Helm Summit: Full Schedule

8:00am PST

Registration & Breakfast

Wednesday February 21, 2018 8:00am - 9:00am PST
Bar area

9:00am PST

Welcome Remarks + History of Helm

Speakers

Adam Reese

Software Engineer, Microsoft

Adam Reese is a core maintainer for the Kubernetes Helm project. As a Senior Engineer at Microsoft, he has contributed to many open source projects. Over his 15 years building software, Adam has built everything from distributed pipeline processors to embedded systems.

Taylor Thomas

Nike

Taylor is a core maintainer of the Helm project and works as a Software Developer at Nike for his day job. He is a Gopher by way of Node.js and is an regular open source contributor

Wednesday February 21, 2018 9:00am - 9:30am PST
Ballroom

9:30am PST

Helm Journey at Ubisoft

Over the last year and a half, we have heavily invested in Helm. As such, we would like to share how helm is used internally at Ubisoft, by our core team and game teams, to deliver our services platform to games in production. Internally, we have over 60 charts, each with several versions (over 830 in total), a continuous delivery pipeline, several experiment with internal servers and some use cases that could be interesting for the community. In the presentation will tackle the following examples and use cases:
- Continuous Delivery using Helm
- Building and hosting our own chart repositories
- Our experience contributing to ChartMuseum
- Working in teams- Developing on Windows
- Using external charts- Assembly/umbrella charts
- Plugins to help manage values- Chart best practices

Speakers

Rodrigue Cloutier

Technical Lead, Ubisoft

Rodrigue's unusual career path, starting in marine biology and progressing into game programming, allowed him build a solid experience in complex systems. As a senior technical lead at Ubisoft, he is now heading up the implementation of a new online platform, backed by Kubernetes... Read More →

Wednesday February 21, 2018 9:30am - 10:00am PST
Ballroom

10:00am PST

Managing Environments as a Chart

Turn your Kubernetes environment into a Helm Chart-of-Charts.
Helm allows us to easily scope deployments of multiple charts, services and features using only a single values file per environment.

I'll explain how to manage such a configuration by setting smart default values, managing secrets you don't want to expose, and automating Helm runs into CI/CD while also handling failures.

I'll touch on how we arrived at our particular use case of Helm, how we use it to manage dev/stage/production environments, how we think it has benefited us, and what we hope to achieve in the long run.

Speakers

Brian Richardson

Cloud Engineer, IBM

Managing Environments as a Chart Brian Richardson Helm Summit 2018 pptx

Wednesday February 21, 2018 10:00am - 10:20am PST
Ballroom

10:20am PST

Hosting a Chart Repository with ChartMuseum

Learn how to host your own Helm Chart Repository in minutes using ChartMuseum, backed by your favorite cloud storage provider.

Speakers

Josh Dolitsky

Founder & Chief Engineer, bloodorange.io

Hosting a Chart Repository with ChartMuseum pdf

Wednesday February 21, 2018 10:20am - 10:40am PST
Ballroom

10:50am PST

Break

Wednesday February 21, 2018 10:50am - 11:05am PST
Ballroom

11:15am PST

Managing 6000 releases with Lostrómos, a Helm operator

What happens when you need to simultaneously deploy 6,000 unique instance of your application to different Kubernetes clusters at different providers worldwide? We built Lostrómos, a helm operator to power our deployment. Come learn all about it!

Speakers

Nicole Hubbard

Architect, WP Engine

Nicole Hubbard is an Architect at WP Engine where she focuses on building container based infrastructure, automation and helping teams deploy their applications.

Wednesday February 21, 2018 11:15am - 11:45am PST
Ballroom

11:45am PST

Testing and Releasing Charts Using Helm and ChartMuseum

A method of testing and a precursor to Continuous Delivery with Helm: quickly deploy a private Chartmuseum using Helm in order to store and release your own charts as artifacts. Increase confidence in your releases and standardize the process across environments, paving the way towards Continuous Deployment for your project.

