Using HPE MapR Data Fabric Storage Container

About This Lab

Welcome to this Lab on HPE CSI Driver for Kubernetes with MapR Data Fabric Storage and SQL Server 2019 Lab.

Kubernetes allows you to create and manage containers. From a perspective of SQL Server, kubernetes provides:

• A scalable architecture to deploy containerized applications and data platforms such as SQL Server
• Persistent storage for stateful containers like SQL Server
• Built-in load balancers to abstract application connections to SQL Server
• Built-in high availability for stateful containers like SQL Server
• An ecosystem for Operators to simplify application deployment and manage high availability

MapR Data fabric or kubernetes benefits :

• Persist data for containerized applications
• Scale data and performance as containers grow
• Benefit from MapR tickets, for end-to-end security
• Multi-tenant deployment and access

The purpose of this demo is to show you how to use the MapR storage function for SQL Server in a kubernet environment.

Note: The Kubernete cluster and MapR Data Fabric cluster and CSI driver are already deployed

Learning Objectives

When you complete this live demo, you will be able to:

• Understand Container Storage Interface (CSI) Driver for Kubernetes
• Know how to install the HPE CSI driver MapR for kubernetes
• Understand Kubernetes deployment
• Understand Dynamic Provisionning , Persitent Storage on Kubernetes
• Understand the basics of deploying SQL Server on an Kubernetes cluster
• Connect and run queries against SQL Server deployed on Kubernetes

Business Applications of this Lab

• Developers looking to deploy a database container for their applications on Kubernetes.
• Database Administrators looking to understand how to deploy database platforms like SQL Server in a Kubernetes cluster using Kubernetes.

Infrastructure description

You’ll need a local systems or Virtual Machines that you are able to install software on.

The infrastructure of the platform is composed of:

• 1 Master node : arvine
• 3 Worker nodes : gamaret, garonoir, syrah
• 1 MapR Data Fabric cluster 6.2 with 3 nodes , the same nodes for kubernetes cluster

• Microsoft SQL Server 2019 container
• RedHat Enterprise server 8.2
• Kubernetes 1.19.6 or release higher

• 1 Node with JupyterLab-JupyterHub and BASH kernel
• active ssh connection between your JupyterLab-JupiterHub host and one of the Kubernetes cluster nodes (for example with the master node)

Lab Details

This Lab uses Kubernetes,Docker , SQL Server 2019 , Azure Data Studio, SQL Command Line Tools, and the Kubernetes CLI (kubectl).

Lab Modules

Terminology

Throughout this demonstration we take for example the connection with the user stundent4 but all the configurations apply to any user.

Setup

Clone the repository :

$ git clone https://github.com/colussim/HPE-Data-Fabric-Jupyter-playbook.git
$ cd HPE-Data-Fabric-Jupyter-playbook
$

• Copy the student4-datafabric-csi-jupyter directory into your JuperterLab environment or into the directory of your JuperterHub user.
• Create the same user used in your JupyterLab-JupyterHub environment on your master node of your kubernetes cluster.
• Generate a ras ssh key for this user.
• Update the authorized_keys file with the public keys (in master node)
• To make kubectl work for your non-root user, run these commands, which are also part of the result of the kubeadm init command:

  mkdir -p $HOME/.kube
  sudo cp -i /etc/kubernetes/admin.conf $HOME/.kube/config
  sudo chown $(id -u):$(id -g) $HOME/.kube/config

• Copy the sql-mapr directory contained in the tudent4-datafabric-csi-server directory of the repository into the home directory of your create user on your master node

Usage

Load the Notebook : HPEMAPR-SQL-Intro.ipynb

Conclusion

As you can see, it is quite easy to set up and use HPE MapR CSI Driver .

This version include all the latest capabilities in the CSI specification :

• Dynamic Provisionning : Parameter Overloading
• Volume Snapshots
• Data Sources : Cloning - Volume Snapshot and Restore
• Volume Expansion

Do not hesitate to contact me for more information

Resources

HPE MapR CSI Driver for Kubernetes on GitHub

HPE MapR CSI documentation

Thank You