RiseML Blog
Company blog
Comparing Google’s TPUv2 against Nvidia’s V100 on ResNet-50
Comparing Google’s TPUv2 against Nvidia’s V100 on ResNet-50
Google recently added the Tensor Processing Unit v2 (TPUv2), a custom-developed microchip to accelerate deep learning, to its cloud…
Training ImageNet on a TPU in 12.5 hours with GKE and RiseML
Training ImageNet on a TPU in 12.5 hours with GKE and RiseML
Google’s Tensor Processing Unit (TPU), a custom-developed accelerator for deep learning, offers a fast and cost-efficient alternative to…
RiseML supports distributed deep learning with Horovod and TensorFlow
RiseML supports distributed deep learning with Horovod and TensorFlow
Most deep learning models are trained on a single node, because parallelisation is difficult due to strong computational dependencies…
Benchmarking Google’s new TPUv2
Benchmarking Google’s new TPUv2
NOTE: We published a follow-up article with more up-to-date benchmark results here.
Accelerating I/O bound deep learning on shared storage
Accelerating I/O bound deep learning on shared storage
When training a neural network, one typically strives to make the GPU the bottleneck. All data should be read from disk, pre-processed, and…