A fully functioning, flexibly deployable application for visualizations of personal financial data.
Use a package manager and install docker-machine
and kubectl
and minikube
and run:
./deploy_locally_k8s.ps1
This is a bit deprecated (for now), but you can also run via:
docker-compose up --build -d
Install MongoDB, NodeJS and the required NodeJS packages. Then, start a local MongoDB instance bound to port 27017. Finally, run:
node server.js --database_info 127.0.0.1:27017
This is a bit deprecated, due to more interest in GKE as a deployment platform. Tweaking is needed.
Use an AWS Linux machine, clone this repo, and then run:
cd Midas/cloud/aws && ./deploy_on_aws_k8s.sh
Ensure billing is enabled for GCP.
Use the provided Ansible to spin up the cluster:
cd Midas/cloud/gcp/ansible/playbooks
ansible-playbook -i hosts midas.yaml --tags="midas_gke_cluster"
Connect to the default cluster via Google Cloud shell, then run:
git clone https://github.com/AlexDHoffer/Midas.git
cd Midas/cloud/gcp
./deploy_on_gke.sh
Step 2: Spin up Compute Engine (VM) that gathers data and publishes it to the database through the external service endpoint.
Use the provided Ansible to provision the VM:
cd Midas/cloud/gcp/ansible/playbooks
ansible-playbook -i hosts midas.yaml --tags="midas_data_vm"
External service endpoint portion TBD.
TBD