DevOps For Everyone: Bringing DevOps Success to Every App and Every Role in your Organization
- 1. Bringing DevOps Success to Every App and Every Role in Your Organization Siva Rama Krishna DevOps for Everyone
- 2. 2 $> who am i @sivachunduru linkedin.com/in/chunduru slideshare.net/sivachunduru 20 years of diversified experience Developer Analyst Architect Evangelist Business Development Presales Consultant Product Manager
- 3. Evolution of Development and Deployment Application InfrastructureDeployment and PackagingApplication ArchitectureDevelopment Process HostedVirtual ServersN-TierAgile ~ 2000 Plan Release Build Code Test Operate Monito r Deploy DevOps Microservices Containers Cloud~ 2010 Now Waterfall Monolithic Physical Server Datacenter~ 1980 ~ 1990 3
- 4. Pillars of Current Digital Transformation Plan Release Build Code Test Operate Monito r Deploy DevOps Microservices Containers Cloud The ROI of Microservices, Containers and DevOps can be profound on Cloud, no matter an organization’s industry. 4
- 5. 02 DevOps – Organizational Roles 03 DevOps – Fitment to different Applications 04 DevOps – Measure it 05 Q & A 01 DevOps – What is it and it’s pipeline?
- 6. 02 DevOps – Organizational Roles 03 DevOps – Fitment to different Applications 04 DevOps – Measure it 05 Q & A 01 DevOps – What is it and it’s pipeline?
- 7. 7 DevOps Movement Crystallized at Flickr http://www.slideshare.net/jallspaw/10-deploys-per-day-dev-and-ops-cooperation-at-flickr As of 2009 • 51 total employees • 10 deploys per day • 3 billion photos stored • 40k photos viewed/sec.
- 8. DevOps … not a market, but a tool-centric philosophy that supports a continuous delivery value chain 8 … is the practice of operations and development engineers participating together in the entire service lifecycle, from design through the development process to production support https://theagileadmin.com/what-is-devops
- 9. Plan Code Build Test Package Release Deploy Monitor Harness the delivery process Plan Code Build Test Package Release Deploy Monitor 9 Agile Continuous Integration Plan Code Build Test Package Release Deploy Monitor Continuous Delivery If Agile software development was the opening act to a great performance, Continuous Delivery is the headliner. - Kurt Bittner, Principle AnalystApp/Team Infra Cloud Infra Observability Kit
- 10. 10 Gartner DevOps Model - Catalog of DevOps Practices People Process Culture Technical Debt Test-driven Development Test-Driven Deployment Test Everything Instrument Everything Trust Culture Engineering Culture Collective Ownership Autonomous Teams Joint Meetings Learning Culture Job RotationFull-Stack Teams Site Reliability Engineers Minimum Viable Process Common Metrics Feature Flags Optimize Flow Chaos Monkey Value Streams Collaborative Culture Never Done Small Batches Automated Testing Minimum Viable Product Release Automation Automated Builds Canary Rollouts Fail Forward Version Everything Servant Leadership Platform Engineers Technology Continuous Monitoring Integrated Tool Chains Monitor Everything Infrastructure as Code Developer Self- Service Continuous Testing Continuous Integration ChatOpsOne-Step Build, Test, Deploy Continuous Delivery Feature Teams Starting point of DevOps could be from any of the categories: People, Process, Culture or Technology.
- 11. 02 DevOps – Organizational Roles 03 DevOps – Fitment to different Applications 04 DevOps – Measure it 05 Q & A 01 DevOps – What is it and it’s pipeline?
- 12. 12 They Support in Production A Small Team Owns the Product They Implement They Architect They Care They Fix Things Ownership is Key to Success Everyone involved in creating, architecting, developing, testing, operating and supporting the product owns the product
- 13. • Silos • Wall of Confusion • Lack of Ownership • Slower Change Management 13 • Sense of Ownership • No “Wall of Confusion” • Better Collaboration • Agile, Fast Rate of Change Organization Structure Typical Enterprise Organization Structure Head of IT Head of Operations Head of DBAs Head of Infrastructure Head of App Dev Head of QA Head of Development Cross-functional Product Teams Product Lead Architect Sys Admin DBA JavaScript Developer Developer QA Network Admin Storage Admin Graphic ArtistNoSQL Admin Business Analyst Security
- 14. Roles Activities Agile planning Software version control Automated builds Automated tests Automated deploys Issue tracking Collaboration 14 Tools JIRA Git, BitBucket, GitLabs Jenkins, CircleCI Junit, Selenium Jenkins, CircleCI JIRA Wiki, Slack DevOps For Software Development Architect Frontend Developer Backend Developer QA Graphic Artist Business Analyst Security Product Owner
- 15. Roles Activities Infrastructure as code Version control Key management services Containers Container orchestration Release Canary, Blue/Green Serverless 15 Tools Chef, Puppet, Ansible, Helm, Terraform (aaS) DevOps For Software Deployment Network Admin Security Storage Admin Release Engineer NoSQL Admin DBA System Engineer Sys Admin Functions, Cloud Events Istio Kubernetes Docker Key Management, Vault Git, BitBucket, GitLabs
- 16. Roles Activities Log collection, analysis and visualization System health monitoring, metrics, alarms and notifications Product usage metrics Collaboration and communication 16 Tools ElasticSearch, Kibana, Splunk, OCI Logging Prometheus, Grafana, Monitoring and Alarms, Notifications Service Business Intelligence Slack, Pager Duty, Emails, Wiki DevOps For Service Operations System Admin Security Storage Admin Network Admin DBA Developer Product Owner System Operator
- 17. 02 DevOps – Organizational Roles 03 DevOps – Fitment to different Applications 04 DevOps – Measure it 05 Q & A 01 DevOps – What is it and it’s pipeline?