Speakers

Stef Arnold

Software Engineering Manager, SUSE

Stef Arnold is a Senior Software Engineer on the containers as a service team at SUSE. Previously, Stef has worked in both the Software Engineering and DevOps disciplines in the financial and education industries.

helmsummit 2018 draft Feb21 pdf

Wednesday February 21, 2018 11:45am - 12:05pm PST
Ballroom

12:15pm PST

Lunch

Wednesday February 21, 2018 12:15pm - 1:15pm PST
Bar area

1:15pm PST

Helm at Reddit: From Local Dev, Staging, to Production

How Reddit uses Helm in local dev, staging, and production. An overview of the primary pieces (Helm and Docker repos, CI), supporting tooling, and some best practices we've identified.

Speakers

Greg Taylor

Engineering Manager, Reddit, Inc

Greg Taylor leads the Release Engineering team within the Reddit's Infrastructure division. He and his team steward the internal Kubernetes-based infrastructure product (InfraRed) and build tooling and process to empower service owners to get their ideas to production. Greg has recently... Read More →

Helm at Reddit From Local Dev, Staging, to Production pptx

Wednesday February 21, 2018 1:15pm - 1:35pm PST
Ballroom

1:35pm PST

Securing Helm

Providing a secure Helm installation is important when running in production environments. It’s important to understand the security implications of running Helm on your clusters. This session will share best practices for securing Helm in different operational contexts.

Speakers

Lachie Evenson

Principal Program Manager, Microsoft

Lachlan is a Principal Program Manager on the open source team at Azure. As a cloud native ambassador, emeritus Kubernetes steering committee member and release lead, Lachlan has deep operational knowledge of many Cloud Native projects. He spends his days building and contributing... Read More →

Wednesday February 21, 2018 1:35pm - 2:05pm PST
Ballroom

2:05pm PST

I Can Haz Services?

Hear ye, hear ye! Announcing the new and improved Monocular! We added support for managing your app’s service dependencies right from the UI.

Have you ever copy/pasted your database credentials into a secret, then mounted the secret into your PodSpec in a chart? No more! Service catalog makes it easy to automatically create the service and mount the credentials into your app, right from your chart. After you install your chart, you can even see which apps are using which services. We call that the service dependency graph.

Before, you did all this in YAML. Now you can do it all in Monocular.

Here’s the breakdown. Directly in the new Monocular UI, you can:

- Update values for Helm charts before deploying them
- See all the services available to your charts
- Provision (create) and bind (get credentials) for managed services available in the service catalog
- … from any cloud provider, regardless of where you’re running
- See existing service instances and credentials used by applications

Speakers

Angus Lees

Principal Engineer, Bitnami

Angus Lees is a principal software engineer with Bitnami, working on and with Kubernetes. He has decades of experience with Linux and the Free Software community, including 9 years working with Google’s internal cluster and networking infrastructure, several years as an OpenStack... Read More →

Aaron Schlesinger

Cloud Developer Advocate, Microsoft

Aaron is a developer advocate at Microsoft Azure and a core maintainer of the Athens Project. He is an emeritus core maintainer and chair of the Kubernetes SIG-Service-Catalog and a contributor to various other projects in the Kubernetes community. He enjoys distilling his wide ranging... Read More →

Wednesday February 21, 2018 2:05pm - 2:35pm PST
Ballroom

2:45pm PST

Afternoon Coffee and Snack Break

Wednesday February 21, 2018 2:45pm - 3:05pm PST
Bar area

3:05pm PST

Testing Lessons Learned From The Community Charts

The community stable and incubator charts have come a long way in the past couple years. One area that has helped to keep changes coming while keeping the charts stable is testing. Over the past 6 months the testing setup has grown to provide faster more thorough automated feedback. In this session we’ll explore how and why those improvements work so that can apply them in your own environments.

