How to run llama b bfwith ghs

Lambda labs has half-off GH200s right now to get more people used to the ARM tooling. This means you can maybe actually afford to run the biggest open-source models! The only caveat is that you'll have to occasionally build something from source. Here's how I got llama 405b running with full precision on the GH200s.

Create instances

Llama 405b is about 750GB so you want about 10 96GB GPUS to run it. (The GH200 has pretty good CPU-GPU memory swapping speed -- that's kind of the whole point of the GH200 -- so you can use as few as 3. Time-per-token will be terrible, but total throughput is acceptable, if you're doing batch-processing.) Sign in to lambda labs and create a bunch of GH200 instances. Make sure to give them all the same shared network filesystem.

Save the ip addresses to ~/ips.txt.

Bulk ssh connection helpers

I prefer direct bash & ssh over anything fancy like kubernetes or slurm. It's manageable with some helpers.

# skip fingerprint confirmation
for ip in $(cat ~/ips.txt); do
    echo "doing $ip"
    ssh-keyscan $ip >> ~/.ssh/known_hosts
done

function run_ip() {
    ssh -i ~/.ssh/lambda_id_ed25519 ubuntu@$ip -- stdbuf -oL -eL bash -l -c "$(printf "%q" "$*")"  /dev/null
}
function runall() { ips="$(cat ~/ips.txt)" run_ips "$@"; }
function runrest() { ips="$(tail -n+2 ~/ips.txt)" run_ips "$@"; }

function ssh_k() {
    ip=$(sed -n "$k"p ~/ips.txt)
    ssh -i ~/.ssh/lambda_id_ed25519 ubuntu@$ip
}
alias ssh_head='k=1 ssh_k'

function killall() {
    pkill -ife '.ssh/lambda_id_ed25519'
    sleep 1
    pkill -ife -9 '.ssh/lambda_id_ed25519'
    while [[ -n "$(jobs -p)" ]]; do fg || true; done
}

Set up NFS cache

We'll be putting the python environment and the model weights in the NFS. It will load much faster if we cache it.

# First, check the NFS works.
# runall ln -s my_other_fs_name shared
runhead 'echo world > shared/hello'
runall cat shared/hello

# Install and enable cachefilesd
runall sudo apt-get update
runall sudo apt-get install -y cachefilesd
runall "echo '
RUN=yes
CACHE_TAG=mycache
CACHE_BACKEND=Path=/var/cache/fscache
CACHEFS_RECLAIM=0
' | sudo tee -a /etc/default/cachefilesd"
runall sudo systemctl restart cachefilesd
runall 'sudo journalctl -u cachefilesd | tail -n2'

# Set the "fsc" option on the NFS mount
runhead cat /etc/fstab # should have mount to ~/shared
runall cp /etc/fstab etc-fstab-bak.txt
runall sudo sed -i 's/,proto=tcp,/,proto=tcp,fsc,/g' /etc/fstab
runall cat /etc/fstab

# Remount
runall sudo umount /home/ubuntu/wash2
runall sudo mount /home/ubuntu/wash2
runall cat /proc/fs/nfsfs/volumes # FSC column should say "yes"

# Test cache speedup
runhead dd if=/dev/urandom of=shared/bigfile bs=1M count=8192
runall dd if=shared/bigfile of=/dev/null bs=1M # First one takes 8 seconds
runall dd if=shared/bigfile of=/dev/null bs=1M # Seond takes 0.6 seconds

Create conda environment

Instead of carefully doing the exact same commands on every machine, we can use a conda environment in the NFS and just control it with the head node.

# We'll also use a shared script instead of changing ~/.profile directly.
# Easier to fix mistakes that way.
runhead 'echo ". /opt/miniconda/etc/profile.d/conda.sh" >> shared/common.sh'
runall 'echo "source /home/ubuntu/shared/common.sh" >> ~/.profile'
runall which conda

# Create the environment
runhead 'conda create --prefix ~/shared/311 -y python=3.11'
runhead '~/shared/311/bin/python --version' # double-check that it is executable
runhead 'echo "conda activate ~/shared/311" >> shared/common.sh'
runall which python

Install aphrodite dependencies

Aphrodite is a fork of vllm that starts a bit quicker and has some extra features.
It will run the openai-compatible inference API and the model itself.

You need torch, triton, and flash-attention.
You can get aarch64 torch builds from pytorch.org (you do not want to build it yourself).
The other two you can either build yourself or use the wheel I made.

If you build from source, then you can save a bit of time by running the python setup.py bdist_wheel for triton, flash-attention, and aphrodite in parallel on three different machines. Or you can do them one-by-one on the same machine.

runhead pip install 'numpy



<h4>
  
  
  triton & flash attention from wheels
</h4>



<pre class="brush:php;toolbar:false">runhead pip install 'https://github.com/qpwo/lambda-gh200-llama-405b-tutorial/releases/download/v0.1/triton-3.2.0+git755d4164-cp311-cp311-linux_aarch64.whl'
runhead pip install 'https://github.com/qpwo/lambda-gh200-llama-405b-tutorial/releases/download/v0.1/aphrodite_flash_attn-2.6.1.post2-cp311-cp311-linux_aarch64.whl'

triton from source

k=1 ssh_k # ssh into first machine

pip install -U pip setuptools wheel ninja cmake setuptools_scm
git config --global feature.manyFiles true # faster clones
git clone https://github.com/triton-lang/triton.git ~/shared/triton
cd ~/shared/triton/python
git checkout 755d4164 # 



<h4>
  
  
  flash-attention from source
</h4>



<pre class="brush:php;toolbar:false">k=2 ssh_k # go into second machine

git clone https://github.com/AlpinDale/flash-attention  ~/shared/flash-attention
cd ~/shared/flash-attention
python setup.py bdist_wheel
pip install --no-deps dist/*.whl
python -c 'import aphrodite_flash_attn; import aphrodite_flash_attn_2_cuda; print("flash attn ok")'

