Yesterday I attended an expert session of industry use cases on Kubernetes

Here’s a summary of the session:

Centralized Version Control System

Only one central repository, it contains all history. Developers have to get latest version of code from central repository and then make changes in code then commit or merge those changes into central repository. Working on branches in difficult as developer often faces merge conflicts. It does not provide offline access. It is slower as every command need to communicate with server. If CVCS Server is down, developers cannot work. Example: Subversion (SVN), CVS, Perforce,etc

Distributed Version Control System

Every developer has their own server, they will have copy of entire history in their local machine. Everyone can work locally and disconnected. We can work offline as a client copies the entire repository on their local machine. It is faster as mostly user deals with local copy without hitting server every time. If DVCS server is down, developer can work using their local copies. Example: Git, Mercurial,etc

Why Git over CVS?

· Easy to collaborate in teams using Git

· Easy to create branches in Git

· While merging if there are many conflicts you can solve them and commit without affecting the main branch then merge

· REST API implementation in Git

Resource Optimization

It is necessary to ensure that all the resources are utilized efficiently for this resource optimization is necessary. We can use monitoring tools and check which resources are utilized how much during the day and then reallocate resources for some other use . Setting limits on resources is necessary so that we have resources left to be allocated at time of failure of some server.

Benefits

· Increased revenue

· Conflict resolution

· Improve project delivery

· Reduce administration costs

Chaos Testing:

Chaos testing means to purposefully crash a production system.Chaos Monkey is a tool invented in 2011 by Netflix to test the resilience of its IT infrastructure. It works by intentionally disabling computers in Netflix’s production network to test how remaining systems respond to the outage. This concept can help to check how cluster behaves if something fails and help to find weaknesses and help in increasing resiliency of the system

SCC(Security Context Constraints):

They can control

· Whether a pod can run privileged containers

· Capabilities that a container can request

· Use of host directories as volumes

· SE linux context of the container

· Container user ID

· Use of host namespaces and networking

· Allocation of FS group