Microservices Architecture and Migration Strategies

9/20/2022

This post introduces the concept of microservices architecture as a way to jumpstart DevOps practices, highlighting benefits like improved scalability, fault isolation, and faster feature delivery. It outlines key steps for building with microservices, including identifying decomposable application aspects based on business domains and determining essential metrics to monitor (CPU, memory, API response time, error rate). It also details best practices for deployment and management, such as infrastructure automation and consumer-driven contract testing to ensure API compatibility. The post emphasizes the synergy between microservices and DevOps, enabling independent team management and accelerated feedback loops.

6/17/2019

This post introduces and explains the microservices architecture, contrasting it with monolithic architectures. It details three distinct strategies for migrating to microservices: the Strangler Method, the Lego Strategy, and the 'nuclear option'. It also outlines key goals for a successful transition (Architecture, Automation, Extensibility, Proactivity) and provides an example of Verizon's success. The post concludes by highlighting the role of modern tools like Kubernetes and CI/CD in simplifying microservices adoption.

2/14/2019

This post details how Verizon Connect migrated from a monolithic, Java-based application architecture to a microservices architecture using GitLab. They achieved a reduction in datacenter rebuilds from 30 days to 8 hours by focusing on architecture, automation, extensibility, and future-proofing. GitLab was chosen as a single solution to replace disparate tools like Jira, BitBucket, and Jenkins, providing integrated code review, CI/CD, issue tracking, source code management, audit management, and ChatOps capabilities. This enabled them to achieve deployment flexibility, platform agnosticism, streamlined processes, and focus developers on differentiating tasks.

11/26/2018

This post discusses the challenge of managing an increasing number of microservice repositories and the need for an integrated solution. It highlights GitLab's approach of providing 'everything under one roof' to streamline the DevOps lifecycle, enabling faster deployments and improved stability. It also mentions upcoming features like group-level Kubernetes clusters, a global Docker registry browser, and the ability to define multiple pipelines, building on existing multi-project pipeline capabilities and Kubernetes integration.

8/16/2016

This post introduces the concept of microservices architecture, defining it as a way to design software applications as smaller, independent, cloud-based services rather than monolithic data centers. It details the benefits such as system-wide stability, ability to scale teams independently, diverse technology adoption per service, and improved architecture through enforced module boundaries. It also outlines drawbacks including increased latency, complexity of distributed systems, multi-step rollouts, the need to accommodate failure modes, infrastructure complexity, and the management of multiple projects/repositories. The post then connects these concepts to GitLab's value proposition for microservices, highlighting its integrated approach to CI/CD, issue tracking, and project management for independent services, and discusses when microservices are appropriate for teams.