Comments (12)
Yes, that should be possible, though I haven't done it. The GPU code just relies on device placement, so if you can construct a TF graph which can name all of the 16 GPUs as different devices, it should work...
from mesh.
I can run the mnist example on a GPU. Does not appear to be utilizing CPU resources. However, when using 4 GPU, only the first device is actually utilized.
Hopefully we can get a developer response on this... I can't see what would need to be modified in mnist.py to make distributed GPU training work.
EDIT: specifying your devices by name ['gpu:0, 'gpu:1', 'gpu:2']
instead of [''] * mesh_size
solves the problem for me
from mesh.
@PSZehnder Does mesh tensorflow supports multi node training ( i.e. each node has #x GPUs attached to it)?
I'm using 2 nodes each with 8 GPUs and would like to train on the entire (2 nodes *8 gpus )=16 GPUs.
How do I configure mesh tensorflow to train in a multi node setup?
from mesh.
@nshazeer Does mesh tensorflow supports multi node training ( i.e. each node has #x GPUs attached to it)?
I'm using 2 nodes each with 8 GPUs and would like to train on the entire (2 nodes *8 gpus )=16 GPUs.
How do I configure mesh tensorflow to train in a multi node setup?
from mesh.
@nshazeer , Thanks for your reply. If I can make the 16 GPUs visible ,How the data loading will be done in a 2 node * 8 GPUs ?
Will the data be loaded through 1 CPU in node0 ( where I run the script, so 1 CPU sends data to 16 GPUs) or the data loading will be done from the 2 cpus ( node0 and node1), so each CPU sends the data which is relevant to the 8 GPUs it its connected to. ?
from mesh.
Facing the same issue.
from mesh.
facing same issue. can someone share some answers for this?
from mesh.
Also seeing this issue. Monitoring GPU usage shows that only one GPU is being utilized when running BERT.
from mesh.
The current MNIST example is just using a single GPU in AMD/RocM platforms.
from mesh.
@nict-wisdom do you have a snippet showing how you used the ProfilerHook
, I am a bit struggling with it atm.
from mesh.
Met the same problem, anyone on this team can reply this issue?
from mesh.
We are also facing the same issue. Any help in this context will be highly appreciated.
from mesh.
Related Issues (20)
- Tensorflow Mesh needs documentation. Will this be provided anytime soon? HOT 1
- the `model_executor.py` example is broken
- OpenNMT-tf
- [Wrong Code Comments] In moe.py, there are two wrong code comments
- performing the opposite of mtf.lowering HOT 1
- How to assign values to specific slice of a data block on a specific GPU?
- How to use tf.contrib.opt.ScipyOptimizerInterface or tfp.optimizer.lbfgs_minimize with MeshTF ?
- [MOE-transformer] How do you build static graph of MOE-Model?
- Ability to add Custom Tensorflow Hooks
- Beam search
- How to freeze embedding layers
- Mesh-tf model conversion to onnx? HOT 2
- About the mixture of expert model
- mask_1_flat and mask_2_flat applied to gates twice?
- Getting "NanLossDuringTrainingError: NaN loss during training."
- When running BERT on GPU: Resource exhausted: failed to allocate memory HOT 1
- Does load-balanced loss help the loss converge?
- AttributeError: module 'tensorflow.python.framework.ops' has no attribute 'register_tensor_conversion_function' HOT 4
- Optimizer momentums not properly populated training model with DTensors HOT 1
- Error while importing Meshtensorflow
Recommend Projects
-
React
A declarative, efficient, and flexible JavaScript library for building user interfaces.
-
Vue.js
🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
-
Typescript
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
-
TensorFlow
An Open Source Machine Learning Framework for Everyone
-
Django
The Web framework for perfectionists with deadlines.
-
Laravel
A PHP framework for web artisans
-
D3
Bring data to life with SVG, Canvas and HTML. 📊📈🎉
-
Recommend Topics
-
javascript
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
-
web
Some thing interesting about web. New door for the world.
-
server
A server is a program made to process requests and deliver data to clients.
-
Machine learning
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
-
Visualization
Some thing interesting about visualization, use data art
-
Game
Some thing interesting about game, make everyone happy.
Recommend Org
-
Facebook
We are working to build community through open source technology. NB: members must have two-factor auth.
-
Microsoft
Open source projects and samples from Microsoft.
-
Google
Google ❤️ Open Source for everyone.
-
Alibaba
Alibaba Open Source for everyone
-
D3
Data-Driven Documents codes.
-
Tencent
China tencent open source team.
from mesh.