What is DevOps?

What is DevOps

• "DevOps" lacks a concrete definition and is more of philosophy, a way of working
• DevOps stresses communication, collaboration, integration, automation, and measurement of cooperation
• It is practiced between software developers and other information-technology (IT) professionals
• DevOps is an approach to technology management that has the purpose of creating and delivering software applications to the users

What DevOps Isn't

• DevOps is not the same as Agile
• Agile can be a development methodology within an overall DevOps-style approach and share values such as short , continual improvement
• DevOps is not Continuous Integration, but CI is often a part of DevOps-style behavior
• DevOps is not "the developers running Operations."
• DevOps is automating things to the point where Operations runs itself in response to authorized actions.
• DevOps is not a software development methodology
• DevOps is not automation, but you can't have DevOps without it

How DevOps links with Cloud Computing

• DevOps is the automation of agile methodology
• Enables developers to respond to the needs of the business in near real-time.
• It helps remove the existing latency around software development
• Cloud computing can provide a standard and centralized platform for testing, deployment, and production for DevOps automation
• DevOps automation is becoming cloud-centric as providers support DevOps systemically on their platform
• Cloud providers leverage usage-based accounting, which tracks the use of resources by application, developer, user, and data
• DevOps is driving the interest and the growth of the cloud

Software development methodologies

• In the Traditional Waterfall Model complete product requirements are clear and fixed and the product definition is stable
• The Agile Development Model is characterized by requirements that change frequently and development needs to be fast
• The DevOps Approach is characterized by changing requirements, needing Agile development and operations
• The evolution of software development methodologies moved from traditional Waterfall Models, to Agile Development Models, and finally, to DevOps

Waterfall Model

• The stages for the Waterfall Model are:
  • Complete Requirement is gathered, and SRS is developed
  • The System is Planned and Designed using the SRS
  • Implementation of the System takes place
  • System is tested, and its quality is assured
    • System is deployed to the end users
    • Regular Maintenance of the system is done

Waterfall Model Challenges

• Code deployment time was huge after development stage
• Pressure to work on old, pending, and new code was high due to slow deployment time,
• It was difficult to maintain 100% uptime of the production environment
• Infrastructure automation tools did not work effectively
• The number of servers to be monitored increased over time, raising the complexity of the system
• It was challenging to provide feedback and diagnose issues in the product effectively

DevOps requirements

• A system with little to no waiting time is helpful for code deployment
• Systems should be running with 99% uptime
• A feedback system should be established
• Better collaboration is desirable

Dev & Ops challenges

• DevOps solutions such as continuous integrations ensure quick deployments of code, faster testing and feedback
• Devs waiting time for code deployment is effectively reduced
• DevOps solutions such as containerization/ virtualization ensures great reliability for service uptime
• It also provides configuration management to help organize plans and manage infrastructure
• Continuous Monitoring establishes effective feedbacks system to assure continuous administration

DevOps Principles

• Customer-Centric Action: DevOps team must take customer-centric action to constantly invest in service/product improvement
• End-To-End Responsibility: DevOps team need to provide performance support until they become end-of-life to enhance responsibility
• Continuous Improvement: Focuses on continuous improvement to minimize waste to speed up the improvement of product or services offered
• Automate everything: Automation is important for not only software development but also infrastructure landscape
• Work as one team: The developers, designers, testers and all other roles should work with complete collaboration
• Monitor and test everything: Monitoring, test, and robust procedure are keys for DevOps

Four DevOps values

• Development and test on a production-like environment
• Accelerated Deploy which means using proper processes and tools
• Constant collaboration and feedback
• Constantly checking operation quality

Waterfall vs. Agile vs. DevOps concerns

• Waterfall and Agile methodology is usually a development design
• DevOps focuses on the entire life cycle
• The Waterfall and Agile methodology show minimal concern for operations
• DevOps emphasizes main concern
• A Waterfall methodology views maintenance as a redevelopment phase
• Agile Methodology as a redevelopment phase
• DevOps replaces a running system, as is a maintenance procedure from previous versions of methodologies
• The waterfall focuses on designing a packaged system ready to be deployed
• Agile focuses on Various intermediate demo versions of the system
• DevOps focuses on A continuously updated running system
• Quality of the products or deployments assure to be Test-driven by end of testing or processing for previous versions
• DevOps focuses on Test-driven development and monitoring-driven operation with feedback to Dev

