Adding site content to /site directory (see #194)

Signed-off-by: Luke Wilson <luke.wilson@dell.com>

site/CONTRIBUTORS.md

@@ -0,0 +1,6 @@
+# Omnia Maintainers
+- Luke Wilson and John Lockman (Dell Technologies)
+<img src="images/delltech.jpg" height="90px" alt="Dell Technologies">
+
+# Omnia Contributors
+<img src="images/delltech.jpg" height="90px" alt="Dell Technologies"> <img src="images/pisa.png" height="100px" alt="Universita di Pisa">

site/INSTALL.md

@@ -0,0 +1,105 @@
+## TL;DR Installation
+ 
+### Kubernetes
+Install Kubernetes and all dependencies
+```
+ansible-playbook -i host_inventory_file kubernetes/kubernetes.yml
+```
+
+Initialize K8s cluster
+```
+ansible-playbook -i host_inventory_file kubernetes/kubernetes.yml --tags "init"
+```
+
+### Install Kubeflow 
+```
+ansible-playbook -i host_inventory_file kubernetes/kubeflow.yaml
+```
+
+### Slurm
+```
+ansible-playbook -i host_inventory_file slurm/slurm.yml
+```
+
+# Omnia  
+Omnia is a collection of [Ansible](https://www.ansible.com/) playbooks which perform:
+* Installation of [Slurm](https://slurm.schedmd.com/) and/or [Kubernetes](https://kubernetes.io/) on servers already provisioned with a standard [CentOS](https://www.centos.org/) image.
+* Installation of auxiliary scripts for administrator functions such as moving nodes between Slurm and Kubernetes personalities.
+
+Omnia playbooks perform several tasks:
+`common` playbook handles installation of software 
+* Add yum repositories:
+    - Kubernetes (Google)
+    - El Repo (for Nvidia drivers)
+    - EPEL (Extra Packages for Enterprise Linux)
+* Install Packages from repos:
+    - bash-completion
+    - docker
+    - gcc
+    - python-pip
+    - kubelet
+    - kubeadm
+    - kubectl
+    - nfs-utils
+    - nvidia-detect
+    - yum-plugin-versionlock
+* Restart and enable system level services
+    - Docker
+    - Kubelet
+
+`computeGPU` playbook installs Nvidia drivers and nvidia-container-runtime-hook
+* Add yum repositories:
+    - Nvidia (container runtime)
+* Install Packages from repos:
+    - kmod-nvidia
+    - nvidia-container-runtime-hook
+* Restart and enable system level services
+    - Docker
+    - Kubelet
+* Configuration:
+    - Enable GPU Device Plugins (nvidia-container-runtime-hook)
+    - Modify kubeadm config to allow GPUs as schedulable resource 
+* Restart and enable system level services
+    - Docker
+    - Kubelet
+
+`master` playbook
+* Install Helm v3
+* (optional) add firewall rules for Slurm and kubernetes
+
+Everything from this point on can be called by using the `init` tag
+```
+ansible-playbook -i host_inventory_file kubernetes/kubernetes.yml --tags "init"
+```
+
+`startmaster` playbook
+* turn off swap
+*Initialize Kubernetes
+    * Head/master
+        - Start K8S pass startup token to compute/slaves
+        - Initialize software defined networking (Calico)
+
+`startworkers` playbook
+* turn off swap
+* Join k8s cluster
+
+`startservices` playbook
+* Setup K8S Dashboard
+* Add `stable` repo to helm
+* Add `jupyterhub` repo to helm
+* Update helm repos
+* Deploy NFS client Provisioner
+* Deploy Jupyterhub
+* Deploy Prometheus
+* Install MPI Operator
+
+
+### Slurm
+* Downloads and builds Slurm from source
+* Install package dependencies
+    - Python3
+    - munge
+    - MariaDB
+    - MariaDB development libraries
+* Build Slurm configuration files
+

site/PREINSTALL.md

@@ -0,0 +1,27 @@
+# Pre-Installation Preparation
+
+## Assumptions
+Omnia assumes that prior to installation:
+* Systems have a base operating system (currently CentOS 7 or 8)
+* Network(s) has been cabled and nodes can reach the internet
+* SSH Keys for `root` have been installed on all nodes to allow for password-less SSH
+* Ansible is installed on either the master node or a separate deployment node
+```
+yum install ansible
+```
+
+## Example system designs
+Omnia can configure systems which use Ethernet- or Infiniband-based fabric to connect the compute servers.
+
+![Example system configuration with Ethernet fabric](images/example-system-ethernet.png)
+
+![Example system configuration with Infiniband fabric](images/example-system-infiniband.png)
+
+## Network Setup
+Omnia assumes that servers are already connected to the network and have access to the internet.
+### Network Topology
+Possible network configurations include:
+* A flat topology where all nodes are connected to a switch which includes an uplink to the internet. This requires multiple externally-facing IP addresses
+* A hierarchical topology where compute nodes are connected to a common switch, but the master node contains a second network connection which is connected to the internet. All outbound/inbound traffic would be routed through the master node. This requires setting up firewall rules for IP masquerade, see [here](https://www.server-world.info/en/note?os=CentOS_7&p=firewalld&f=2) for an example.
+### IP and Hostname Assignment
+The recommended setup is to assign IP addresses to individual servers. This can be done manually by logging onto each node, or via DHCP.

site/_config.yml

@@ -0,0 +1,4 @@
+theme: jekyll-theme-minimal
+title: Omnia
+description: Ansible playbook-based tools for deploying Slurm and Kubernetes clusters for High Performance Computing, Machine Learning, Deep Learning, and High-Performance Data Analytics
+logo: images/omnia-logo.png

site/metalLB/README.md

@@ -0,0 +1,10 @@
+# MetalLB 
+
+MetalLB is a load-balancer implementation for bare metal Kubernetes clusters, using standard routing protocols.
+https://metallb.universe.tf/
+
+Omnia installs MetalLB by manifest in the playbook `startservices`. A default configuration is provdied for layer2 protocol and an example for providing an address pool. Modify metal-config.yaml to suit your network requirements and apply the changes using with: 
+
+``` 
+kubectl apply -f metal-config.yaml
+```

site/metalLB/metal-config.yaml

@@ -0,0 +1,21 @@
+apiVersion: v1
+kind: ConfigMap
+metadata:
+  namespace: metallb-system
+  name: config
+data:
+  config: |
+    address-pools:
+    - name: default
+      protocol: layer2
+      addresses:
+      - 192.168.2.150/32
+      - 192.168.2.151/32
+      - 192.168.2.152/32
+      - 192.168.2.153/32
+      - 192.168.2.154/32
+      - 192.168.2.155/32
+      - 192.168.2.156/32
+      - 192.168.2.157/32
+      - 192.168.2.158/32
+      - 192.168.2.159/32