Install aphrodite

You can use my wheel or build it yourself.

aphrodite from wheel

# skip fingerprint confirmation
for ip in $(cat ~/ips.txt); do
    echo "doing $ip"
    ssh-keyscan $ip >> ~/.ssh/known_hosts
done

function run_ip() {
    ssh -i ~/.ssh/lambda_id_ed25519 ubuntu@$ip -- stdbuf -oL -eL bash -l -c "$(printf "%q" "$*")"  /dev/null
}
function runall() { ips="$(cat ~/ips.txt)" run_ips "$@"; }
function runrest() { ips="$(tail -n+2 ~/ips.txt)" run_ips "$@"; }

function ssh_k() {
    ip=$(sed -n "$k"p ~/ips.txt)
    ssh -i ~/.ssh/lambda_id_ed25519 ubuntu@$ip
}
alias ssh_head='k=1 ssh_k'

function killall() {
    pkill -ife '.ssh/lambda_id_ed25519'
    sleep 1
    pkill -ife -9 '.ssh/lambda_id_ed25519'
    while [[ -n "$(jobs -p)" ]]; do fg || true; done
}

aphrodite from source

# First, check the NFS works.
# runall ln -s my_other_fs_name shared
runhead 'echo world > shared/hello'
runall cat shared/hello

# Install and enable cachefilesd
runall sudo apt-get update
runall sudo apt-get install -y cachefilesd
runall "echo '
RUN=yes
CACHE_TAG=mycache
CACHE_BACKEND=Path=/var/cache/fscache
CACHEFS_RECLAIM=0
' | sudo tee -a /etc/default/cachefilesd"
runall sudo systemctl restart cachefilesd
runall 'sudo journalctl -u cachefilesd | tail -n2'

# Set the "fsc" option on the NFS mount
runhead cat /etc/fstab # should have mount to ~/shared
runall cp /etc/fstab etc-fstab-bak.txt
runall sudo sed -i 's/,proto=tcp,/,proto=tcp,fsc,/g' /etc/fstab
runall cat /etc/fstab

# Remount
runall sudo umount /home/ubuntu/wash2
runall sudo mount /home/ubuntu/wash2
runall cat /proc/fs/nfsfs/volumes # FSC column should say "yes"

# Test cache speedup
runhead dd if=/dev/urandom of=shared/bigfile bs=1M count=8192
runall dd if=shared/bigfile of=/dev/null bs=1M # First one takes 8 seconds
runall dd if=shared/bigfile of=/dev/null bs=1M # Seond takes 0.6 seconds

Check all installs succeeded

# We'll also use a shared script instead of changing ~/.profile directly.
# Easier to fix mistakes that way.
runhead 'echo ". /opt/miniconda/etc/profile.d/conda.sh" >> shared/common.sh'
runall 'echo "source /home/ubuntu/shared/common.sh" >> ~/.profile'
runall which conda

# Create the environment
runhead 'conda create --prefix ~/shared/311 -y python=3.11'
runhead '~/shared/311/bin/python --version' # double-check that it is executable
runhead 'echo "conda activate ~/shared/311" >> shared/common.sh'
runall which python

Download the weights

Go to https://huggingface.co/meta-llama/Llama-3.1-405B-Instruct and make sure you have the right permissions. The approval usually takes about an hour. Get a token from https://huggingface.co/settings/tokens

runhead pip install 'numpy



<h3>
  
  
  run llama 405b
</h3>

<p>We'll make the servers aware of each other by starting ray.<br>
</p>

<pre class="brush:php;toolbar:false">runhead pip install 'https://github.com/qpwo/lambda-gh200-llama-405b-tutorial/releases/download/v0.1/triton-3.2.0+git755d4164-cp311-cp311-linux_aarch64.whl'
runhead pip install 'https://github.com/qpwo/lambda-gh200-llama-405b-tutorial/releases/download/v0.1/aphrodite_flash_attn-2.6.1.post2-cp311-cp311-linux_aarch64.whl'

We can start aphrodite in one terminal tab:

k=1 ssh_k # ssh into first machine

pip install -U pip setuptools wheel ninja cmake setuptools_scm
git config --global feature.manyFiles true # faster clones
git clone https://github.com/triton-lang/triton.git ~/shared/triton
cd ~/shared/triton/python
git checkout 755d4164 # 



<p>And run a query from the local machine in a second terminal:<br>
</p>

<pre class="brush:php;toolbar:false">k=2 ssh_k # go into second machine

git clone https://github.com/AlpinDale/flash-attention  ~/shared/flash-attention
cd ~/shared/flash-attention
python setup.py bdist_wheel
pip install --no-deps dist/*.whl
python -c 'import aphrodite_flash_attn; import aphrodite_flash_attn_2_cuda; print("flash attn ok")'

runhead pip install 'https://github.com/qpwo/lambda-gh200-llama-405b-tutorial/releases/download/v0.1/aphrodite_engine-0.6.4.post1-cp311-cp311-linux_aarch64.whl'

A good pace for text, but a bit slow for code. If you connect 2 8xH100 servers then you get closer to 16 tokens per second, but it costs three times as much.

further reading

• theoretically you can script instance creation & destruction with the lambda labs API https://cloud.lambdalabs.com/api/v1/docs
• aphrodite docs https://aphrodite.pygmalion.chat/
• vllm docs (api is mostly the same) https://docs.vllm.ai/en/latest/

The above is the detailed content of How to run llama b bfwith ghs. For more information, please follow other related articles on the PHP Chinese website!