- 18. Plan Code Build Test Package Release Deploy Monitor 18 DevOps Pipeline - Build Ant & Maven Gradle Node.js – npm, grunt, bower, gulp Cloud CLI / SDK Containers Shell
- 19. Plan Code Build Test Package Release Deploy Monitor 19 DevOps Pipeline - Test Unit Testing Web App Testing Static Code Analyzer
- 20. Plan Code Build Test Package Release Deploy Monitor 20 DevOps Pipeline - IaC Deployment plans Infra Provisioning Infra Configuration Infra Management Security
- 21. 21 A Winning DevOps approach – Pick the right application Systems of Innovation Systems of Differentiation Systems of Record Traditional DevOps + - Governance + - Change Systems of Record - typically ERP-type applications Systems of Differentiation - typically business-specific applications, often COTS applications with customization Systems of Innovation - typically new web-based, agile-development-focused applications
- 22. Polyglot / Cloud Native Applications • MicroServices • Distributed Computing • 12-Factor Applications • API-based Interoperability
- 23. Integration / Business Process Applications • Internal / External Compute? • Repo Creation/Pre-population? • Adapters / Connectors? • Human Approvals?
- 24. Machine Learning Apps Ops Data is Perfect for ML • User Experience Data • App/Infra Performance Metrics • Events/Alerts • Configurations • Transaction Payloads • Log Records UserExperience Application MiddleTier DataTier Virtualization Infrastructure Anomaly Detection Clustering Correlation Prediction
- 25. 02 DevOps – Organizational Roles 03 DevOps – Fitment to different Applications 04 DevOps – Measure it 05 Q & A 01 DevOps – What is it and it’s pipeline?
- 26. • Code Quality – MTBF – MTTR – Bug Escape Distance • Code Complexity – Impacts the speed • Unit Test Coverage – Lower values turns to risk • Commit Rates – Longer the rates, more code to be merged – can cause problems, delays, bugs • Environment Stability – Continuous Automated Tests – Combine Tests & System Monitoring – Real time Application Monitoring – Faster Platform Upgradation • Business Metrics – # of deployments – # of release candidates built – Time to production – Software stack under CD pipeline What would be improved over time? All Dev All Ops Side Side 26
- 27. 27 Represent Effectiveness and Cost of DevOps Success Application : ABCD Avg. monthly values prior to DevOps (P) Month 1 M2 … M12 YTD Avg. monthly values using DevOps (Q)Week 1 Week 2 Week 3 Week 4 Total Number of release candidates x releases y Average time from release to production x min commit to production y min Fastest time for a release candidate to production x min a commit to production y min
- 28. 28 Represent Effectiveness and Cost of DevOps Success Other Cost Elements that show impact on TCO are, Opex: Cost avoidance/optimization, doing more with what you have, fail fast and fail cheap. Capex: Improved utilization, cloud-based systems and convergence. Application : ABCD Effectiveness (Q-P) Month 1 Month 2 ……… Month 12 Avg. Yearly value # of releases (A) Avg. Hours saved to production (B) # of DevOps resources (C ) Avg. resource billing rate (D) TCO difference (A*B*C*D) Business Plan Develop Test Build Release End User 0 days 3 days 10 days 20 days 3 days 13 days 36 days 1 days 6 days 2 days 3 days 1 days Efficiency = 13 days / 36 days = 36.11%
- 29. Summary: DevOps for Everyone 29 Software Developer Release Manager Automation Architect Security Engineer Evangelist Test Engineer Distributed Computing Microservices COTS Apps* 12-Factor Apps Systems of Innovation
- 30. 30
Editor's Notes
- #11: There is no official catalog of DevOps practices. As such, individual organizations must not only define what DevOps is to them, but also identify the activities that come into the DevOps focus. This figure is Gartner's initial attempt at providing some "bounded rationality" to the question: "What does DevOps consist of?“ …organization style attributed to Google, an SRE team = "… combines software development, networking and system engineering expertise to build and run large scale, massively distributed, fault tolerant software systems and infrastructure.“ …organization style modeled after Amazon.com's "two pizza" teams, a Feature team = “….have an integrated team of developers, quality assurance (QA), production engineers, and program or product managers, typically numbering between six and 10 individuals in total” ChatOps —ChatOps is an approach generally credited to GitHub. It is the pattern of automating operations tasks with an automated chat robot, or "bot," from within a chat room. Specifically, DevOps team members issue commands to perform code deployments, for example, via Internet Relay Chat (IRC) to a chat bot, which then executes them through scripts, etc. This enables the entire team to have real time collaboration, and ensures that everyone is aware of the current status of an operation in progress. A DevOps project can start from within any of the four basic areas or paths. It is less important to have a specific starting point than to just start and iterate toward the desired objective. In addition, it is not necessary to fully implement every practice within a category —organizations should stop at the point where they've determined that they've implemented the capabilities necessary for their business.
- #22: Apply a Layered Operations Strategy Reflecting the Reality That Not All IT Services Have the Same Support Requirements. Different applications cause different rates of change versus stability. Gartner’s ‘Pace-Layered Application Strategy’ outlines a way to categorize applications into: 1. Systems of record (typically ERP-type applications) 2. Systems of differentiation (typically business-specific applications, often COTS applications with customization) 3. Systems of innovation (typically new web-based, agile-development-focused applications) Stabilize IT Operations Investment in Systems of Record, and Begin to Shift IT Management Funding to Systems of Innovation and Systems of Differentiation That Will Drive Greatest Incremental Value for the Business DevOps as a concept is well aligned with projects with high degrees of uncertainty, so exhaustive planning is not going to be optimal.
- #27: Measuring Availability / Uptime Measuring the Business Differentiation How quickly we can deploy business differentiation to the market? # of deployments New functionality