# BioSimSpace Website & Notebook Installation

This directory contains all of the files needed to set up and configure the BioSimSpace website and jupyter notebooks (including building all custom docker images)

Setting up the initial kubernetes cluster

This is all build in Azure

Create the cluster

$ az group create --name biosimspace --location eastus
$ az aks create --resource-group biosimspace --name biosimspace --node-count 6 --generate-ssh-keys --node-vm-size Standard_D4_v3
$ az aks get-credentials --resource-group biosimspace --name biosimspace
$ kubectl get nodes

Creates a cluster called 'biosimspace' in the 'biosimspace' workgroup, and then makes sure that I can connect using kubectl

This sets up six D4v3 nodes. Each node has 4 fast cores, 16 GB RAM and 100 GB disk. These cost $0.24 per node hour, so $34.56 per day, $1036.80 per month and $12614.40 per year. There is the option for reserving these for 3 years. This would bring the cost down to $0.106 per node hour, so $15.26 per day, $457.90 per month or $5571.36 per year

(Using UK pricing, £0.179 per hour, £25.776 per day, £773.28 per month and £9408.24 per year. There is a 3 year option that brings down the annual cost to £4152 per year (£12456 upfront cost. This is £11.34 per day).

Next, creating a public static IP address...

$ az network public-ip create -g MC_biosimspace_biosimspace_eastus -n static_ip --dns-name biosimspace --allocation-method Static
$ az network public-ip show -g MC_biosimspace_biosimspace_eastus -n static_ip --query "{ address: ipAddress }"
{
  "address": "52.168.53.185"
}

I then setup up the router, website, notebook and workshops are detailed in the subdirectories below.