Mark Richards

Microservices Architecture and Design

9:00 AM MDT

Microservices is one of the latest software architecture styles that promises to deliver benefits such as ease of testing, fast and easy deployments, fine-grained scalability, architectural modularity, and high overall agility. Unfortunately, coupled with these benefits comes a lot of complexity. This product-agnostic architecture class provides you with a full understanding of the Microservices architecture style and what hybrids and alternatives exist, which will help guide you in making the right architecture and design decisions for your organization. In this workshop I will start by discussing the core concepts of the Microservices architecture style, the primary benefits and drawbacks of Microservices, how Microservices differs from SOA, and also how to create hybrid architectures. Taking a deeper dive, I will then talk about the hard stuff in microservices, including how to identify services, various service template design strategies, event-driven services, service communication patterns, and distributed transactions. Finally, I'll talk about the hardest part of microservices, which is how to distribute your monolithic data.

Workshop Outline:

Microservices Core Concepts

• Topology overview
• Distributed architecture considerations
• Service components and granularity
• Bounded context concepts and challenges
• Microservices drivers and advantages
• API layer design
  
  
  

  Hybrid Architectures and Migration Techniques

• Service-based architecture
• Microservices migration patterns
• Architectural modularity
  
  
  

  Microservices Design Techniques

• Service identification
• Service design
• Event-driven services
• Service communication patterns
• Distributed Logging
• Error handling techniques
  
  
  

  Microservices Data Considerations

• Distributed Data Challenges
• Creating Data Domains
• Dealing with Common Data
• Deferred Data Migration
• Distributed Transactions
• Eventual Consistency Patterns

Reactive Architecture Patterns 1

8:30 AM MDT

Reactive architecture patterns allow you to build self-monitoring and self-healing systems that can react to both internal and external conditions without human intervention. How would you like to design systems that can automatically grow as the business grows, automatically handle varying load (cyber Monday?), and automatically handle (and repair) internal and external errors, all without human interaction? I'll show you how to do this with your current technology stack (no special languages, tools, frameworks, or products). In this two-part session I will leverage both slides and live coding using Java and RabbitMQ to describe and demonstrate how to build reactive systems. Get ready for the future of software architecture - that you can start implementing on Monday.

Part 1 Agenda:

• Reactive systems introduction
• Channel Monitor Pattern (self-awareness)
• Consumer Supervisor Pattern (scalability and elasticity)
• Workflow Event Pattern (error handling)

Reactive Architecture Patterns 2

10:30 AM MDT

Reactive architecture patterns allow you to build self-monitoring and self-healing systems that can react to both internal and external conditions without human intervention. How would you like to design systems that can automatically grow as the business grows, automatically handle varying load (cyber Monday?), and automatically handle (and repair) internal and external errors, all without human interaction? I'll show you how to do this with your current technology stack (no special languages, tools, frameworks, or products). In this two-part session I will leverage both slides and live coding using Java and RabbitMQ to describe and demonstrate how to build reactive systems. Get ready for the future of software architecture - that you can start implementing on Monday.

Part 2 Agenda

• Thread Delegate Pattern (scalability)
• Threshold Adjust Pattern (automatic configuration and self-learning)
• Producer Control Flow Pattern (self-healing)

Streaming Architecture Using Kafka (UberConf)

1:00 PM MDT

There are many different uses for Apache Kafka. It can be used as a streaming broker, event broker for transactional data, and even a database. This session is about understanding streaming architecture and how to implement it using Apache Kafka. I start this session by talking about some of the streaming architecture patterns, then dive into how Apache Kafka works using the Core API. Using live coding examples in Apache Kafka, I also talk about the differences between Kafka and regular messaging (RabbitMQ, ActiveMQ, etc.) and when you should use each. I end this session by putting everything together, showing an actual streaming architecture using Kafka within a Microservice ecosystem for gathering various metrics for business and operational monitoring and reporting.

Agenda:

• Streaming architecture patterns
• Kafka overview
• Kafka producers and consumers
• Kafka vs. standard messaging
• Streaming examples in a microservices ecosystem

Microservices AntiPatterns

2:45 PM MDT

Even though teams are gaining more experience in designing and developing microservices, nevertheless there is still a lot to learn about this highly distributed and somewhat complicated architecture style. Unfortunately, lots of microservices anti-patterns and pitfalls emerge during this learning curve. Learning about these anti-patterns and pitfalls early on can help you avoid costly mistakes during your development process. While anti-patterns are things that seem like a good idea at the time and turn out bad (see martinfowler.com/bliki/AntiPattern.html), pitfalls are those practices that are never a good idea at the time - ever. In this session I will cover some of the more common anti-patterns you will likely encounter while creating microservices, and most importantly describe some of the techniques for avoiding these anti-patterns.

