Skip to content

Commit

Permalink
Add example for distributed pytorch
Browse files Browse the repository at this point in the history
  • Loading branch information
Michaelvll committed Dec 12, 2024
1 parent bc4a42e commit 179520e
Show file tree
Hide file tree
Showing 3 changed files with 81 additions and 0 deletions.
31 changes: 31 additions & 0 deletions examples/distributed-pytorch/README.md
Original file line number Diff line number Diff line change
@@ -0,0 +1,31 @@
# Distributed Training with PyTorch

This example demonstrates how to run distributed training with PyTorch using SkyPilot.

**The example is based on [PyTorch's official minGPT example](https://github.com/pytorch/examples/tree/main/distributed/minGPT-ddp)**

## Using distributed training


The following command spawn 2 nodes with 1 L4 GPU each.

`sky launch -c train.yaml`

In the [train.yaml](./train.yaml), we use `torchrun` to launch the training and set the arguments for distributed training using environment variables provided by SkyPilot.

```yaml
run: |
cd examples/mingpt
MASTER_ADDR=$(echo "$SKYPILOT_NODE_IPS" | head -n1)
torchrun \
--nnodes=$SKYPILOT_NUM_NODES \
--nproc_per_node=$SKYPILOT_NUM_GPUS_PER_NODE \
--master_addr=$MASTER_ADDR \
--master_port=8008 \
--node_rank=${SKYPILOT_NODE_RANK} \
main.py
```
## Using `rdvz`
25 changes: 25 additions & 0 deletions examples/distributed-pytorch/train-rdzv.yaml
Original file line number Diff line number Diff line change
@@ -0,0 +1,25 @@
name: minGPT-ddp-rdzv

resources:
cpus: 4+
accelerators: L4

num_nodes: 2

setup: |
git clone --depth 1 https://github.com/pytorch/examples || true
cd examples
git filter-branch --prune-empty --subdirectory-filter distributed/minGPT-ddp
# SkyPilot's default image on AWS/GCP has CUDA 11.6 (Azure 11.5).
uv pip install -r requirements.txt "numpy<2" "torch==1.12.1+cu113" --extra-index-url https://download.pytorch.org/whl/cu113
run: |
cd examples/mingpt
MASTER_ADDR=$(echo "$SKYPILOT_NODE_IPS" | head -n1)
torchrun \
--nnodes=$SKYPILOT_NUM_NODES \
--nproc_per_node=$SKYPILOT_NUM_GPUS_PER_NODE \
--rdzv_backend=c10d \
--rdzv_endpoint=$MASTER_ADDR:8008 \
--rdzv_id $RANDOM \
main.py
25 changes: 25 additions & 0 deletions examples/distributed-pytorch/train.yaml
Original file line number Diff line number Diff line change
@@ -0,0 +1,25 @@
name: minGPT-ddp

resources:
cpus: 4+
accelerators: L4

num_nodes: 2

setup: |
git clone --depth 1 https://github.com/pytorch/examples || true
cd examples
git filter-branch --prune-empty --subdirectory-filter distributed/minGPT-ddp
# SkyPilot's default image on AWS/GCP has CUDA 11.6 (Azure 11.5).
uv pip install -r requirements.txt "numpy<2" "torch==1.12.1+cu113" --extra-index-url https://download.pytorch.org/whl/cu113
run: |
cd examples/mingpt
MASTER_ADDR=$(echo "$SKYPILOT_NODE_IPS" | head -n1)
torchrun \
--nnodes=$SKYPILOT_NUM_NODES \
--nproc_per_node=$SKYPILOT_NUM_GPUS_PER_NODE \
--master_addr=$MASTER_ADDR \
--master_port=8008 \
--node_rank=${SKYPILOT_NODE_RANK} \
main.py

0 comments on commit 179520e

Please sign in to comment.