Speakers

Matt Farina

Distinguished Engineer, SUSE

Matt works as a Distinguished Engineer where he focuses on cloud native technologies. He is an author, speaker, and regular contributor to open source. Matt has a particular interest in developer tooling and experience, CI/CD, dependency management, and, of course, cloud native technologies... Read More →

Wednesday February 21, 2018 3:05pm - 3:35pm PST
Ballroom

3:35pm PST

A Newbie's Onboarding to Helm: First Impressions

Newbie to the Helm product and user community goes through the process of a.) installing and configuring a Helm scenario in a demo environment b.) presenting that demo to a customer environment and c.) contributing to the Helm community. This talk will present the experience, lessons learned and suggestions for improvement to the onboarding process for new Helm users and contributors.

Speakers

John Yandziak

Customer Engineer, Google

Wednesday February 21, 2018 3:35pm - 3:40pm PST
Ballroom

3:40pm PST

Proposed Chart Best Practices for PVCs

It would be nice to have documentation around common use cases and best practices for PVCs within helm charts. Some suggested best practices to be covered include default configurations for helm charts that use PVCs, complex concepts like multiple mounts to one PVC, and more.

Speakers

Robert Kernick

Consultant, Oteemo

Wednesday February 21, 2018 3:40pm - 3:45pm PST
Ballroom

3:45pm PST

NetworkPolicy Resources: Proposed Best Practices for Charts

Until now, there has only been very limited discussion in the Helm development community about the details of including NetworkPolicy resources in charts. In this talk I propose some best practices for resource templating, structuring values, and ways to think about NetworkPolicy generally, by analogy with existing "bare metal" systems and already-accepted chart best practices like RBAC.

Speakers

Joe Thompson

Solutions Architect, Mesosphere

I'm a solutions architect for Mesosphere and a regular participant in the Helm and SIG-Apps communities. Prior to Mesosphere, I worked at Capital One, CoreOS and Red Hat (among others), providing practical solutions and training in and for Kubernetes and other cloud environments... Read More →

Wednesday February 21, 2018 3:45pm - 3:50pm PST
Ballroom

3:50pm PST

Helm trick you'd want to know! - Deployment update based on ConfigMap

Kubernetes provides a great and easy way to manage the life cycle of our applications. There are various options available to configure different kinds of applications. We can specify how many instances or pods we want running for our app, how we want to update it, or even pass in some configuration files needed by some apps. Now we don’t want these config files baked into our image as once the config file changes, we will have to change the image again. That’s where the amazing ConfigMap comes to the rescue. In today’s talk, I’m going to focus on how Helm provides a brilliant and necessary feature related to ConfigMaps. When a config file content is changed, we also want our deployment to get updated, and Helm provides a way to do that without having to run any additional commands! In this talk, we will demonstrate how you can achieve this. So in this world of automating everything, let’s make use of this too

Speakers

Rajashree Mandaogane

Senior Software Engineer, Apple

Rajashree, an NC State graduate is a Software Engineer at Rancher Labs. She loves programming in Golang and working on container orchestration. She lives by the motto, “You’ll never know what you can do, until you try”. This has led her not only to being a developer, but also... Read More →

Wednesday February 21, 2018 3:50pm - 3:55pm PST
Ballroom

3:55pm PST

Can you keep a secret? A practical discussion about protecting secrets in Helm

Helm is a fantastic package manager that makes configuring applications on Kubernetes a breeze. Managing secrets within those apps, however, tends to be an afterthought because of challenging it can be. Whether it's certificates, database passwords, or API tokens, figuring out how we handle their security is a necessity. In this talk, we'll talk best practices and the tools that exist today for keeping your secrets secure when using Helm.

Speakers

Zach Sais

IBM

HelmSecretsPresentation pdf

Wednesday February 21, 2018 3:55pm - 4:00pm PST
Ballroom

4:00pm PST

Towards a Standard Containerized App Registry

The number and complexity of applications regularly deployed to production Kubernetes environments is increasing, making managing collections of container images as important as managing individual container images. Development around Kubernetes application toolchains, such as Helm or KPM, has created a vibrant but fragmented ecosystem. The community is working on standards for consolidating interfaces around these toolchains. This talk will discuss the work to standardize the uploading and downloading of applications to an application registry, making the process both familiar and secure across different tools.

