Downloaded 241 times
![5
// Vector sum in C
void vector_add(int n, const float* a, const float* b, float* c)
{
for (int idx = 0; idx < n; ++idx)
c[idx] = a[idx] + b[idx];
}
// Vector sum in CUDA
__global__ void vector_add(int n, const float* a, const float* b, float* c)
{
int idx = blockIdx.x * blockDim.x + threadIdx.x;
if (idx < n)
c[idx] = a[idx] + b[idx];
}
CUDA C++ PROGRAMMING](https://image.slidesharecdn.com/146387dockercon16-160622172714/75/Using-Docker-for-GPU-Accelerated-Applications-5-2048.jpg)
![5
// Vector sum in C
void vector_add(int n, const float* a, const float* b, float* c)
{
for (int idx = 0; idx < n; ++idx)
c[idx] = a[idx] + b[idx];
}
// Vector sum in CUDA
__global__ void vector_add(int n, const float* a, const float* b, float* c)
{
int idx = blockIdx.x * blockDim.x + threadIdx.x;
if (idx < n)
c[idx] = a[idx] + b[idx];
}
CUDA C++ PROGRAMMING](https://crownmelresort.com/image.slidesharecdn.com/146387dockercon16-160622172714/75/Using-Docker-for-GPU-Accelerated-Applications-5-2048.jpg)
Uploaded byNVIDIA
Using Docker for GPU Accelerated Applications
The document discusses the use of Docker for GPU-accelerated applications, highlighting the challenges and solutions in GPU computing, especially within NVIDIA's ecosystem. It covers applications in various fields such as deep learning, medical imaging, and autonomous machines, while stressing the importance of proper driver management and container support. The presentation emphasizes how Docker facilitates the workflow for developing and deploying GPU applications effectively.
More Related Content
Similar to Using Docker for GPU Accelerated Applications
Using Docker for GPU Accelerated Applications
- 1. Felix Abecassis, SystemsSoftware Engineer Jonathan Calmels, Systems Software Engineer USING DOCKER FOR GPU ACCELERATED APPLICATIONS
- 2. 2 AGENDA GPU COMPUTING CUDA Ecosystem Applications NVIDIA DOCKER Challenges OurSolution DEMOS GPU Isolation Machine Learning Remote Deployment
- 3.
- 4. 4 HETEROGENEOUS COMPUTING GPU CPU + Optimizedfor parallel tasks Optimized for serial tasks
- 5. 5 // Vector sumin C void vector_add(int n, const float* a, const float* b, float* c) { for (int idx = 0; idx < n; ++idx) c[idx] = a[idx] + b[idx]; } // Vector sum in CUDA __global__ void vector_add(int n, const float* a, const float* b, float* c) { int idx = blockIdx.x * blockDim.x + threadIdx.x; if (idx < n) c[idx] = a[idx] + b[idx]; } CUDA C++ PROGRAMMING
- 6.
- 7.
- 8. 8 APPLICATIONS: DEEP LEARNING INTERNET& CLOUD Image Classification Speech Recognition Language Translation Language Processing Sentiment Analysis Recommendation MEDICINE & BIOLOGY Cancer Cell Detection Diabetic Grading Drug Discovery MEDIA & ENTERTAINMENT Video Captioning Video Search Real Time Translation SECURITY & DEFENSE Face Detection Video Surveillance Satellite Imagery AUTONOMOUS MACHINES Pedestrian Detection Lane Tracking Recognize Traffic Sign
- 9. 9 GPU-ACCELERATED DEEP LEARNING TENSORFLOWCNTK TORCH CAFFE WATSON CHAINER THEANO MATCONVNET
- 10.
- 11. 11 CHALLENGES A Typical Cluster Ubuntu14.04 Drivers 352 4x Maxwell CentOS 7 Drivers 346 4x Kepler Ubuntu 16.04 Drivers 362 8x Pascal CUDA 7.5 CUDA 7.0 cuDNN 3 CUDA 7.5 cuDNN 4 CUDA 8.0 Patches
- 12. 12 PACKAGING DRIVER FILES? FROMubuntu:14.04 RUN apt-get update && apt-get install --no-install-recommends -y gcc make libc-dev wget RUN wget http://us.download.nvidia.com/XFree86/Linux- x86_64/361.42/NVIDIA- Linux-x86_64-361.42.run RUN sh NVIDIA-Linux-x86_64-361.42.run --silent --no-kernel-module Never install the driver in the Dockerfile, not portable!
- 13. 13 BRINGING GPU SUPPORTTO DOCKER
- 14. 14 INTERNALS Docker Daemon NVIDIA DockerPlugin NVIDIA Drivers NVIDIA Docker Docker Registry 1 2 3 4 5 6 HTTP + SSH HTTP + UNIX LDCACHE + NVML HTTPS CUDA 8.0 CAFFE CUDA 7.5 TENSORFLOW HTTPS HTTPSDocker CLI
- 15. 15 DOCKERHUB IMAGES CUDA 7.0 runtime CUDA7.5 runtime CUDA 8.0 runtime cuDNN v4 runtime CUDA 7.0 devel cuDNN v4 runtime CUDA 7.5 devel cuDNN v5 runtime CUDA 8.0 devel NVIDIA/Caffe 0.14.2 cuDNN v4 devel cuDNN v4 devel NVIDIA/Caffe 0.14.5 cuDNN v5 devel DIGITS 3.0 CNTK BVLC/Caffe TensorFlow DIGITS 3.3 Ubuntu 14.04 Ubuntu 16.04
- 16. 16 GPU APPLICATIONS WORKFLOW Research/DevelopTest/Package Deploy Video Transcoding Image Processing Deep Learning HPC Visualization Container Based Applications GPU-Accelerated Data Center
- 17.
- 18.