Schedule

Look through the lens of the DevOps engineer with Jibby Ayo-Ani to discover the power of cloud computing while using it to its utmost power to protect your company against everything from security vulnerabilities to natural disasters.

In the world of software development or technology in general, performance often gets overlooked or is looked at late. Daniel Drozdzewski examines the philosophical aspects and benefits of keeping performance at the forefront of your mind.

New software in production comes with a lot of risks, especially for companies with high availability requirements. However, it’s possible to make significant infrastructure changes while maintaining your company’s reliability. Jose Nino and Lita Cho outline deploying hybrid topologies with Kubernetes and Envoy.

Rob Skillington and Łukasz Szczęsny explore scaling monitoring, alerting, and configurational complexity for a single view of your applications, databases, infrastructure, and operations across all regions using M3 and Prometheus.

Laurent Bernaille examines the lessons he learned operating large Kubernetes clusters.

Developers working with Kubernetes still wonder what the optimal development workflow looks like. Join Ellen Korbes to take a look at the capabilities of the tooling available in the current landscape and see if it can offer end-to-end workflows that perform effectively in the real world.

Andy Kwiatkowski takes a deep dive into how Shopify saved a million dollars a year in infrastructure costs by rolling its own autoscaler.

Out-of-the-box Kubernetes makes it easy to deploy and scale your applications within one Kubernetes cluster in one single region. But it's also possible to deploy an application over multiple clusters in different regions, so it becomes truly highly available even if a complete region fails. Learn how to deploy one application across multiple federated Kubernetes clusters with Bastian Hofmann.

Regardless of all the technical benefits that Kubernetes brings, team interactions are still key for successfully delivering and running services. Manuel Pais explores how team design affects the success of Kubernetes adoption.

Join Heidi Waterhouse to learn about how it's easy to miss several important parts of the calculation when you make a build versus buy calculation, including maintenance, updating, security, availability, and finding operators.

Even software, written in high-level cross-platform language with no assembly can fail multiple ways when ported to a different CPU architecture. Ignat Korchagin examines the issues Cloudflare encountered when porting its software stack to ARM64.

Sweden's Television manages online products that range from providing news to TV series and are used by millions of people. To make sure that it creates content that engages, entertains, and educates, it started its own platform for collecting and analyzing user data. Ismail Elouafiq highlights the architectural choices the company made and the lessons it learned in building its data ecosystem.

Config changes cause more outages than code changes. Jamie Wilkinson introduces you to the UTM theorem, which shows that your application is a terrible software interpreter, and you should apply the same risk mitigation as for code deployment. This is harder with config languages; an application isn't a full interpreter. You can address this risk with progressive rollouts and config minimization.

Jenn Strater walks you through the best practices she's learned since transitioning from a software engineer at various product companies to working for a company that focuses on build automation.

Jan Lehnardt examines the challenges and solutions of changing the storage and distributed systems underpinnings from AP to CP (as per CAP) while retaining the unique multimaster replication APIs that CouchDB is renowned for.

Switching databases requires a lot of effort from engineering teams, and Christian Grabowski walks you through steps you can take to reduce the amount of work needed to achieve payoff. NS1 created an abstraction layer of wire protocols between old and new databases, which allowed it to develop advanced functionality in new services, while legacy services required minimal changes.

Games can teach the software world a lot about performance, monitoring, change, and beyond. Jon Manning and Paris Buttfield-Addison review current best practices for gathering large amounts of data from ongoing user activity, finding the best modifications to make in the analysis of that data, and how to roll out changes while hundreds of thousands of users are right in the middle of something.

Peter Bourgon dives into how Fastly is thinking about state at the edge and provides the first public look into the architecture and communication model of a global-scale data system that Fastly is currently prototyping.

Dmitrii Dolgov takes a deep dive into how to troubleshoot intricate performance issues in PostgreSQL using such tools as strace, perf, extended Berkeley Packet Filter (eBPF). And stay curious.

Molly Struve gives you the tools and strategies you need to build a monitoring system that will scale with your team and your infrastructure.

Extended Berkeley Packet Filter (eBPF) is one of the most exciting topics in the performance analysis space. IOVisor (Linux Foundation’s eBPF project) makes eBPF tools. Lorenzo Fontana is here to to explain how eBPF works, what purpose it serves, and the tooling available for Kubernetes.

