Tesla

Subscribe
SPEED UP TRAINING WITH GPU-ACCELERATED TENSORFLOW
The fastest, easiest way to get started with deep learning on GPUs
Speed Up Training with GPU-Accelerated TensorFlow

Expected Performance Results

Each of the models described in the previous section output either an execution time/minibatch or an average speed in examples/second, which can be converted to the time/minibatch by dividing into the batch size.

The Inception v3 model for image training supports training on single and multiple GPUs. The graphs below show the expected performance on systems with a CPU and 1, 2, and 4 Tesla GPUs per node.

 

Image-Classification Training Performance

 

Inception v3 Training Scalable Performance on Multi-GPUs Per Node

 

CREATE YOUR IDEAL SOLUTION.