📚 Week 2 · Unit II · Lecture 6
Introduction to DevOps
& DevOps and Agile

Understanding the cultural shift that bridged the Dev–Ops divide, and how DevOps and Agile work together to deliver software at speed.

Dr. Mohsin Furkh DarSchool of Computer Sciences
DateTue, 16 Jun 2026 · 9:00 – 10:00 AM
ProgrammeBTech CSE – 5th / 6th Semester
Today's Agenda
1
What is DevOps? — Definitions & Origins
2
The Dev vs Ops Wall of Confusion
3
Core DevOps Principles (CAMS)
4
DevOps and Agile — Relationship & Differences
5
DevOps Lifecycle Overview
6
Real-World Benefits & Industry Context
Context & Recap
Where We Are in the Course
✅
Unit I (Completed): We covered the Waterfall Model, the Dev vs Ops Conflict, Agile Movement (2000), Agile vs Waterfall, Iterative Development, the Agile Manifesto, Customer Collaboration, and Responding to Change. You now have a solid foundation in why traditional approaches struggled.
🚀
Unit II Begins Today: We now explore DevOps — the natural evolution that takes Agile ideas further, extends them to Operations, and ties together culture, automation, measurement, and sharing.
The Problem We Solved
  • Software delivered in big, slow batches (Waterfall)
  • Agile fixed the Dev side — smaller, faster iterations
  • But Ops was still a separate silo — the conflict remained
The Problem DevOps Solves
  • Breaks the Dev ↔ Ops wall permanently
  • Automates the path from code to production
  • Enables continuous, reliable software delivery
Definition
What Is DevOps?

DevOps is a set of practices that combines software development (Dev) and IT operations (Ops), aiming to shorten the development lifecycle and provide continuous delivery with high software quality.

— Wikipedia / Industry Consensus Definition
🧑‍💻
A Culture
DevOps is first and foremost a mindset shift — shared ownership between developers and operations teams.
⚙️
A Practice
Automation, CI/CD pipelines, infrastructure-as-code, continuous monitoring — translating culture into action.
📐
A Movement
Emerged from the 2009 Velocity Conference talk by John Allspaw & Paul Hammond: "10+ deploys per day."
💡
Key Insight: DevOps is NOT just a set of tools. Tools enable DevOps but do not define it. You can use Jenkins, Docker, and Kubernetes yet still have silos — if the culture is wrong, DevOps fails.
Origins
How DevOps Was Born
2001
Agile Manifesto Published
The 17 signatories publish the Agile Manifesto. Dev teams adopt iterative practices — but Ops teams are left out. Tension increases between fast-moving Dev and change-resistant Ops.
2008
Patrick Debois Starts the Conversation
At the Agile Conference 2008, Patrick Debois (later called "the father of DevOps") presents ideas about bridging Dev and Ops. He sees firsthand the pain of siloed teams on a Belgian government project.
2009
Velocity Conference: "10 Deploys per Day"
John Allspaw & Paul Hammond from Flickr deliver the landmark talk showing that Dev and Ops working together can push 10+ production deployments per day safely. Patrick Debois responds by organizing the first DevOpsDays in Ghent, Belgium.
2013
The Phoenix Project
Gene Kim, Kevin Behr & George Spafford publish The Phoenix Project — a novel that popularizes DevOps principles for mainstream IT audiences. The "Three Ways" of DevOps are introduced.
Now
DevOps is Mainstream
Major organizations (Amazon, Netflix, Google, Microsoft) deploy hundreds to thousands of times per day. DevOps is the default approach for cloud-native software delivery.
Core Problem
The "Wall of Confusion"
🧱
The fundamental conflict: Developers are incentivized to ship change fast. Operations teams are incentivized to keep systems stable. These goals are naturally in tension — and that tension creates the "Wall of Confusion."
🧑‍💻 Development Team Goals
  • Ship new features quickly
  • Respond to user feedback fast
  • Experiment & iterate
  • Deploy frequently
  • Innovation over stability
🛠️ Operations Team Goals
  • Maintain uptime & availability
  • Prevent system failures
  • Slow down risky changes
  • Deploy infrequently, predictably
  • Stability over innovation
🎯 What DevOps Does

DevOps tears down this wall by creating shared responsibility, shared tools, and shared metrics between Dev and Ops. Both teams co-own the entire software lifecycle — from code to production to monitoring.

Core Principles
CAMS — The Four Pillars of DevOps

Coined by Damon Edwards and John Willis, CAMS represents the foundational elements of DevOps culture. We'll study each pillar in depth in upcoming lectures.

🎭
C — Culture
People and processes first. Remove blame, build psychological safety, foster shared ownership between Dev, Ops, QA, and Business teams. "Culture eats strategy for breakfast."
🤖
A — Automation
Automate repetitive tasks: testing, deployments, provisioning, monitoring. Automation creates speed, consistency, and reduces human error in the delivery pipeline.
📊
M — Measurement
Track everything — deployment frequency, lead time, mean time to recovery (MTTR), change failure rate. Data-driven decisions replace gut feel and finger-pointing.
🤝
S — Sharing
Share knowledge, tools, and feedback openly across teams. Internal open-source practices, post-mortems without blame, cross-team transparency.
📌
Remember: CAMS is ordered deliberately. Without Culture, Automation is just machinery people resent. Without Measurement, you can't know if Automation is working. Without Sharing, improvements stay local.
Lifecycle
The DevOps Infinity Loop

DevOps operates as a continuous, never-ending loop connecting development activities (Plan → Code → Build → Test) with operations activities (Release → Deploy → Operate → Monitor). The figure-eight (∞) shape represents this unbroken cycle.

