DevOps and Software Engineering Overview

DevOps Implementation

• DevOps is a combination of architecture, technical practices, and cultural norms to increase the speed and safety of delivering applications and services.
• DevOps integrates and automates software development and IT operations. This improves and shortens the system development life cycle, enabling continuous delivery with high software quality.

Software Engineering Timeline

• Key dates and events in software engineering history are listed, like the invention of Pascal and the creation of Unix.

Traditional Development

• The inventors' main responsibility is designing and developing new functionalities.
• Features and functionality are developed within a dedicated development environment.
• New products are periodically delivered to operators with relevant instructions.
• Feedback from operators is occasionally incorporated for future deliveries.
• Inventors are rewarded for innovative features in development.

Traditional Operations

• Operations personnel maintain and manage the system's operational efficiency.
• Receiving new products from developers, installing, and running production systems is the core responsibility of operations tasks.
• Operations staff monitor system performance and identify problems, failures, and deployment issues.
• Feedback from Operations is offered to developers for future enhancements.
• Operations are penalized for system downtime.

The History of Separation of Dev and Ops

• Specialization in development (Dev) and operations (Ops) stems from the increasing complexity of computer systems and the accelerating pace of technology.
• The growing demand for new features, large amounts of data, and complex calculations contribute to this specialization.
• Increasingly abstract architectures and more specialized tools further solidify the separation.

Conflicts Caused by Different Goals

• Conflicts arise between development and operations teams due to differing priorities and goals.
• Development emphasizes pushing changes, while operations may resist changes to maintain stability.

Dev & Ops work Together

• Feedback loops between developers and operations personnel are critical for better understanding and cooperation.
• Real-time metrics from operations allow developers to learn from real-world conditions.
• Real-time metrics from development help operations anticipate needs and provide early input.
• Collaboration between cross-functional teams ensures end-to-end system development with infrastructure and software code.
• Shared goals of feature delivery and system stability are essential.

DevOps: Convergence of Dev and Ops

• DevOps integrates and applies best practices from manufacturing and leadership to the IT value stream.
• It's influenced by methodologies like Lean, the Toyota Production System, resilience engineering, safety culture, and human factors.
• Value stream mapping, Kanban boards, and total productive maintenance were initially codified by the Toyota Production System.
• The Lean enterprise Institute research led to applications of Lean principles in other sectors, such as services and healthcare.

Toyota Production System (TPS)

• Just-in-time production, emphasizing making only what's needed, when it's needed.
• Jidoka (Autonomation) emphasizes automation with a human touch.

Toyota Production System (TPS) (Con't)

• Integrated socio-technical system developed by Toyota to efficiently organize manufacturing and logistics; customers' relations are minimized, so is cost and waste.
• Prioritizes intelligent work strategies, minimizing inventory, increasing cash flow, and reducing physical space, thus facilitating efficient delivery.

Lean

• Continuous improvement process stemmed from manufacturing practices.
• Encourages respect for people.
• Five key principles emphasize Defining value, Value stream mapping, Creating flow, Using a pull system, and Pursuing perfection.
• Identifying and eliminating waste.

Wastes in Software Development

• Activities that cause delays, decreasing value as time progresses.
• Examples include partially finished work, extra steps, features not needed by customers, additional complexity, context switching.
• Ways to reduce waste include streamlining workflows, eliminating unnecessary features, reducing task switching, and improving the management of dependencies.

The Technology Value Stream

• The conversion process of a business idea into a technology-enabled service or feature.
• Value is generated when the service or feature operates in a production environment.
• Lead time begins with the request and finishes on fulfillment. Process time captures the time to process a customer request while excluding waiting time.

Break the Silos: Communication, Collaboration, Integration

• Developers play a crucial role in maintaining system stability by responding promptly to bugs and deployment issues.
• Operators are key to accelerate feature delivery. Their anticipatory roles in production and trust in developers are crucial.
• Increased collaboration and communication to resolve bugs and failures are key.

Break Silos: Dev, Ops & Business

• Development teams must incorporate business and customer needs.
• Operations understand better how to support business goals.
• Business comprehension of changing capabilities in features, functionality, and operations is essential.

Business Demand: Continuously Deliver Valuable Software

• Businesses rely on IT for timely deployment of new features, fast time-to-value, new technology incorporation, adapting to customer demands and decreasing delivery times, maintaining low Mean Time To Repair (MTTR).

Accelerate State of DevOps 2024 Report

• Provides research and data from extensive surveys measuring software delivery performance.
• Key metrics such as lead time for changes, deployment frequency, change failure rate, and failed deployment recovery time are emphasized.
• Throughput and stability are discussed as important facets of software delivery.

The Three Ways of DevOps By Gene Kim

• First way: fast, left-to-right work flow from development to operations and customers
• Second way: fast and constant feedback flowing from right to left across all stages in the value stream
• Third way: generative, high-trust culture that supports a dynamic, disciplined, scientific approach. This facilitates experimentation, risk-taking, organizational learning from both successes and failures.

The first Way: The Principle of Flow(1/5-5/5)

• Make work visible: using visual boards, electronic cards for work representation.
• Limit work in Progress (WIP); using limits to avoid multitasking.
• Reduce batch sizes: to enable faster lead-times and error detection.
• Reduce the number of handoffs: reducing communication steps and conflicts.
• Identify and elevate constraints; eliminating hardships and waste in the value stream.

The Second Way: The Principle of Feedback(1/6 to 6/6)

• See problems as they occur; creating feedback and feedforward loops in the system
• Different types of feedback in software development; such as developer, CI, exploratory, acceptance, stakeholder and user
• Each feedback type takes different amounts of time.
• Swarm and solve problems to build new knowledge, preventing problems from spreading by containing them and fixing them at an early stage.
• Push quality improvement closer to the source; by having developers accountable for quality.
• Enable optimizing for downstream work centers: by understanding external and internal customers, and next-step downstream priorities.

The Third Way: The Principles of Continual Learning and Experimentation

• Enabling organizational learning and a safety culture.
• Define three cultures (pathological, bureaucratic, and generative) based on how organizations handle information and failures.
• Institutionalize the improvement of daily work, improving work by dedicating time to technical debt, defects, and problem areas in code or environments.
• Transform local discoveries into global improvements; creating systems to share and learn from successes, failures and best practices
• Inject resilience patterns into daily work; considering systems and workflows to improve capacity and minimize risks.
• Leaders reinforce a learning, complementary, and supportive culture where problems can be solved efficiently together.

DevOps vs Agile

• Key differences between Agile, a software development methodology, and DevOps, a collaborative culture for development and operations.

DevOps Lifecycle

• The iterative feedback loop of the DevOps process: plan, build, test, code, release, deploy, operate, and monitor.

This quiz covers the essential aspects of DevOps implementation and traditional software engineering practices. Learn about the key components that improve software delivery, significant historical events, and the responsibilities of inventors in the development process. Test your knowledge on how DevOps integrates software development and IT operations.