March 1, 2020

299 words 2 mins read

Navigating in stormy waters: An approach to traffic management with Istio

Navigating in stormy waters: An approach to traffic management with Istio

History repeats itself. Some years ago, software engineers started to implement frameworks to ease the development of software applications. Laurentiu Spilca walks you through how microservices are currently delivered and what Istio can do for you in regard to traffic management.

Talk Title Navigating in stormy waters: An approach to traffic management with Istio
Speakers Laurentiu Spilca (Endava)
Conference O’Reilly Software Architecture Conference
Conf Tag Engineering the Future of Software
Location Berlin, Germany
Date November 5-7, 2019
URL Talk Page
Slides Talk Slides
Video

History repeats itself. Some years ago, software engineers started to implement frameworks to ease the development of software applications, and part of functionality was pulled out of the applications and added to frameworks. Today, the applications have become vast and turned into interconnected systems. Architectures became more complex with small components that communicate with each other; we called this microservices. One of the current challenges is choosing which level of abstraction to use when deploying such systems. We started with infrastructure and then containerized everything. The next step was finding out we need orchestration, and then we repeated the first step. We’ve taken functionality out of the framework and added it to the deployment and orchestration logic. Laurentiu Spilca explores the future of implementing traffic management in service-oriented architectures with Istio. He outlines how microservices are currently delivered: in containers and orchestrated with a tool like Kubernetes. He digs into what Istio is, what can you do with it in regard to traffic management, and how it actually does this. You’ll see two live coding examples, one with progressively rolling out functionalities like Istio and a description of implementation use cases for this (such as A/B testing and safe rollouts) and an example with fault injection and a description of implementation use cases (resilience testing).

comments powered by Disqus