CUDA Tegra、GeForce、Quadro、Tesla的区别
2017-10-30 17:07
375 查看
部分内容引自:http://blog.csdn.net/ziv555/article/details/51279424
Tegra系列产品是转为移动和嵌入式设备而设计的,如平板电脑和手机,典型的如Tegra K1,Tegra X1,TegraX2等(为嵌入式芯片,功耗低,gpu和cpu芯片在同一块硅片上)
GeForce面向图像用户,典型的如NVIDIA TITAN Xp、GeForce GTX 1080等(通常用于桌面显示,价格亲民,普通大众使用的较多,也可进行科学计算,只是精确度稍低)
Quadro用于专业绘图设计,典型的如Quadro 6000、Quadro FX 5800(计算能力均在2.0及以下)(Quadro支持高速OpenGL渲染)
Tesla用于大规模的并行计算,典型的如Tesla P100、Tesla M60、Tesla K80(Tesla不支持OpenGL。Tesla专为企业部署而设计,拥有更高级别的专业显存,专为处理计算类应用程序而设计。)
Tesla和Geforce的区别(因此Geforce通常比较便宜,Tesla通常很贵)
1:T内部和外部RAM都有ECC保护,而G无保护,ECC的保护,大大降低了内存错误的发生几率;
2:T全部通过0错误测试,保证计算结果的正确性,而G仅针对图形应用的测试,使用T计算得出结果是准确的,可信的。而使用G计算出的结果往往出现错误,经常同一个程序跑几遍,得出的结果却不一样,因此对于严谨的科学计算来说,G的这种表现是致命的;
3:T是Nvidia原厂生产,G是第三方厂商,相比质量更好;
Tegra系列猜测由于成本、体积、需求点等问题同样没有ECC内存保护(查看Tegra X1确实没有):
ECC是否支持查看方式在官方示例代码中有,目录为NVIDIA_CUDA-8.0_Samples/1_Utilities/deviceQuery,其中还含有其他一些性能的获取和显示。
如TX1的运行结果为:
./deviceQuery/deviceQuery Starting...
CUDA Device Query (Runtime API) version (CUDART static linking)
Detected 1 CUDA Capable device(s)
Device 0: "NVIDIA Tegra X1"
CUDA Driver Version / Runtime Version 8.0 / 8.0
CUDA Capability Major/Minor version number: 5.3
Total amount of global memory: 3995 MBytes (4188774400 bytes)
( 2) Multiprocessors, (128) CUDA Cores/MP: 256 CUDA Cores
GPU Max Clock rate: 72 MHz (0.07 GHz)
Memory Clock rate: 13 Mhz
Memory Bus Width: 64-bit
L2 Cache Size: 262144 bytes
Maximum Texture Dimension Size (x,y,z) 1D=(65536), 2D=(65536, 65536), 3D=(4096, 4096, 4096)
Maximum Layered 1D Texture Size, (num) layers 1D=(16384), 2048 layers
Maximum Layered 2D Texture Size, (num) layers 2D=(16384, 16384), 2048 layers
Total amount of constant memory: 65536 bytes
Total amount of shared memory per block: 49152 bytes
Total number of registers available per block: 32768
Warp size: 32
Maximum number of threads per multiprocessor: 2048
Maximum number of threads per block: 1024
Max dimension size of a thread block (x,y,z): (1024, 1024, 64)
Max dimension size of a grid
4000
size (x,y,z): (2147483647, 65535, 65535)
Maximum memory pitch: 2147483647 bytes
Texture alignment: 512 bytes
Concurrent copy and kernel execution: Yes with 1 copy engine(s)
Run time limit on kernels: Yes
Integrated GPU sharing Host Memory: Yes
Support host page-locked memory mapping: Yes
Alignment requirement for Surfaces: Yes
Device has ECC support: Disabled
Device supports Unified Addressing (UVA): Yes
Device PCI Domain ID / Bus ID / location ID: 0 / 0 / 0
Compute Mode:
< Default (multiple host threads can use ::cudaSetDevice() with device simultaneously) >
deviceQuery, CUDA Driver = CUDART, CUDA Driver Version = 8.0, CUDA Runtime Version = 8.0, NumDevs = 1, Device0 = NVIDIA Tegra X1
Result = PASS
获取该内容是否支持的函数和部分代码如下
cudaSetDevice(dev);
cudaDeviceProp deviceProp;
cudaGetDeviceProperties(&deviceProp, dev);
printf(" Device has ECC support: %s\n", deviceProp.ECCEnabled ? "Enabled" : "Disabled");
Tegra系列产品是转为移动和嵌入式设备而设计的,如平板电脑和手机,典型的如Tegra K1,Tegra X1,TegraX2等(为嵌入式芯片,功耗低,gpu和cpu芯片在同一块硅片上)
GeForce面向图像用户,典型的如NVIDIA TITAN Xp、GeForce GTX 1080等(通常用于桌面显示,价格亲民,普通大众使用的较多,也可进行科学计算,只是精确度稍低)
Quadro用于专业绘图设计,典型的如Quadro 6000、Quadro FX 5800(计算能力均在2.0及以下)(Quadro支持高速OpenGL渲染)
Tesla用于大规模的并行计算,典型的如Tesla P100、Tesla M60、Tesla K80(Tesla不支持OpenGL。Tesla专为企业部署而设计,拥有更高级别的专业显存,专为处理计算类应用程序而设计。)
Tesla和Geforce的区别(因此Geforce通常比较便宜,Tesla通常很贵)
1:T内部和外部RAM都有ECC保护,而G无保护,ECC的保护,大大降低了内存错误的发生几率;
2:T全部通过0错误测试,保证计算结果的正确性,而G仅针对图形应用的测试,使用T计算得出结果是准确的,可信的。而使用G计算出的结果往往出现错误,经常同一个程序跑几遍,得出的结果却不一样,因此对于严谨的科学计算来说,G的这种表现是致命的;
3:T是Nvidia原厂生产,G是第三方厂商,相比质量更好;
Tegra系列猜测由于成本、体积、需求点等问题同样没有ECC内存保护(查看Tegra X1确实没有):
ECC是否支持查看方式在官方示例代码中有,目录为NVIDIA_CUDA-8.0_Samples/1_Utilities/deviceQuery,其中还含有其他一些性能的获取和显示。
如TX1的运行结果为:
./deviceQuery/deviceQuery Starting...
