AZ-400 Chapter 1-3 Flashcards
What is devops? (3 things)
Collaboration
Continuous Delivery
Breaking Down Silos
What is the struggle between op and dev?
Op wants as few changes as possible while dev is rewarded for more changes
What is fastlaning?
Adding urgent, unplanned work to the top of the sprint board as Kanban
What are the benefits of devops culture? (2 things)
Reduced cycle time (lower change failure rate), increased employee satisfaction
What is cycle time?
Time between starting work on a task and the moment a user can use it
Metrics for measuring results (4 things)
Cycle Time and Lead Time
Amount of work in progress
Mean time to recovery
Change rate (change failure rate included)
What is lead time?
Time between a request for a feature and the user reciving it
7 Key DevOps Practices
Config Management
Release Management
CI
CD
Infrastructure as code
Test automation
App Performance Monitoring
7 DevOps Habits
Team auto…
Team autonomy
managing technical debt
flow of customer value
hypothesis driven development
evidence gathered in PROD
live-site culture
infrastructure as a resource
Where should config files live?
In source control as JSON or YAML. This prevents configuration drift
The version and config of a release is stored as?
An artifact
What kind of metrics are used for app performance monitoring? (3)
Response time, # of requests, use patterns
What are the 5 stages of DevOps evolution?
Normalizing the tech stack
Standardizing and reducing variability
Expand DevOps Practices
Automate infrastructure delivery
Provide self-service capabilities
What is SRE?
Site reliability engineering
Goal is to improve stability and reliability
Use automation to reduce repetitive tasks
Should spend 50% of time on Op work, the rest on automation, deployment pipelines and reviewing alerts
What is reliability and availability?
Reliability - likelihood of services performing predictably
Availability - percentage of time the system is fully operational
Reliability is a subset of availability
availability % = (uptime seconds - downtime seconds) / uptime *100