Containers & Kubernetes Architecture
Tuesday, November 10, 2020
When you combine the efficiency of containers, agility of serverless and flexibility of event-driven services, you end up with a more reusable, interoperable and scalable architecture with minimal management overhead.
In this talk, we’ll explore the open-source Knative Eventing and its managed version Cloud Run. We’ll explore what they provide for event-driven serverless containers and we’ll deep dive into some real-world reference architectures.
At the end of this session, you’ll have a solid understanding on how Knative and Cloud Run can power your event-driven apps.
We know that costs for public cloud services can quickly get out of hand, especially as we adopt, automate, and scale CI/CD practices. Without cloud costs visibility the invoice would be a black box. Still for many users, desperately seeking to reduce their cloud costs, the root cause of these costs is hidden within subtleties of cloud resources. This makes it difficult for teams to plan work, test their services, and manage costs. Join this session to learn how to manage development resources to reduce cloud costs. Attendees will learn how to optimize Kubernetes workloads.
The goals of this talk is to explain the different between two common continuous deployment strategies for Kubernetes clusters.
There are two common strategies to continuous deployment to Kubernetes-based environments. The first one is a push based model, where a deployment pipeline (e.g. Jenkins / Azure Devops /…) pushes new applications to a container cluster. A second approach is a pull based model (e.g. Flux / Argo ) where services running on the cluster will pull new application configuration into the cluster.
In this talk we'll discuss both approaches, and do a compare and contrast between them. We'll also see a demo of both approaches to continuous delivery on a Kubernetes Cluster.
Many Enterprises have existed long before Public cloud and Virtual Machines were a reality. They are heavily invested in their own data centers running on physical machines. They use traditional technologies to deploy software using Puppet or Ansible and they daydream about modern microservices architecture and envy the next blog post coming out of Google’s and Facebooks of the world. With Kubernetes, this has changed. We show to those enterprises, that its possible to run your software in the same infrastructure that was once only accessible to a select few companies. We also show , in the absence of VM’s and managed cloud offerings on bare metal it's still possible to manage Kubernetes on physical hardware and with a little investment in processes and monitoring, you can run a billion dollar empire and modernize your software to microservice architecture and follow modern cloud native processes like CI/CD and 12 factor app.
"Containers are the new ZIP format to distribute software" is a fitting description of today's development world. However, it is not always that easy and this talk highlights the development of Elastic's container strategy over time:
* Docker images: A new distribution model.
* Docker Compose: Local demos and a little more.
* Helm Chart: Going from demo to production.
* Kubernetes Operator: Full control with upgrades, scaling,...
Besides the strategy we are also discussing specific technical details and hurdles that appeared during the development. Or why the future will be a combination of Helm Chart and Operator (for now).