CUDA Device Query (Runtime API) version (CUDART static linking)
Detected 1 CUDA Capable device(s)
Device 0: "NVIDIA Tegra X1"
CUDA Driver Version / Runtime Version 8.0 / 8.0
CUDA Capability Major/Minor version number: 5.3
Total amount of global memory: 3995 MBytes (4188774400 bytes)
( 2) Multiprocessors, (128) CUDA Cores/MP: 256 CUDA Cores
GPU Max Clock rate: 72 MHz (0.07 GHz)
Memory Clock rate: 13 Mhz
Memory Bus Width: 64-bit
L2 Cache Size: 262144 bytes
Maximum Texture Dimension Size (x,y,z) 1D=(65536), 2D=(65536, 65536), 3D=(4096, 4096, 4096)
Maximum Layered 1D Texture Size, (num) layers 1D=(16384), 2048 layers
Maximum Layered 2D Texture Size, (num) layers 2D=(16384, 16384), 2048 layers
Total amount of constant memory: 65536 bytes
Total amount of shared memory per block: 49152 bytes
Total number of registers available per block: 32768
Warp size: 32
Maximum number of threads per multiprocessor: 2048
Maximum number of threads per block: 1024
Max dimension size of a thread block (x,y,z): (1024, 1024, 64)
Max dimension size of a grid
4000
size (x,y,z): (2147483647, 65535, 65535)
Maximum memory pitch: 2147483647 bytes
Texture alignment: 512 bytes
Concurrent copy and kernel execution: Yes with 1 copy engine(s)
Run time limit on kernels: Yes
Integrated GPU sharing Host Memory: Yes
Support host page-locked memory mapping: Yes
Alignment requirement for Surfaces: Yes
Device has ECC support: Disabled
Device supports Unified Addressing (UVA): Yes
Device PCI Domain ID / Bus ID / location ID: 0 / 0 / 0
Compute Mode:
< Default (multiple host threads can use ::cudaSetDevice() with device simultaneously) >
deviceQuery, CUDA Driver = CUDART, CUDA Driver Version = 8.0, CUDA Runtime Version = 8.0, NumDevs = 1, Device0 = NVIDIA Tegra X1
Result = PASS
获取该内容是否支持的函数和部分代码如下
cudaSetDevice(dev);
cudaDeviceProp deviceProp;
cudaGetDeviceProperties(&deviceProp, dev);
printf(" Device has ECC support: %s\n", deviceProp.ECCEnabled ? "Enabled" : "Disabled");
相关文章推荐
- Nvidia Tesla和Quadro、GeForce的区别
- NVIDIA Tesla/Quadro和GeForce GPU比较
- NVIDIA之Tesla、GeForce和Quadro系列GPU对比
- 几款科学计算显卡对比(GTX Titan X、GTX 980、Tesla K40 K80 及quadro K4200) 科学计算显卡的两个主要性能指标: 1、CUDA compute capabili
- 截至2012年5月23日19点58分支持CUDA的NVIDIA的GPU列表(Tesla,Quadro,NVS)
- 截至2012年5月23日19点58分支持CUDA的NVIDIA的GPU列表(Tesla,Quadro,NVS)
- nvidia显卡系列:geforce/quadro/tesla
- Quadro和Geforce系列的区别
- "GeForce GTX 850M" 的CUDA参数
- Ubuntu16.04 + cuda 8.0 + Geforce GT630 OEM 配置
- 拷贝global memory,cudaMemcpyToSymbol 和cudaMemcpy函数是否有区别
- Cuda 学习教程二:GeForce GTX 950M 显卡参数
- 服务器tesla配置驱动安装cuda
- ubuntu16.04 安装GeForce1080驱动+CUDA8.0
- CUDA和OpenCL有什么区别?
- 几款科学计算显卡对比(GTX Titan X、GTX 980、Tesla K40 K80 及quadro K4200)
- CUDA学习笔记之Tesla图形与计算架构和通用计算模型
- CUDA 与 OpenCL 区别
- GPU CUDA编程中threadIdx, blockIdx, blockDim, gridDim之间的区别与联系
- 深度学习主机环境配置: Ubuntu16.04 + GeForce GTX 1070 + CUDA8.0 + cuDNN5.1 + TensorFlow