CNCF-hosted Co-located Events North America 2024

This presentation addresses the high latency issues during workflow submissions via Argo Workflows UI and Argo Workflows SDK, focusing on the integration between client applications and Argo Workflows. It explores the inefficiencies caused by the lack of WorkflowTemplate caching and offers solutions to mitigate latency, enhance performance, and reduce load on the Kubernetes API server. Attendees will learn practical optimization strategies to significantly improve workflow submission times and overall system responsiveness.

In this talk, we will explore how to effectively manage inter Applications dependencies in Argo CD. Argo CD's Application Custom Resource Definition allows you to group Kubernetes Manifests logically, treating them as a single entity. While Applications are designed to be autonomous, this can pose challenges in a microservices architecture where components are isolated into separate Applications. This session will discuss patterns, best practices, and lessons learned to manage dependencies using the current capabilities of Argo CD. The session will begin with an overview of the Application CRD and its role as the atomic unit in Argo CD, highlighting how it enables the logical grouping of Kubernetes objects. We will then discuss Application characteristics and how it impacts microservices, using real-world examples. The we'll go over the current limitations in setting up Application dependencies in Argo CD, followed by strategies and patterns for managing inter-Application dependencies

Argo Workflows, a K-native DAG-style workflow engine for Kubernetes, is a powerful tool for orchestrating complex workflows. Additionally Java is the most popular language for developers in complex organisations. However, using Argo with the Argo Java SDK can be challenging due to limited examples and unclear documentation. This session aims to provide a comprehensive, step-by-step guide on interacting with Argo Workflows using Fabric8’s Custom Resource Definitions (CRD) in Java. Argo Workflows are implemented as 'Custom Resources' in Kubernetes. By leveraging the Fabric8 library’s Custom Resource Definitions, developers can seamlessly interact with Argo Workflows through the Kubernetes API. Join us to learn how to harness the power of Argo Workflows in Java, simplifying your workflow management on Kubernetes.