Google Cloud Developer Advocate Nikita Namjoshi introduces how distributed training models can dramatically reduce machine learning training times, explains how to make use of multiple GPUs with Data Parallelism vs Model Parallelism, and explores Synchronous vs Asynchronous Data Parallelism.

Mesh TensorFlow → https://goo.gle/3sFPrHw
Distributed Training with Keras tutorial → https://goo.gle/3FE6QEa
GCP Reduction Server Blog → https://goo.gle/3EEznYB
Multi Worker Mirrored Strategy tutorial → https://goo.gle/3JkQT7Y
Parameter Server Strategy tutorial → https://goo.gle/2Zz3UrW
Distributed training on GCP Demo → https://goo.gle/3pABNDE

Chapters:
0:00 Introduction
00:17 Agenda
00:37 Why distributed training?
1:49 Data Parallelism vs Model Parallelism
6:05 Synchronous Data Parallelism
18:20 Asynchronous Data Parallelism
23:41 Thank you for watching

Watch more ML Tech Talks → https://goo.gle/mltechtalks
Subscribe to TensorFlow → https://goo.gle/TensorFlow


#TensorFlow #MachineLearning #ML


product: TensorFlow General;