-
Notifications
You must be signed in to change notification settings - Fork 1
Project 1
Since the Post Processing, Model Execution, and Post-processing files have the same installation instructions I have mentioned the installation instructions for the 3 microservices here itself for your simplicity. For all the remaining microservices the instructions are mentioned separately.
Make sure that you have python 3.7 or above version installed. We recommend having version 3.7.3 installed. It is suggested to run this repository in a virtual environment so that it doesn't interfere with your local system's default installations. Please refer to the guides for creating a virtual environment in Windows, Ubuntu and macOS.
We recommend using python3 and pip3 whenever you are running the code to use the Python 3 version in case there is more than one version of python installed.
Do add all these libraries in your virtual environment or the conda environment. But you will have to use pip for installing pytemperature. Anaconda comes with most of the libraries mentioned below so we strongly encourage you to use Anaconda. Libraries along with the versions are mentioned above for -
numpy==3.0.3
matplotlib==3.0.3
pika==1.1.0
pytemperature==1.0
metpy==0.11.0
boto3==2.49.0
botocore==1.14.9
We recommend installing these versions as mentioned above which you can easily do by replacing the libraryname with libraryname==version( just like mentioned above ) for each of the commands for installation as mentioned below. But the default most recent stable versions which will be downloaded when you do the steps as mentioned below should work fine as they did for us. We have mentioned the versions for incase some incompatibility among some versions which get updated in the future.
Conda using terminal commands
conda install -c anaconda numpy
conda install -c conda-forge matplotlib
conda install -c conda-forge pika
conda install -c conda-forge metpy
conda install -c anaconda boto3
conda install -c anaconda botocore
Pip using terminal commands
pip3 install numpy
pip3 install matplotlib
pip3 install pika
pip3 install pytemperature
pip3 install metpy
pip3 install boto3
pip3 install botocore
Installing Erlang( RabbitMQ dependency ) for Windows, Ubuntu, and macOS.
Installing RabbitMQ for Windows , Ubuntu and macOS.
To access NEXRAD files from the AWS servers, we need to set up an AWS account here.
Don't worry it's free. Once an account is set up, we need to make use of the authorization key provided under the Project Name > My Security Credentials tab. Next click on the link "Access Keys". Click the button "Create New Access Key". An access key ID and the secret access key will be displayed. Set up the AWS configuration for your machine following the instructions here .
Add your AWS Access Key ID and AWS Secret Key to the post_processing.py file.
Make sure that all the above dependencies are installed and functional. The installation guide above explains all the things including the activation of servers and the environment. RabbitMQ should be enabled and running and you don't need to run that in a separate terminal. You can enable and exit that terminal. The server will keep on running in the background.
- Clone the git repository in your system
git clone https://github.com/airavata-courses/IU-Witcher-2020.git
- Checkout
release-1.0
branch
git checkout release-1.0
All microservices will be running in separate terminals so each microservice will need a separate environment. Here are the mentioned environments which need to be for each microservice which will all be running in the separate terminals.
- Data Retrieval needs enabling of virtual/conda environment in a separate terminal.
- Model Execution needs enabling of virtual/conda environment in a separate terminal.
- Post processing needs enabling of virtual/conda environment in a separate terminal.
Move to the current working directory in all the terminals -
cd IU-Witcher-2020/
After this, the common installation and enabling of the environment( for the 3 microservices ) is done. Now you can move to separate pages for installation of other microservices and running in the order mentioned below.
Run all the microservices below in the respective order for the architecture to fully function.
Run RabbitMQ