CSDV3017 · Lecture 2 · 09 Jun 2026 · 12:00–13:00
DevOps Practices
& History
Agile vs DevOps · Kanban · Scrum

BTech CSE 6th Semester · School of Computer Sciences · UPES Dehradun

Faculty: Dr. Mohsin Furkh Dar
Mode: Online · MS Teams
Session: Tue 12:00 – 13:00
Week: 1 of 6

Recap from Lecture 1

  • DevOps = Culture + Automation + Measurement + Sharing
  • The Dev–Ops "wall of confusion" and why it needed solving
  • The DevOps ∞ lifecycle: Plan → Code → Build → Test → Release → Deploy → Operate → Monitor
  • DORA metrics: Deployment Frequency, Lead Time, Change Failure Rate, MTTR

Today's Learning Objectives

  • Trace the full history and evolution of DevOps practices
  • Identify and explain the eight core DevOps practices
  • Distinguish Agile from DevOps — scope, focus, and overlap
  • Explain the Kanban method and its WIP-limit discipline
  • Understand the Scrum framework: roles, events, and artifacts
Origins & Evolution

History of DevOps

DevOps has roots going back to Lean manufacturing and Agile software development. Its evolution spans over two decades of industry pain and innovation.

1980s – 1990s
Waterfall & the Silo Era
Software delivered in long cycles (12–24 months). Dev, QA, and Ops as completely separate departments. Deployments were rare, risky, and manual. ITIL frameworks governed Operations.
2001
The Agile Manifesto
17 developers sign the Agile Manifesto in Snowbird, Utah. Iterative development, customer collaboration, and responding to change over rigid plans. Agile transforms how Dev teams work — but Ops is left unchanged.
2007 – 2008
The Friction Point
Patrick Debois and Andrew Shafer meet at an Agile conference and identify the "Agile Infrastructure" problem: Dev teams became faster, but Ops couldn't keep up. The gap widened.
2009
DevOps is Born
John Allspaw & Paul Hammond: "10+ Deploys Per Day: Dev & Ops Cooperation at Flickr" — Velocity Conference. Patrick Debois organises the first DevOpsDays in Ghent, Belgium. The hashtag #DevOps is born.
2013 – 2015
Mainstream Adoption
Gene Kim et al. publish The Phoenix Project (2013) and The DevOps Handbook (2016). Docker released (2013) — containers revolutionise deployment. AWS, Azure, GCP embrace DevOps tooling.
2018 – Present
DevOps Matures: DevSecOps, GitOps, AIOps
Security integrated into pipelines (DevSecOps). GitOps uses Git as the single source of truth for infrastructure. AI begins augmenting monitoring and incident response (AIOps). DORA research benchmarks elite performers.
Core DevOps Practices

Eight Foundational
DevOps Practices

DevOps is not a single tool — it is a collection of interconnected practices that collectively shorten the feedback loop between writing code and delivering value.

  • 🔗

    1. Continuous Integration (CI) Pipeline

    Developers merge code into a shared repository frequently. Each merge triggers an automated build and test suite. Detects integration issues early. Tools: Jenkins, GitHub Actions, GitLab CI.

  • 🚢

    2. Continuous Delivery & Deployment (CD) Pipeline

    Code passing CI is automatically prepared for release (Continuous Delivery) or automatically deployed to production (Continuous Deployment). Eliminates manual release bottlenecks.

  • 🏗️

    3. Infrastructure as Code (IaC) Automation

    Infrastructure (servers, networks, databases) defined and managed through code, not manual clicks. Version-controlled, repeatable, and auditable. Tools: Terraform, Ansible, Puppet, Chef.

  • 📊

    4. Continuous Monitoring & Observability Feedback

    Real-time visibility into application health, performance, and security. Metrics, logs, and traces (the three pillars of observability). Tools: Prometheus, Grafana, ELK Stack, Datadog.

  • 🤝

    5. Collaboration & Communication Culture

    Shared on-call, joint post-mortems, cross-functional teams. Breaking down silos through shared tools, dashboards, and blameless retrospectives. "You build it, you run it."

  • 🔒

    6. DevSecOps — Shift Left Security Security

    Security practices integrated into every stage of the pipeline — not bolted on at the end. SAST, DAST, dependency scanning in CI. Security as a shared responsibility.

  • 📦

    7. Microservices & Containerisation Architecture

    Applications decomposed into small, independently deployable services. Containers (Docker) ensure consistency across environments. Kubernetes orchestrates at scale.

  • 🔁

    8. Feedback Loops & Continuous Improvement Culture

    Short, frequent feedback cycles at every stage — from unit tests to production metrics to customer analytics. Kaizen mindset: always improve. Retrospectives drive learning.

