Docker Meetup #50: Introduction to InfraKit, Swarm Mode & Load Balancing
Details
Please join us for Docker's 50th meetup! We will learn about Docker’s new toolkit, InfraKit! We also have exciting talks on Swarm Mode and Load Balancing.
Agenda:
6:00pm - 6:15pm - Arrival and Refreshments
6:15pm - 7pm - Talk #1: Introduction to InfraKit, by Bill Farner (https://twitter.com/wfarner?lang=en) & David Chung (https://www.linkedin.com/in/davidchungsf)
7pm - 7:45pm - Talk #2: Using Docker Swarm Mode and Healthchecks to Deploy Applications Without Loss by Nishant Totla (https://twitter.com/nishanttotla?lang=en) and Dongluo Chen (https://twitter.com/dongluochen?lang=en)
7:45pm - 8pm - Talk #3: How to Make Containers Discovery & Influence Loadbalancers by Roman Naumenko (https://twitter.com/naumenko_roman?lang=en)
8pm - 8:15pm - Questions, networking and close
Talk #1: Introduction to InfraKit
Abstract:
InfraKit is a new toolkit from Docker for creating and managing declarative, self-healing infrastructure. In this talk, we will describe the design of InfraKit, the roles of different plugin types, and demonstrate a few introductory examples of using InfraKit.
Speaker Biographies:
David Chung (https://www.linkedin.com/in/davidchungsf) is a software engineer working on infrastructure automation at Docker. Previously he co-founded Conductant, Inc. to focus on cluster orchestration and automation. David has worked on products from large Internet services to enterprise software. Before Conductant, David advised companies on adoption of container and orchestration technologies like Docker and Apache Aurora. Before that, he was a principal software engineer and tech lead on Zynga's successful Farmville franchise, and a software engineer at Google on products like AppEngine, Secure Data Connector, Checkout, and storage backend for Google Apps.
Bill Farner (https://twitter.com/wfarner?lang=en) is a Software Engineer specializing in distributed systems and cluster orchestration. He was a creator of the Apache Aurora project, and led the team using Aurora to orchestrate Twitter's infrastructure. He previously worked at Google on production engineering and storage. Currently, he works at Docker on Infrakit.
Talk #2: Using Docker Swarm Mode and Healthchecks to Deploy Applications Without Loss
Abstract:
Docker swarm mode enables users to manage their applications with service primitives. In this talk we demonstrate how to do service upgrades without impacting your application. The Healthcheck feature provides health indication for a container. Coming up in Docker 1.13 release, Docker Swarm can connect healthcheck result with load balancer to implement no-loss service upgrade.
Speakers Biographies:
Nishant Totla is a software engineer at Docker, and works on the core open source team. He is currently working on Docker SwarmKit and Docker Swarm. Prior to Docker, he was a PhD student at UC Berkeley, doing research on programming languages. In his spare time, he enjoys long-distance running, biking, and other outdoor activities. Nishant tweets at @nishanttotla (https://twitter.com/nishanttotla).
Dongluo Chen (https://twitter.com/dongluochen?lang=en) is a software engineer at Docker focusing on orchestration and container development. Before Docker he was software engineer manager at Microsoft Azure building and automating global data centers. He worked at France Telecom (Orange) and the Ohio State University as research scientist in networking area.
Talk #3: How to Make Containers Discovery and Influence Loadbalancers
Abstract:
Docker allows easy composition of multiple services on a developer machine. However, services communication via loadbalancers is a problem that often overlooked. This presentation shows how to run multiple services automatically discovered and registered with contemporary loadbalanced: HAProxy, Nginx, Traefik or Fabio. Simple compose files used to achieve this.
Speaker Biography:
Roman Naumenko (https://twitter.com/naumenko_roman?lang=en) is a DevOps Engineer at Gliffy working on fun stuff in the cloud. He's originally from Russia, having graduated from Obninsk Institute for Nuclear Power Engineering, where he learned how to operate things safely. Before that he spent a couple of years in a Military University, learning how to run things reliably. Roman missed the chance to become a PhD while working on epidemiological analysis of radiation effects among Chernobyl Liquidators - but he learned how to work with statistical data, risks and probabilities. Roman worked for Blackberry on large scale logging systems, hardware certification and other interesting stuff. Incredibly, he still prefers Blackberry to any other smartphone. He loves the startup environment and culture, thinks San Francisco is the best place on Earth and is persuaded that infrastructure automation and distributed systems are sexy.
Docker Meetup #50: Introduction to InfraKit, Swarm Mode & Load Balancing