Talks
Speakers
Events
Topics
Sign in
Home
Talks
Speakers
Events
Topics
Leaderboard
Use
Analytics
Sign in
Suggest modification to this talk
Title
Description
RailsConf 2018: Some Funny Things Happened on The Way to A Service Ecosystem by Chris Hoffman So this one time we ran Rails services over AMQP (instead of HTTP). You may be thinking “Why? Was that their first mistake?” It was not. From handling shared models poorly to deploying & operating services with inadequate tooling, we've made...so many mistakes. I'm glad we did; they helped us grow. We have patterns & practices for everything we’ve seen so far, from API contracts to distributed consensus protocols. Over this talk’s 40 minutes, you'll learn things it took us 3 years to grasp. Whether you have a monolith or pine for when you did, you'll be more prepared and better armed than we were.
Date
Summarized using AI?
If this talk's summary was generated by AI, please check this box. A "Summarized using AI" badge will be displayed in the summary tab to indicate that the summary was generated using AI.
Show "Summarized using AI" badge on summary page
Summary
Markdown supported
In the video, titled "Some Funny Things Happened on The Way to A Service Ecosystem," Chris Hoffman discusses the transition from a monolithic architecture to a service-based ecosystem, sharing experiences from his work at OpTaurus, an e-commerce company. He reflects on the challenges faced during the shift, emphasizing the importance of learning from mistakes in adopting microservices. The presentation covers several key insights: - **Monolith to Services**: Hoffman introduces the concept of moving from a monolith to a service architecture, sharing background about OpTaurus and illustrating the difficulties encountered in implementing services, such as issues with authentication (auth) and inter-service communication. - **Learning from Mistakes**: He emphasizes that their early attempts were fraught with errors, particularly in managing shared models, query optimizations, and handling latency issues between services, ultimately leading to project cancellations. - **Data Sharing Challenges**: Hoffman discusses how data sharing between services posed serious complications, advocating for a model where services broadcast information rather than synchronize data, which often led to data inconsistency. - **Service Development Insights**: The speaker advises starting with small, manageable service projects that add functionality instead of extracting from the monolith, highlighting the need to ensure project visibility to gain organizational support. - **Technical Approaches**: Specific strategies for implementing services are outlined, including a recommendation to avoid complex, asynchronous protocols in favor of simpler HTTP-based communication. - **Operational Ownership**: Hoffman asserts that developers should take ownership of operations in a microservices environment, advocating for conventions that empower engineers to effectively manage production incidents. In conclusion, Hoffman shares several important takeaways for those looking to implement microservices, including the need to: - Avoid Active Record callbacks to simplify data extraction. - Focus on projects that deliver noticeable benefits to gain political support within organizations. - Establish strong operational practices and conventions to streamline deployment and maintenance of services. This talk aims to prepare teams for the potential pitfalls they may encounter on their journey to a service ecosystem, providing actionable insights gained from real-world experiences at OpTaurus.
Suggest modifications
Cancel