Open source tools for dashboarding and metrics have seen massive adoption in recent years. Riding the hype, the new, shiny tools are inevitably confronted with overblown expectations and problematic usage patterns, causing frustration and criticism. Björn Rabenstein outlines how to use dashboards and metrics effectively rather than condemning them altogether.

The internet is underpinned by one critical protocol—Border Gateway Protocol (BGP). It's famously simple—by design. That simplicity helped it withstand the test of time and led many organizations to build complicated systems to drive their routing and shape their traffic. It's good enough for almost anything. Go beyond "good enough" with Nathanael Jean-Francois and take BGP to the next level.

Gilles Dubuc takes a deep dive into how Wikipedia interprets large amounts of real user performance data and the many pitfalls you can fall into when doing so.

Ignite is happening at Velocity. Join us for a fun, high-energy evening of five-minute talks—all aspiring to live up to the Ignite motto: Enlighten us, but make it quick.

Liz Rice empowers you to start writing extended Berkeley Packet Filter (eBPF) programs so you can unleash what has been described as superpowers for Linux.

Build pipelines are commonly used in the industry to build and roll out changes to cloud accounts. Typically, wide permissions are granted to those systems, making them an interesting attack vector. Take a look with Andreas Sieferlinger at typical vulnerabilities and examine the case of the confused deputy—a trusted third-party party—and how these vulnerabilities can be mitigated in real-life.

Time and money are generally the resources we focus on when building applications. Yet we can’t buy trust; it builds slowly and can be broken quickly when we don’t factor it in to our development process. Jennifer Davis examines how to leverage security practices to enable an all-team approach to security.

GDPR was likely one of the biggest challenges in data management that occurred in 2018. Yulia Trakhtenberg dives into a one-year retrospective about how it was executed in reality at a large-scale data organization.

Learn from the systems analysts of yore and how their practices apply to today's cloud native stack. Jesse Butler gleans some powerful ideas from the past to enhance your observability, resilience, and security in the cloud.

Homin Lee details constructing and using knowledge graphs to help DevOps teams make sense of the overwhelming volume of metric, log, trace, and event data generated by today's observability systems.

If you're interested in learning a framework of reference to enable continuous deployment to Kubernetes for business-critical production applications, join in. Priyanka Sharma bridges the gap between how to make large-scale migrations of production applications and the nitty gritty details that engineering managers and leaders need to consider.

Baruch Sadogursky analyzes real-world software update fails and how multiple DevOps patterns that fit a variety of scenarios could have saved the developers. Manually making sure that everything works before sending an update and expecting the user to do acceptance tests before they update is most definitely not on the list of such patterns.

Jumpstart your networking at Velocity by coming to Speed Networking before the keynotes begin. Bring your business cards and prepare a minute of chitchat about yourself, your projects, and your interests.

Program chairs Christopher Guzikowski, James Turnbull, and Ines Sombra open the first day of keynotes.

Details to come.

Karthik Gaekwad explores why the Oracle Container Engine for Kubernetes (OKE) is one of Oracle Cloud's most popular platforms. You'll learn the good and some ugly lessons learned along the way on how to manage, operate, and scale Kubernetes at a cloud provider scale.

By empowering you to ask new questions of your software, observability fuels curiosity about the world as it is, not how you expect it to be. In the end, after all, Christine Yen explains, "Nines don't matter if your users aren't happy."

This talk shows how good abstractions make it possible to identify and apply solutions to seemingly unrelated problems from different disciplines to build better systems with less effort.

Ana Oprea examines SRE and security best practices for designing, operating, and scaling dependable infrastructure.

We never change the amount of work or technical debt; we just shift it, and with it, we change how it emerges and appears. Heidi Waterhouse explains how you can handle this level of uncertainty.

The program chairs close day one of keynotes.

Join other attendees during lunch at Velocity to share ideas, talk about the issues of the day, and maybe solve a few problems. If you aren’t sure which topic to pick, don’t worry—it's not a long-term commitment. Try two or three, and settle on a different one tomorrow.

Join us in the Expo Hall for drinks and food at the Expo Hall Reception.