DevOps life cycle

• The stages of DevOps are: Build, Code, Test, Plan, Release, Deploy, Operate, and Monitor
• The processes within DevOps (development, testing, delivery, integration, deployment, and monitoring) are continuous
• This means DevOps is a continuous cycle of improvement and automation

DevOps tools

• Tools used in DevOps: Jira, Git, Puppet, Chef, Ansible, Maven, Eclipse, Gradle, Junit, Jenkins, Nagios, Splunk, Sensu, New Relic

DevOps stages

• In the 'planning' phase to 'coding' phase, the vision of the project is decided when code is written
• In Continuous Development, the software is continuously developed with deliverables broken down into multiple sprints of short cycles
• Code is maintained by using Version Control tools, in a single source of truth
• Ops can quickly rollback the deployed code and revert back to the previous stable state if there are any problems
• Tools such as Git and SVN are used for maintaining the different versions of the code with tools like Ant, Maven, Gradle for building
• In Continuous testing, developed software is continuously tested for bugs using tools like Selenium and TestNG

Continuous Integration

• This is the stage when existing code integrates with new functionality
• The system is checked to ensure the updated and constantly developed code reflect smoothly with the changes by the end users
• The developers commit changes to the source code more frequently
• Allows for early detection of problems, along with code reviews, unit testing, integration testing, and packaging

Benefits

• Multiple people can work separately on different parts of the code to reduce integration risks
• Focus on functionality to ensure higher code quality
• If a commit causes a build disruption, teams are notified immediately
• The impartial system in place reduces the frequency of quarrels among team members
• Different versions and builds of the code can help isolate and trace bugs efficiently, which makes things easier for QA teams
• Deploying projects is less tedious and time-consuming because it is automated

Jenkins

• Jenkins integrated with Git/SVN, can automatically schedule jobs and ready itself for builds and tests
• With testing tools, like Selenium, Jenkins can schedule/ automate test case execution
• With deployment tools, Jenkins can trigger deployments planned by configuration management/ containerization tools
• Can be integrated with monitoring tools (Splunk/ ELK/ Nagios/ NewRelic) to continuously monitor the status & performance of the server

Continuous Delivery

• Code changes are automatically prepared for a release to production
• With proper implementation, developers can always have a deployment-ready artifact
• Jenkins can be used to support continuous delivery

Continuous Deployment

• The code is deployed to the production environment
• Ensures the code is correctly deployed on all servers, while the system administrator is responsible for scaling up servers
• Configuration management tools (Puppet, Chef, SaltStack and Ansible) are used for executing the deployment
• Containerization tools (Docker and Vagrant) produce consistency across environments.

Continuous Deployment Process Flow

• Ensuring the code is correctly deployed on all servers is important
• Management tools perform consistent functional requirements and performance
• Releasing deployments, scheduling updates, and ensuring the configs are consistent are performed by the CM and containerization tools

Configuration Management Tools

• Responsible for the act of implementing and maintaining consistency in an application's functional requirements and performance
• In detail, releasing deployments to servers, scheduling updates, and keeping configurations consistent are all management tools performances
• Tools usually work based on the master-slave architecture
• Puppet is considered one of the greatest
• Most used services are Chef, Ansible, and SaltStack

Containerization Tools

• Are tools which help in stability throughout application build across testings
• Eliminates failures through the use of Packaging and replicating dependencies/packages that developers use
• Best tool to use is considered Docker
• Prominent tools that use them are: Amazon ECS, Azure Container Service and Google Container Engines

Continuous Delivery vs Continuous Deployment

• Continuous delivery has an operator that is checked and then moved to production
• Continuous deployment produces automatically without explicit operational approval

Continuous Monitoring

• Crucial in aims to improve quality of software by monitoring its performance
• The Operations team participates and monitors user activity for bugs/system health
• Tools used in the process are Splunk, Elk Stack, Nagios, NewRelic and Sensu

DevOps is a philosophy that stresses communication, collaboration, integration, automation, and measurement of cooperation between software developers and IT professionals. It's not the same as Agile, Continuous Integration, or developers running operations. DevOps automates processes so operations can manage themselves.