📋 PlanUser Stories / Backlog
💻 CodeVersion Control (Git)
🔨 BuildCompile / Package
🧪 TestAutomated Testing
🚀 ReleaseVersioning / Approval
📦 DeployTo Staging / Prod
⚙️ OperateConfig / Scale
📡 MonitorLogs / Alerts
Dev Side (Left Loop)
  • Plan: Feature planning, sprint backlogs
  • Code: Developers write & commit code
  • Build: CI server compiles the code
  • Test: Automated unit, integration, regression tests
Ops Side (Right Loop)
  • Release: Version tagged, change approved
  • Deploy: Pushed to staging then production
  • Operate: Scaling, patching, configuration
  • Monitor: Logs, metrics, user feedback
Relationships
DevOps and Agile — Partners, Not Rivals
🔗
Key Relationship: Agile is the what and how of building software iteratively. DevOps is the how of getting that software to users reliably and continuously. They are complementary practices, not alternatives.
Dimension Agile DevOps
Primary Focus Development process & team collaboration Full lifecycle from Dev through Operations
Team Scope Dev + QA + Product (no Ops) Dev + QA + Ops + Security (all silos)
Feedback Loop Sprint reviews every 2–4 weeks Continuous feedback from production monitoring
Key Output Working software at end of sprint Continuously deployed, monitored software
Delivery Speed Faster than Waterfall (sprints) Even faster (minutes to hours per release)
Automation Level Low to moderate (tests, story tracking) High (CI/CD pipelines, IaC, monitoring)
Mindset Iterative, adaptive, customer-focused Collaborative, automated, data-driven
Relationship Deep Dive
How Agile Enables DevOps

Agile practices create the pre-conditions that make DevOps possible. You cannot practice DevOps effectively without already having some Agile discipline in place.

🔄
Small, Frequent Commits
Agile sprints encourage small feature increments, which are exactly what CI/CD pipelines are designed to handle efficiently. Large batches break CI/CD.
🧪
Test-Driven Development
Agile's emphasis on TDD means code arrives at the CI pipeline with tests already written — enabling automated quality gates in DevOps pipelines.
👥
Cross-Functional Teams
Agile's cross-functional team model is a prerequisite for DevOps — it breaks the developer-only silo and invites Ops and security members into the team early.
🎯
Definition of Done
In DevOps-matured Agile teams, "Done" means not just coded and tested, but deployed to production and monitored — extending Agile's DoD into operations.
🎯 Bottom Line

Agile without DevOps = Fast development, slow delivery (code piles up waiting for manual deployment). DevOps without Agile = Fast pipeline, but large, risky batches of code going through it. Together, they create a complete system of continuous value delivery.

Industry Evidence
The Business Case for DevOps

The DORA (DevOps Research & Assessment) team at Google studies elite-performing engineering organizations annually. Their findings consistently show dramatic improvements from DevOps adoption.

208×
More frequent software deployments
106×
Faster lead time from commit to deploy
2,604×
Faster mean time to recover from failures
7×
Lower change failure rate
Source: 2019 State of DevOps Report — Elite vs Low performers
Real-World Examples
  • Amazon: Deploys every 11.7 seconds on average
  • Netflix: 100s of deployments per day, zero downtime
  • Etsy: From 2 deploys/week to 50+ per day after DevOps adoption
Key DORA Metrics (4 Golden Signals)
  • Deployment Frequency — how often you deploy
  • Lead Time for Changes — commit → production
  • Mean Time to Recovery — how fast you fix incidents
  • Change Failure Rate — % deployments causing incidents
Clarifications
Common DevOps Misconceptions
❌
"DevOps means Developers do Operations work" — False. DevOps means Dev and Ops collaborate and share responsibility. Developers don't replace SREs or sysadmins; they work together on shared goals.
❌
"DevOps is just CI/CD" — False. CI/CD is one practice within DevOps. DevOps also includes cultural transformation, monitoring, security (DevSecOps), and organizational change.
❌
"DevOps replaces Agile" — False. They are complementary. Agile manages how we develop software; DevOps manages how we deliver and operate it. Most mature teams practice both.
✅
DevOps is: A cultural philosophy + a set of technical practices + organizational change + continuous improvement. Its success depends more on people and culture than on any specific tool or technology.
Lecture Summary
Key Takeaways

Lecture 6 — Introduction to DevOps & DevOps and Agile

1
DevOps = Culture + AutomationIt bridges the Dev–Ops divide through shared ownership, not just tooling. Culture comes first.
2
Wall of ConfusionDev wants speed; Ops wants stability. DevOps resolves this conflict through shared goals and practices.
3
CAMS FrameworkCulture, Automation, Measurement, Sharing — the four pillars that define true DevOps adoption.
4
DevOps Lifecycle (∞)An 8-phase continuous loop: Plan → Code → Build → Test → Release → Deploy → Operate → Monitor.
5
Agile + DevOps TogetherAgile handles iterative development; DevOps handles continuous delivery and operations. They are partners.
6
Proven Business ValueDORA research: Elite performers deploy 208× more often with 7× lower change failure rates than low performers.
📚 Next Lecture

Lecture 7 (Mon, 22 Jun): Minimum Viable Product & Application Deployment — We'll explore the MVP concept, why shipping fast beats shipping perfect, and how applications are deployed in a DevOps environment.

Dr. Mohsin Furkh Dar · School of Computer Sciences · UPES Dehradun

CSDV3016P / CSDV3017 · DevOps Overview · Summer 2026