Speakers

Jimmy Zelinskie

Product Manager, CoreOS

Jimmy Zelinskie is the product manager for the everything related to Kubernetes applications at CoreOS. Together with Antoine Legrand, he worked to create the App Registry project, which intends unify registry protocols for tooling in the Kubernetes ecosystem.

Wednesday February 21, 2018 4:00pm - 4:05pm PST
Ballroom

4:05pm PST

WheelHouse – A Public Hub for Helm Charts

Have you ever wished that you had a public location that served as your private Helm Chart repository? Some place that you could quickly push charts to / pull charts from without having to set up a HTTP server or a ChartMuseum instance?

A few of us at SUSE have often felt that having something like a 'DockerHub' for Helm Charts would be extremely useful. Through this discussion, we would like to gauge interest for something like this, identify potential collaborators, and lay the foundation for the design, development, and operation of such a service.

Speakers

Nikhil Manchanda

Engineering Manager, Microsoft

Hello there! I'm an Engineering Manager at Microsoft helping customers build cloud applications using Azure, Helm, Kubernetes and other container technologies. I was previously one of the founders of the OpenStack Trove project. My main areas of expertise are databases and IoT, but... Read More →

Wednesday February 21, 2018 4:05pm - 4:10pm PST
Ballroom

4:25pm PST

Closing Remarks

Speakers

Taylor Thomas

Nike

Taylor is a core maintainer of the Helm project and works as a Software Developer at Nike for his day job. He is a Gopher by way of Node.js and is an regular open source contributor

Wednesday February 21, 2018 4:25pm - 4:30pm PST
Ballroom

4:30pm PST

Happy Hour & Speed Networking

Wednesday February 21, 2018 4:30pm - 6:30pm PST
Loyal Legion 710 SE 6th Ave, Portland, OR 97214

8:30am PST

Breakfast

Thursday February 22, 2018 8:30am - 9:30am PST
Bar area

9:30am PST

Opening Remarks + Future of Helm

Speakers

Matt Butcher

Principal Software Development Engineer, Microsoft Azure

Matt does cloud native open source development at Microsoft, where he has worked on Brigade, Helm, Krustlet and others. Matt is the author of a bunch of books and articles, most recently O'Reilly's book "Learn Helm" (with Matt Farina and Josh Dolitsky). When not coding, Matt enjoys... Read More →

Matt Fisher

Software Engineer, Microsoft

Matt is a core maintainer of the Helm project, and a Software Engineer at Microsoft. Off hours, Matt enjoys spending time camping, woodworking and spending time with his family.

Thursday February 22, 2018 9:30am - 10:00am PST
Ballroom

10:00am PST

By Any Other Name: A Proposal for How Aliases, Dependencies and Value Mapping Should Work

The current alias, dependency and import system has some rough edges that make it difficult or impossible to do some notionally simple tasks. In this talk I propose some ideas for revamping how they can be used, treating these sets of features as a unified whole that act in concert to simplify chart authoring.

Speakers

Joe Thompson

Solutions Architect, Mesosphere

Thursday February 22, 2018 10:00am - 10:05am PST
Ballroom

10:05am PST

Tiller CRD and Controller

This talk will be a quick pitch for a possible Tiller design for Helm 3 using CRDs and controllers. I'll show some example CRD definitions and flow chart of how the basic controller logic will work

Speakers

Taylor Thomas

Nike

Taylor is a core maintainer of the Helm project and works as a Software Developer at Nike for his day job. He is a Gopher by way of Node.js and is an regular open source contributor

Helm 3 CRD pdf

Thursday February 22, 2018 10:05am - 10:10am PST
Ballroom

10:10am PST

More Charts, More Problems

Package managers are hard. Helm learned a lot of lessons from others’ mistakes, but also repeated some. For example, having a single index file per repository is not scalable. It’s the same mistake that NPM made and it causes slower CI, high memory consumption, slower searches, and more. In this lightning talk we’ll explore several solutions to the problem, such as compression and partitioning, their strengths, and their weaknesses.