Agenda

• Function-Driven Identification AntiPattern
• I Was Taught to Share AntiPattern
• The Timeout AntiPattern
• Hop on the Bus AntiPattern
• Data-Driven Migration AntiPattern

Microservices Pitfalls

4:30 PM MDT

Even though teams are gaining more experience in designing and developing microservices, nevertheless there is still a lot to learn about this highly distributed and somewhat complicated architecture style. Unfortunately, lots of microservices anti-patterns and pitfalls emerge during this learning curve. Learning about these anti-patterns and pitfalls early on can help you avoid costly mistakes during your development process. While anti-patterns are things that seem like a good idea at the time and turn out bad (see martinfowler.com/bliki/AntiPattern.html), pitfalls are those practices that are never a good idea at the time - ever. In this session I will cover some of the more common pitfalls you will likely encounter while creating microservices, and most importantly describe some of the techniques for avoiding these pitfalls.

Agenda

• Grains of Sand Pitfall
• Where's the Acid Pitfall
• Logging Can Wait Pitfall
• Jump on the Bandwagon Pitfall
• Give it a Rest Pitfall

The Art of Problem Solving

8:30 PM MDT

As Tech Leaders, we are presented with problems and work to find a way to solve them, usually through technology. In my opinion this is what makes this industry so much fun. Let's face it - we all love challenges. Sometimes, however, the problems we have to solve are hard - really hard. So how do you go about solving really hard problems? That's what this session is about - Heuristics, the art of problem solving. In this session you will learn how to approach problems and also learn techniques for solving them effectively. So put on your thinking cap and get ready to solve some easy, fun, and hard problems.

Agenda:

• the checklist
• visual representations
• you know what happens when you assume…
• you gotta start somewhere
• the world is flat
• hey! keep it clean!
• perception can be reality

Architectural Modularity

9:00 AM MDT

It seems like all we talk about these days is making our architectures more modular. Buy why? In this session I will discuss the drivers and reasons why it is essential to move towards a level of modularity in our architectures. I will discuss and show real-world use cases of distributed modular architectures (specifically microservices and service-based architecture), and then discuss in detail the core differences between microservices and service-based architecture and when you should consider each. I'll end the talk by discussing the most effective way of migrating to modular distributed architectures.

Agenda:

• Drivers for modularity
• Modular distributed architectures
• Microservices vs. service-based architecture
• The move towards modularity

Analyzing Software Architecture (archived)

11:00 AM MDT

One of the expectations of any software architect is to analyze the current technology environment and recommend solutions for improvement. This is otherwise known as continually assessing architecture vitality. Too many times software architects fail to regularly perform this task, leading to emergency refactoring efforts to save a troubled system from failure. The question is, what does it mean to assess an application architecture? In this session we will explore static analysis metrics and tools and techniques for leveraging those metrics for determining structural decay. Using a real-world large-scale application, I'll show you how to leverage code metrics to find (and fix) structural decay before it gets you into trouble.

Agenda

• Architecture and structural decay
• Architecturally significant code metrics and what they mean
• Interpreting results and tracking metrics
• Code analysis tools and how to use them
• Real-world hands-on architecture analysis using tools and metrics

Integration Architecture: Concepts and Patterns

1:30 PM MDT

Very few applications stand alone anymore. Rather, they are combined together to form holistic systems that perform complex business functions. One of the big challenges when integrating applications is choosing the right integration styles and usage patterns. In this session we will explore various techniques and patterns for application integration, and look at what purpose and role open source integration hubs such as Camel and Mule play in the overall integration architecture space (and how to properly use them!). Through actual integration scenarios and coding examples using Apache Camel you will learn which integration styles and patterns to use for your system and how open source integration hubs play an part in your overall integration strategy

Agenda:

• Integration Styles: pros, cons, and when to use
• Integration and Abstraction
• Integration Hubs: use cases, failover, federation, team ownership
• Common integration patterns