Past Meetup

Containers, Functions, Serverless & Stateful Apps on Kubernetes

This Meetup is past

114 people went

Location image of event venue

Details

6:00pm: Networking, food & drinks provided by Oracle
6:30pm: Intro by Codefresh
6:45pm: Talks begin. See details below

πŸ—“οΈ Please RSVP. Space is limited.
β€—β€—β€—β€—β€—β€—β€—β€—β€—β€—β€—β€—β€—β€—β€—β€—β€—β€—β€—β€—β€—β€—β€—β€—β€—β€—β€—β€—β€—β€—β€—β€—β€—β€—β€—β€—β€—β€—β€—β€—β€—β€—β€—β€—β€—β€—β€—β€—β€—β€—β€—β€—β€—β€—β€—β€—β€—β€—β€—β€—β€—β€—β€—β€—β€—β€—β€—β€—β€—β€—β€—β€—β€—β€—β€—β€—β€—β€—β€—β€—β€—β€—β€—β€—β€—β€—β€—β€—β€—β€—β€—β€—β€—β€—β€—β€—β€—

☞ TALK 1: Containers, Functions & Serverless - The Fn Project

SUMMARY: While containers have changed the way we manage and deploy software, functions are changing how we think about it and build it. Developers just want to write and deploy their code. But typically, before you can deploy, you need to allocate the necessary infrastructure including provisioning machines, configuring storage, configuring networking, and so onβ€”all just to run even one line of code. Fortunately, serverless comes to the rescue!

This talk will explore how Containers, Functions, and Serverless are changing software development, and how the Fn project combines these concepts to deliver a serverless framework that eliminates infrastructure concerns entirely. Built on Docker, Fn includes support for writing functions in popular programming languages. All you need is Docker and you're ready to go! With Fn's function==container approach we'll show how easy it is to deploy pre-existing components as functions.

SPEAKER: Varun Madan is the engineering lead for Oracle Functions, a service based on Fn Project (fnproject.io). Over the past 2 years, he has been building container native applications and services at Oracle. Before this, he spent 8 years at Amazon Web Services building different parts of EC2. Varun is @madvar on twitter.

β€—β€—β€—β€—β€—β€—β€—β€—β€—β€—β€—β€—β€—β€—β€—β€—β€—β€—β€—β€—β€—β€—β€—β€—β€—β€—β€—β€—β€—β€—β€—β€—β€—β€—β€—β€—β€—β€—β€—β€—β€—β€—β€—β€—β€—β€—β€—β€—β€—β€—β€—β€—β€—β€—β€—β€—β€—β€—β€—β€—β€—β€—β€—β€—β€—β€—β€—β€—β€—β€—β€—β€—β€—β€—β€—β€—β€—β€—β€—β€—β€—β€—β€—β€—β€—β€—β€—β€—β€—β€—β€—β€—β€—β€—β€—β€—β€—

☞ TALK 2: Top 5 Considerations for Running Stateful Apps on Kubernetes

SUMMARY: Docker containers are great for running stateless microservices, but what about stateful applications such as databases and persistent queues? Kubernetes provides the StatefulSets controller for such applications that have to manage data in some form of persistent storage. While StatefulSets is a great start, a lot more goes into ensuring high performance, data durability and high availability for stateful apps in Kubernetes. Following are 5 best practices that developers and operations engineers should be aware of.
1. Ensure high performance with local persistent volumes and pod anti-affinity rules.
2. Achieve data resilience with auto-failover and multi-zone pod scheduling.
3. Integrate StatefulSet services with other application services through NodePorts & LoadBalancer services.
4. Run Day 2 operations such as monitoring, elastic scaling, capacity re-sizing, backups with caution.
5. Automate operations through Kubernetes Operators that extend the StatefulSets controller.

We will demonstrate how to run a complete E-Commerce application powered by YugaByte DB, when all services are deployed in Kubernetes.

Speaker Bio: Bogdan Matican is a Founding Engineer at YugaByte, working across all aspects of the product, from core database features to the enterprise administration console. He has been working in or around distributed systems for over 5 years, across Facebook (HBase, blobstore, stream processing), UCar (monitoring, data analytics) and research (security, cluster management).