=1024 alpha = (0xbf800000, -1) beta= (0x40000000, 2)
#### args: lda=1024 ldb=1024 ldc=1024
^^^^ elapsed = 0.00060892 sec GFLOPS=3526.7
@@@@ sgemm test OK
Testing dgemm
#### args: ta=0 tb=0 m=1024 n=1024 k=1024 alpha = (0x0000000000000000, 0) beta= (0x0000000000000000, 0)
#### args: lda=1024 ldb=1024 ldc=1024
^^^^ elapsed = 0.00993085 sec GFLOPS=216.244
@@@@ dgemm test OK
==== Running N=10 without streams ====
Testing sgemm
#### args: ta=0 tb=0 m=1024 n=1024 k=1024 alpha = (0xbf800000, -1) beta= (0x00000000, 0)
#### args: lda=1024 ldb=1024 ldc=1024
^^^^ elapsed = 0.00369406 sec GFLOPS=5813.35
@@@@ sgemm test OK
Testing dgemm
#### args: ta=0 tb=0 m=1024 n=1024 k=1024 alpha = (0xbff0000000000000, -1) beta= (0x0000000000000000, 0)
#### args: lda=1024 ldb=1024 ldc=1024
^^^^ elapsed = 0.09741306 sec GFLOPS=220.451
@@@@ dgemm test OK
==== Running N=10 with streams ====
Testing sgemm
#### args: ta=0 tb=0 m=1024 n=1024 k=1024 alpha = (0x40000000, 2) beta= (0x40000000, 2)
#### args: lda=1024 ldb=1024 ldc=1024
^^^^ elapsed = 0.00317717 sec GFLOPS=6759.12
@@@@ sgemm test OK
Testing dgemm
#### args: ta=0 tb=0 m=1024 n=1024 k=1024 alpha = (0xbff0000000000000, -1) beta= (0x0000000000000000, 0)
#### args: lda=1024 ldb=1024 ldc=1024
^^^^ elapsed = 0.07991505 sec GFLOPS=268.721
@@@@ dgemm test OK
==== Running N=10 batched ====
Testing sgemm
#### args: ta=0 tb=0 m=1024 n=1024 k=1024 alpha = (0x3f800000, 1) beta= (0xbf800000, -1)
#### args: lda=1024 ldb=1024 ldc=1024
^^^^ elapsed = 0.00302100 sec GFLOPS=7108.51
@@@@ sgemm test OK
Testing dgemm
#### args: ta=0 tb=0 m=1024 n=1024 k=1024 alpha = (0xbff0000000000000, -1) beta= (0x4000000000000000, 2)
#### args: lda=1024 ldb=1024 ldc=1024
^^^^ elapsed = 0.07566714 sec GFLOPS=283.807
@@@@ dgemm test OK
Test Summary
0 error(s)
Jteson
$ ./batchCUBLAS -m1024 -n1024 -k1024
batchCUBLAS Starting...
GPU Device 0: "NVIDIA Tegra X2" with compute capability 6.2
==== Running single kernels ====
Testing sgemm
#### args: ta=0 tb=0 m=1024 n=1024 k=1024 alpha = (0xbf800000, -1) beta= (0x40000000, 2)
#### args: lda=1024 ldb=1024 ldc=1024
^^^^ elapsed = 0.00372291 sec GFLOPS=576.83
@@@@ sgemm test OK
Testing dgemm
#### args: ta=0 tb=0 m=1024 n=1024 k=1024 alpha = (0x0000000000000000, 0) beta= (0x0000000000000000, 0)
#### args: lda=1024 ldb=1024 ldc=1024
^^^^ elapsed = 0.10940003 sec GFLOPS=19.6296
@@@@ dgemm test OK
==== Running N=10 without streams ====
Testing sgemm
#### args: ta=0 tb=0 m=1024 n=1024 k=1024 alpha = (0xbf800000, -1) beta= (0x00000000, 0)
#### args: lda=1024 ldb=1024 ldc=1024
^^^^ elapsed = 0.03462315 sec GFLOPS=620.245
@@@@ sgemm test OK
Testing dgemm
#### args: ta=0 tb=0 m=1024 n=1024 k=1024 alpha = (0xbff0000000000000, -1) beta= (0x0000000000000000, 0)
#### args: lda=1024 ldb=1024 ldc=1024
^^^^ elapsed = 1.09212208 sec GFLOPS=19.6634
@@@@ dgemm test OK
==== Running N=10 with streams ====
Testing sgemm
#### args: ta=0 tb=0 m=1024 n=1024 k=1024 alpha = (0x40000000, 2) beta= (0x40000000, 2)
#### args: lda=1024 ldb=1024 ldc=1024
^^^^ elapsed = 0.03504515 sec GFLOPS=612.776
@@@@ sgemm test OK
Testing dgemm
#### args: ta=0 tb=0 m=1024 n=1024 k=1024 alpha = (0xbff0000000000000, -1) beta= (0x0000000000000000, 0)
#### args: lda=1024 ldb=1024 ldc=1024
^^^^ elapsed = 1.09177494 sec GFLOPS=19.6697
@@@@ dgemm test OK
==== Running N=10 batched ====
Testing sgemm
#### args: ta=0 tb=0 m=1024 n=1024 k=1024 alpha = (0x3f800000, 1) beta= (0xbf800000, -1)
#### args: lda=1024 ldb=1024 ldc=1024
^^^^ elapsed = 0.03766394 sec GFLOPS=570.17
@@@@ sgemm test OK
Testing dgemm
#### args: ta=0 tb=0 m=1024 n=1024 k=1024 alpha = (0xbff0000000000000, -1) beta= (0x4000000000000000, 2)
#### args: lda=1024 ldb=1024 ldc=1024
^^^^ elapsed = 1.09389901 sec GFLOPS=19.6315
@@@@ dgemm test OK
Test Summary
0 error(s)
对比
1080ti 1080 Jetson Tx2
GFLOPS=5654.24 GFLOPS=3526.7 GFLOPS=576.83
GFLOPS=7299.19 GFLOPS=5813.35 GFLOPS=620.245