The Foundation

Agile: The
Manifesto & Values

Before we compare Agile and DevOps, we must understand what Agile is. The Agile Manifesto (2001) defines four core value statements.

The Agile Manifesto — 4 Core Values

Individuals and interactions over processes and tools · Working software over comprehensive documentation · Customer collaboration over contract negotiation · Responding to change over following a plan

The items on the left are valued more, but the items on the right still have value.

12 Principles Behind the Agile Manifesto (selected)

Principle 1
Customer Satisfaction First
Highest priority is to satisfy the customer through early and continuous delivery of valuable software.
Principle 3
Deliver Frequently
Deliver working software frequently — from a couple of weeks to a couple of months, with a preference for shorter timescales.
Principle 4
Break Down Silos
Business people and developers must work together daily throughout the project.
Principle 12
Reflect & Adjust
At regular intervals, the team reflects on how to become more effective, then tunes and adjusts its behaviour accordingly.
Agile transformed how development teams work — but it said very little about how code would actually reach production. That gap is what DevOps was built to fill.
Key Distinction

Agile vs DevOps:
What's the Difference?

Agile and DevOps are complementary, not competing. They share values but differ in scope, audience, and what they optimise for.

Dimension Agile DevOps
Primary FocusSoftware development processEnd-to-end software delivery & operations
ScopeDev team (planning → working software)Dev + Ops + QA + Security (code → production)
Key GoalRespond to change; deliver working software fastReduce time-to-market; increase reliability
Team InvolvedDevelopment team primarilyCross-functional (Dev, Ops, QA, Security)
Feedback LoopSprint review, customer feedbackProduction metrics, monitoring, real-user data
Delivery CadenceEnd of sprint (1–4 weeks)Continuous — multiple times per day possible
InfrastructureNot in scopeExplicitly in scope via IaC & cloud
Key FrameworksScrum, Kanban, XP, SAFeCI/CD pipelines, GitOps, SRE
Relationship✓ DevOps extends and operationalises Agile — they work best together
🔁
Where They Overlap
Both value iterative delivery, collaboration, feedback loops, and continuous improvement. DevOps teams almost always use Agile practices (Scrum or Kanban) within their workflow.
🔀
Key Analogy
Agile is about how you build the car. DevOps is about how you get it to the customer and keep it running on the road.
Agile Framework

Kanban: Visualise
Your Workflow

Kanban (看板) is Japanese for "visual board." Originally developed by Toyota for lean manufacturing, it was adapted for software by David J. Anderson (~2007).

Definition

Kanban is a visual workflow management method that helps teams manage and improve the flow of work. It focuses on limiting work-in-progress (WIP) to reduce multitasking, identify bottlenecks, and optimise throughput.

The Six Core Kanban Practices

👁️
Visualise the Workflow
Make every piece of work visible on the board. Hidden work creates hidden bottlenecks.
🚦
Limit Work-in-Progress
WIP limits per column stop multitasking and force the team to finish before starting new work.
💧
Manage Flow
Monitor how items move through the board. Optimise for smooth, fast, predictable delivery.
📜
Make Policies Explicit
Define what "done" means for each column. Clear rules reduce miscommunication.
🔄
Feedback Loops
Regular cadences — daily standups, replenishment meetings, retrospectives.
📈
Improve Collaboratively
Use data (lead time, cycle time, throughput) to drive incremental process improvements.
Kanban in Practice

The Kanban Board
& WIP Limits

A Kanban board makes work visible and enforces discipline through WIP limits. When a column is full, the team must pull work to completion before adding new items.

Example: A DevOps team's Kanban board (WIP limits shown in brackets)

Backlog
Feature
Set up Prometheus monitoring
Bug
Fix login timeout on mobile
Ops
Upgrade Kubernetes to v1.30
Feature
Add GitHub Actions workflow
In Progress [WIP: 3]
Feature
Docker multi-stage build
Bug
Pipeline fails on merge
Ops
Configure Nginx reverse proxy
In Review [WIP: 2]
Feature
Terraform EC2 provisioning
Bug
Memory leak in Node.js API
Done ✓
Ops
Jenkins pipeline setup
Feature
Dockerfile for backend service
⏱️ Key Metrics
Lead Time: Time from item creation to delivery.
Cycle Time: Time from work start to completion.
Throughput: Items completed per week.
🛑 WIP Limits in DevOps
WIP limits are especially powerful in CI/CD pipelines — they prevent developers from flooding the pipeline with unreviewed PRs, keeping the queue manageable.
Agile Framework

Scrum: Delivering
in Sprints

Scrum is the world's most widely used Agile framework. Defined in the Scrum Guide by Ken Schwaber and Jeff Sutherland, it organises work into fixed-length iterations called Sprints (1–4 weeks).

Definition

Scrum is a lightweight framework that helps teams generate value through adaptive solutions for complex problems. It is built on empiricism (transparency, inspection, adaptation) and lean thinking (reduce waste, focus on essentials).

