diff --git a/SWARM-Without-Prediction/README.md b/SWARM-Without-Prediction/README.md index 14c25f0674811e40b439b32126b76930338c0573..2279d9f603457d252a5738ab177140bc937deee1 100644 --- a/SWARM-Without-Prediction/README.md +++ b/SWARM-Without-Prediction/README.md @@ -15,6 +15,9 @@ The algorithm will later be used as a backbone for distributed federated learnin 1. **Create Virtual Machines**: - Create *n* EC2 **Ubuntu** instances on Amazon. - Save the private IP addresses in the file `instance_privateIPs.txt` + - **Set up the rules according to the image below** + +  2. **Install Docker and Docker Compose**: - Use the provided script `install-docker.sh` to install Docker and Docker Compose on each VM. **Note:** This script is for installing Docker and Docker Compose on Ubuntu. @@ -64,13 +67,18 @@ To deploy your distributed algorithm, you need to create and build a Docker imag ## Deployment These deployment steps should be performed on the master node of the Docker Swarm: +Use ssh command to connect to the master node using your AWS credentials (keypair.pem file): + + ssh -i <your_keypair.pem> ubuntu@master-node-ip + 1. **Create Neighbors Configuration**: - Create a `neighbors.json` file ([view example](neighbors.json)) that specifies the neighbors for each node in your distributed algorithm network. 2. **Create Docker Config**: - - Use the `docker config create` command to create a Docker config from your `neighbors.json` file: + - In this step, you should create a configuration file that represents `neighbors.json` + - Use the `docker config create` command to create this configuration file: ```bash - docker config create <config_name> neighbors.json + docker config create <configuration file> neighbors.json ``` 3. **Generate Docker Compose File**: diff --git a/SWARM-Without-Prediction/rules.png b/SWARM-Without-Prediction/rules.png new file mode 100644 index 0000000000000000000000000000000000000000..80c14cff98c3b30ddd319229632658775c5d921b Binary files /dev/null and b/SWARM-Without-Prediction/rules.png differ