Registration will begin at 8:45am. Please arrive promptly for opening announcements at 9:30am.

Containers are all the rage among developers and web companies, but they also represent two very substantial benefits to teams and larger organizations. First, they have the potential to dramatically accelerate the application lifecycle from software builds and testing to deployment and upgrades. Second they represent the first truly hybrid-approach to consuming infrastructure, allowing organizations to run the same workloads on any cloud, virtual machine or physical server. Together, they represent a very real opportunity for faster development and cheaper operations, without many of the limitations of existing PaaS platforms. In this session, we’ll introduce the concept of the private Container Service, an enterprise-grade platform that runs and manages containers across any computing platforms.

We're looking at a major revolution in Cloud Computing, the likes of Docker, CoreOS, Mesos, Kubernetes and many other technologies are not only changing the way we harness our computing power but are also changing the way that we talk about it. In this session we'll look at the new metaphors of computing through the lens of Kubernetes. We'll make use of the Kubernetes API and visualization libraries to get to a better understanding of the underlying mechanics of Kubernetes, we'll build clusters, schedule pods, group containers and expose services, and we'll visually demonstrate how they are all wired together.

Container control systems are often based on distributed-consensus stores like Consul or Etcd, but they impose overheads that may not be justified for every task. Learn how Weaveworks implemented IP Address Management using a conflict-free replicated data type (CRDT) to deliver improved availability, performance and flexibility.

LXD is a new container manager, developed on top of the great LXC library by the same development team as LXC itself. It's designed to be fast, safe by default, work transparently across the network and provide a great user experience, whether it be used on a single laptop or on a large cluster. This talk will go over what LXD is, why it came to be, what it's doing today and where we're taking it.

After we setup our initial Kubernetes+CoreOS based cluster on Azure using Weave, we asked ourselves the question: but what about performance? It turned out that different public clouds such as Azure and AWS but also the network implementation you choose for your Docker cluster has a significant impact on the networking performance of your cluster. In this talk I'll address how ran our performance tests on the different setups and the new libnetwork stack in the Docker experimental channel will be discussed as it offers new possibilities to easily swap the included network "batteries" by another implementation.

Kubernetes presents many challenges to the IT person that needs to deploy and administer the cluster and the containers on it. This session will cover how the open source ManageIQ project covers the lifecycle of Kubernetes and OpenShift, and allows a clear view on the system's various layers by showing what runs on each node, relationships between entities, all while still abstracting the different container technologies and terminology for the administrator.

At Improbable.io we are building a Reactive Infrastructure at scale. We are doing this by leveraging etcd and the DNS and making all aspect of our stack dynamic. In this talk I want to delve into how Improbable.io uses that setup and how it provides us with load-balancing, service discovery, monitoring and health checking, allowing us to build a complete multi-cluster stack in just 3 months.

Scrapbook provides an online learning environment and hands-on labs built on top of Docker. To create an interactive learning environment, Scrapbook offers personalised sandboxed environments as learning playgrounds. A side effect is that users can execute malicious code and hack the system from inside the container. In this talk, Ben will share the lessons learned of building Scrapbook, Docker security and the solutions put in place. This talk will give insight into: - Out of the box security with Docker - Docker and Linux security issues - Monitoring for malicious activity - What happens when it all goes wrong.

Secure software distribution is a hard problem. A testament to this fact are the thousands of different software update systems in use today, most of which vulnerable to a myriad of attacks that leave the end users potentially vulnerable to compromise.

With the explosion in popularity of containers, more and more software is being distributed in the form of a Docker image. This gives us the unique opportunity to leapfrog the status quo, and create a better way to ensure the security of software distribution.

Enter Notary, an application built at Docker that aims to make the internet more secure by making it easy for people to publish and verify content. Notary follows a flexible security framework called TUF (The Update Framework), allowing publishers to sign their content offline and manage their keys securely.

In this talk I will go over Notary, its security guarantees, TUF and how we’ve integrated it into Docker 1.8 to provide content trust.

Containers are revolutionizing the way we deploy and maintain our infrastructures, but monitoring and troubleshooting in a containerized environment can still be painful and impractical. Understanding even basic resource usage is difficult - let alone tracking network connections or malicious activity. This talk will cover the current state of the art for container monitoring and visibility, including pros / cons and live demonstrations of each method. Special emphasis will be put on sysdig, an open source troubleshooting tool authored by the presenter. Sysdig offers unprecedented container visibility without the need to break the “sanctity” of the container.

Containers are awesome! But storing and moving them is painful! It's slow! Containers are huge! There's little deduplication! No cryptographic versioning! Images are lost! Worst, we must depend on other services -- or having fast connection to them... No more! `starship` puts your containers on hyperspeed! Instead of old-school HTTP, `starship` uses shiny IPFS -- a new protocol, Git meets Bittorrent. With IPFS, `starship` (a) deduplicates all data between containers, (b) uses p2p download from fastest hosts, and (c) can limit connections to private cloud(s) or local networks. (release + demo `starship` -- an internal tool. introduces IPFS for context)