Skip to content

Joshuaalbert/DSA2000-Cal

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

DSA2000 Calibration and Forward Modelling

This repo contains two components:

  1. The dsa2000_cal package, which implements a library for carrying out forward modelling and calibration of radio interferometers. The package is predominantly written in JAX, using astropy to represent physical quantities. The package is designed to be used with the DSA2000 radio interferometer, but can be used with others, e.g. it is used in the OVRO-LWA project. See the package README for more details.
  2. Workflows which contains scripts and Dockerfiles for running distributed workflows utilising the dsa2000_cal package. This includes setting up a distributed streaming calibration with low-level interface exposed, sharing data between containers through sharded memory and common namespace; streaming forward modelling to simulate observations; as well as non-streaing variants of the above two workflows.

Deployment

To run any workflows, you need to have docker installed, if not already the case make sure your user is in the docker group ( have an admin run this for you sudo usermod -aG docker $USER). Then, you can run the following command in a given workflow folder:

./run.sh

We use Git LFS to store the data, so you will need to have that installed. You can install it with the following commands:

sudo apt-get install git-lfs
# Then inside repo directory
git lfs install
# To track a large file use
git lfs track "path/to/file"

To run with GPU support, you need to have the Nvidia driver and nvidia-docker installed.

If you have the Nvidia driver installed then you should be able to run:

nvidia-smi

and see something like:

+-----------------------------------------------------------------------------+
| NVIDIA-SMI 525.125.06   Driver Version: 525.125.06   CUDA Version: 12.0     |
|-------------------------------+----------------------+----------------------+
| GPU  Name        Persistence-M| Bus-Id        Disp.A | Volatile Uncorr. ECC |
| Fan  Temp  Perf  Pwr:Usage/Cap|         Memory-Usage | GPU-Util  Compute M. |
|                               |                      |               MIG M. |
|===============================+======================+======================|
...

On Ubuntu, you can do this with the following commands:

distribution=$(. /etc/os-release;echo $ID$VERSION_ID)
curl -s -L https://nvidia.github.io/nvidia-docker/gpgkey | sudo apt-key add -
curl -s -L https://nvidia.github.io/nvidia-docker/$distribution/nvidia-docker.list | sudo tee /etc/apt/sources.list.d/nvidia-docker.list
sudo apt-get update
sudo apt-get install -y nvidia-container-toolkit
sudo systemctl restart docker

On CentOS/RHEL/Fedora, you can do this with the following commands:

distribution=$(. /etc/os-release; echo $ID$VERSION_ID)
sudo rpm --import https://nvidia.github.io/nvidia-docker/gpgkey
sudo curl -s -L https://nvidia.github.io/nvidia-docker/$distribution/nvidia-docker.repo -o /etc/yum.repos.d/nvidia-docker.repo
sudo yum install -y nvidia-container-toolkit
sudo systemctl restart docker\

TODO: set polarisations in Beam

About

DSA-2000 Calibration and Forward Modelling

Resources

License

Stars

Watchers

Forks