🧑‍💼
Product Owner
Defines the product vision. Manages and prioritises the Product Backlog. Voice of the customer. Maximises value delivered by the team.
🧭
Scrum Master
Servant-leader and coach. Ensures Scrum is understood and enacted. Removes impediments. Facilitates all Scrum events. Does not manage the team.
👩‍💻
Developers
Cross-functional team of 3–9. Self-organising. Responsible for the Sprint Goal and delivering the Increment. No sub-teams or hierarchies within.
In DevOps contexts, the Scrum team includes developers and operations engineers — the "Developers" role in Scrum is deliberately broad. Infrastructure tasks go on the Sprint Backlog alongside feature work.
Scrum Framework

Scrum Events,
Artifacts & the Sprint

Scrum's five events create a regular heartbeat for the team. Three artifacts provide transparency into the work.

The Sprint Flow (2-week example)

📋 Sprint Planning
Day 1 · 4hrs max
🏃 Sprint
1–4 Weeks
☀️ Daily Scrum
Every day · 15 min
🎯 Sprint Review
Last day · Demo
🔄 Retrospective
Last day · Improve
↩ next sprint
📋 Sprint Planning
Team selects items from Product Backlog, defines Sprint Goal, breaks items into tasks. Time-boxed to 8hrs for 1-month sprint.
☀️ Daily Scrum (Standup)
15-min daily sync. Three questions: What did I do yesterday? What will I do today? Any blockers?
🎯 Sprint Review
Demo of the working Increment to stakeholders. Inspect what was done. Adapt the Product Backlog. Informal — feedback welcome.
🔄 Retrospective
Inspect the process (not the product). Identify improvements for the next sprint. Blameless, psychologically safe.

Three Scrum Artifacts

Product Backlog

Ordered list of everything needed in the product. Owned by Product Owner. Always evolving.

Sprint Backlog

Items selected for this Sprint + the plan to achieve the Sprint Goal. Owned by Developers.

Increment

The sum of completed Sprint Backlog items. Must meet the Definition of Done. Potentially shippable.

Framework Comparison

Kanban vs Scrum:
When to Use Which?

Both are Agile frameworks, but they suit different team contexts. Many DevOps teams use a hybrid approach called Scrumban.

Dimension Kanban Scrum
CadenceContinuous flow — no fixed iterationsFixed-length Sprints (1–4 weeks)
RolesNo prescribed rolesProduct Owner, Scrum Master, Developers
Work PlanningPull items as capacity allowsCommit to Sprint Backlog at Sprint Planning
WIP ControlExplicit WIP limits per columnControlled by Sprint capacity
Changes mid-cycleCan be added any timeNo changes to Sprint Backlog mid-Sprint
Best ForOps, support, maintenance, continuous delivery pipelinesFeature development with clear goals and stakeholder reviews
MetricsCycle time, lead time, throughput, CFDVelocity, Sprint burndown, release burnup
Key VisualisationKanban board with WIP limitsSprint board + burndown chart
Scrumban is a popular hybrid: Scrum's Sprint structure and review ceremonies combined with Kanban's WIP limits and continuous flow. Widely used in DevOps teams that do both feature work and operational tasks.
Lecture Wrap-up

Key Takeaways

What you should be able to explain after today's session:

  • 01DevOps evolved from Waterfall → Agile (2001) → Agile Infrastructure problem (2007) → DevOpsDays Ghent (2009) → mainstream cloud-native adoption (2013+).
  • 02The eight core DevOps practices include CI, CD, IaC, Continuous Monitoring, Collaboration, DevSecOps, Containerisation, and Feedback Loops.
  • 03Agile focuses on how to build software (Dev team). DevOps extends this to how to deliver and operate software (Dev + Ops + QA + Security). They are complementary.
  • 04Kanban is a continuous flow method that visualises work and enforces WIP limits to prevent bottlenecks. Key metrics: cycle time, lead time, throughput.
  • 05Scrum organises work into Sprints with three roles (PO, SM, Developers), five events (Planning, Daily, Review, Retro, Sprint itself), and three artifacts (Product Backlog, Sprint Backlog, Increment).
  • 06Kanban suits operational and support work; Scrum suits feature development. Scrumban combines both — popular in DevOps teams.
📚
Next Session
Tue, 09 Jun · 14:00–15:00 — DevOps Tools: Continuous Dev, Integration & Testing; Deployment & Monitoring; DevSecOps; DevOps vs SRE; Toolchain
💡
Reflection Question
Think of a software project you know. Which framework — Kanban or Scrum — would suit it better, and why?
🔗
Academic Portal
Resources & references: mohsinfurkh.github.io/academic-portal
arrow keys · swipe on mobile