Speakers

Ankush Chadha

Senior Solution Developer, JFrog

Ankush Chadha is a Senior Solution Developer at JFrog and is focused on partner integrations to cover all angles of the CI/CD environment to further enhance the overall experience. Prior to JFrog, Ankush worked as a Release and Build Manager at Oracle (Fusion Middleware). He has more... Read More →

Thursday February 22, 2018 10:10am - 10:15am PST
Ballroom

10:15am PST

Measuring How Helm Is Used

We measure what we want to understand. Some measure how mobile and web applications are used. Some measure how desktop applications are used. Almost everyone is monitoring who uses their websites and what kind of content is useful. And, some measure how CLI applications are used.

This lightning talk looks at what measuring Helm’s usage could look like and what similar applications have done.

Speakers

Matt Farina

Distinguished Engineer, SUSE

Thursday February 22, 2018 10:15am - 10:20am PST
Ballroom

10:20am PST

OverHelmed: Experience of a Helm chart publisher and consumer

We, AppsCode, publish the largest number of charts (currently 6) for a single organization in stable repository. We also explored how to use Helm for managing internal applications. In this talk, I would like to share what worked for us as a chart publisher and consumer. I would also share some pain points we experienced and how we addressed those.

Speakers

Tamal Saha

Founder & CEO, AppsCode Inc.

Tamal Saha is involved in Helm community for over an year both as a chart publisher and consumer. He has also contributed to the Helm project. Currently he runs one of the fastest growing Kubernetes application development companies, AppsCode Inc.

Thursday February 22, 2018 10:20am - 10:25am PST
Ballroom

10:25am PST

Felloe, Spokes, Axel, and Tiller: A Look at the parts of Helm

Helm describes itself as a package manager, which includes package searching and browsing, repository push and pull, and dependency management. However, it also supports template expansion and deployment (aka "release") management. There are dozens of tools in the Kubernetes ecosystem that perform the latter two functions, but only a couple that provide package management. By supporting arbitrary formats and deployment mechanisms, Helm could greatly expand its user base and truly become *the* package manager for Kubernetes.

Speakers

Brian Grant

Principal Engineer, Google

Brian is the co-Technical Lead of Google Kubernetes Engine, former co-Chair of Kubernetes SIG Architecture, Kubernetes Steering Committee Emeritus, and CNCF Technical Oversight Committee member. His experience while technical lead of Google's internal container platform, Borg, motivated... Read More →

Thursday February 22, 2018 10:25am - 10:30am PST
Ballroom

10:30am PST

Break

Thursday February 22, 2018 10:30am - 10:45am PST
Ballroom

10:45am PST

Q&A Panel with Core Maintainers

Join several of the Helm Core Maintainers for an open Q&A session

Thursday February 22, 2018 10:45am - 11:15am PST
Ballroom

11:15am PST

Unconference and Working Groups

Join your fellow conference members in discussions about Helm 3 features and Charts.

Thursday February 22, 2018 11:15am - 1:00pm PST
Ballroom

12:00pm PST

Lunch

Thursday February 22, 2018 12:00pm - 1:00pm PST
Bar area

1:00pm PST

Break

Thursday February 22, 2018 1:00pm - 1:30pm PST
Ballroom

1:30pm PST

Future of Helm Discussion and Working Group

As we finish up the Summit, come join us for a final discussion on the future of the Helm project

Thursday February 22, 2018 1:30pm - 3:00pm PST
Ballroom

3:15pm PST

Closing Remarks

Speakers

Taylor Thomas

Nike

Taylor is a core maintainer of the Helm project and works as a Software Developer at Nike for his day job. He is a Gopher by way of Node.js and is an regular open source contributor

Thursday February 22, 2018 3:15pm - 3:25pm PST
Ballroom

5:00pm PST

Evening Event

Thursday February 22, 2018 5:00pm - 7:00pm PST
Pips & Bounce 833 SE Belmont St Portland, OR 97214