Alex Williams, founder and editor-in-chief of The New Stack, recently hosted a podcast at KubeCon + CloudNativeCon in Seattle with Alexis Richardson (@monadic), CEO of Weaveworks and Andrew Clay Shafer (@littleidea), Senior Director of Technology at Pivotal and discussed some of the essential tools and workflows that are needed to go cloud native.
Although many organizations have made the shift to cloud native, many are still wading through all of the tools, workflows and other organizational choices and decisions that come with making the change.
Importance of Istio and AWS’ AppMesh
The emergence of service meshes like Istio or AWS’ App Mesh helps control, secure and observe microservices running in a cluster. These technologies are increasingly seen by DevOps teams as an essential tool rather than just another tool on the long list of nice-to-have options available today for Kubernetes and microservices.
Istio is a dedicated infrastructure layer for all your service-to-service interactions in a microservices architecture. All actions in Istio are driven through declarative configuration files. Service meshes like Istio enable developers to manage microservice behavior completely through declarative files that are kept in Git with your code, enabling your teams to adopt GitOps workflows.
GitOps and Progressive Delivery Pipelines
GitOps is also a key operational model that empowers developers with the ability to quickly, and seamlessly make updates to microservices running on Kubernetes. Monitoring and observability throughout the process are essential to GitOps and running microservices at scale in Kubernetes.
GitOps works in tandem with a service mesh to help teams implement more advanced deployment strategies such as canary, feature flagging and blue/green and other types of progressive delivery pipelines.
Canary deployments with GitOps, and Istio
With a canary deployment for example, a change is pushed through the CICD pipeline to a subset of your users. A service mesh such as Istio, allows you to control which subset of your users receives the canary traffic. Because you are using GitOps, it’s easy to make changes, and promote the next version, by tagging it, and deploying it through the same pipeline. This is an iterative process that GitOps workflows helps you manage.
Learn more by listening to the podcast below: