Conference Presentations

Agile DevOps West How to Prevent Catastrophic Doom on Your Next Federal DevOps Project
Slideshow

Trying to achieve real continuous deployments into production is hard for everyone, but it’s especially hard for highly regulated or government projects. These types of challenges range from client-specific, such as a set of manual checks and validations that need to be performed, to more generic problems, like how to version microservices and promote potentially breaking changes. Join Ryan Kenney as he discusses ways that he and his team have overcome obstacles to reaching continuous deployment. First Ryan will give an overview of the project and some of the problems they’ve faced. Then he will discuss how he was able to mitigate the lack of CD to an actual production environment by creating "developer production" and focusing the CD around that.

Ryan Kenney
Agile DevOps West Building the Blocks of Trust in Automation
Slideshow

When moving toward automation, establishing trust in the automation test suite is important to unite the team as a whole. Once trust is established in the process and the tests, it becomes crucial to the overall software development lifecycle. Join Sneha Viswalingam as she shares the journey of how her team of manual test engineers contributed to automation, the steps they took to build clean automation and win the confidence of the organization, and how they came to believe that the automation effort has their backs. She'll outline the strategies used to make the tests reliable and the test development and coding standards her team followed, then the tools and training provided for the manual testers to contribute to the automation efforts. Finally, she'll share the steps they took to make their automation efforts useful and visible to other teams and management by using a wallboard to display the metrics.

Sneha Viswalingam
Agile DevOps West Making the Jump from DevOps to DevSecOps
Slideshow

Organizations are moving to DevOps to build and deploy software more rapidly. But as they break down organizational silos to bring together testing, development, and operations, they often avoid or exclude security in their transformational efforts. Leaders in highly regulated organizations are often left wondering, where does my traditional security organization fit into this new DevOps world? How do I know that my applications are becoming more secure, while still getting the advantages of rapid, incremental deployment? Alan Crouch will talk about his experiences with financial, health, and government clients adopting DevSecOps practices to address these challenges. He’ll outline the essential characteristics that make up a strong DevSecOps pipeline and what practical changes you can adopt now. Then, he will describe how quality gates and security testing can be used to shift security left.

Alan Crouch
Agile DevOps West Fishbowl Discussion: How Much Automation Is Enough?
Slideshow

These days, everyone knows some automation is a necessity. More usually feels better. But when are you done? Or when do you stop for now? How can you tell if adding automation is no longer helping, or is even distracting from the real issues? Because the answer is "It depends," you'll want to listen to the wisdom of others who are on the same journey. In a fishbowl discussion, the audience members sit in a circle of chairs in the middle of the room. Several brave souls will fill all but one of the chairs in the "fishbowl." When you want to join as a speaker, you enter the fishbowl and sit in the empty chair, and one of the other speakers will voluntarily leave so that one chair is always available for a new speaker. You'll hear ideas and experiences from experts and peers alike. Come join Ryan Ripley as he facilitates this exciting conversation.

Ryan Ripley
Agile DevOps West Using Component Testing for Ultra-Fast Builds
Slideshow

A best practice of software architecture is to design your applications into independent modules or components, with a published contract for interaction between components. This is a principle of the microservices style of architecture, but it also applies to components created in a large monolith. If we can test the functionality of the component independently, and apply a level of trust that those components work, this opens the door to rethinking our continuous integration and continuous delivery strategy, potentially reducing the need for long test suites and many environments. It will also cause us to rethink our unit testing strategy and the test pyramid. Tim Cochran will talk through the different kinds of component testing, show working examples, and give advice about when to apply them. He will also cover what this might mean for your organization's broader testing strategy.

Timothy Cochran
Agile DevOps West 5 Common Types of Mobile App Bugs Found Using AI
Slideshow

Among all mobile apps, the current error rate is believed to be at 15 percent. With a thousand new apps launching daily and a constant increase of mobile devices, there’s a need for a scalable solution to create and maintain high-quality apps, without hassle. Thanks to artificial intelligence, exploratory testing is advancing and proving to detect mobile bugs at scale. Join Sandy Park as she examines the five most common types of errors found through more than ten thousand hours of AI-powered testing, with actual samples. She will introduce the challenges of each type and explain how the heuristic or rule-based approach was not able to address the issues efficiently. She will cover topics such as broken-element identification and Z-order detection for layered views. Finally, Sandy will share deep learning methods such as RCNN and LSTM, which enhance coverage and reliability.

Sandy Park
Agile DevOps West Case Study: An Engineering-Focused, Scaled Agile Rollout at Standard & Poor's
Slideshow

A large company moves to agile, but when the going gets tough, they abandon all their agile processes and revert to old ways—which are now a combination of Scrum and waterfall—and delivery is worse than before they started. Usually, what happens next is the CTO gets removed, and the new CTO comes in and proclaims again that we are all moving to agile to re-energize the organization, and they start their transformation once more. Have you seen this movie before? The agile transformation for Standard & Poor's played out this way twice, but finally, the third time was the charm—their last transformation was successful. But why was this time different, what makes them think it will stick, and how do they know as they continue to grow that they won’t revert to old ways again?

Stan Guzik
Agile DevOps West Hunting Sasquatch: Finding Intermittent Issues Using Periodic Automation
Slideshow

In pop culture, Sasquatch (aka Bigfoot) is an ape-like creature infrequently seen in the Pacific Northwest of North America—if he even exists. In the software realm, we have our own version of Sasquatch: that irritating, elusive "intermittent issue." Traditionally, we run automated tests on event boundaries, like when we have a successful deployment; we look for problems when we think they may have been introduced. Logically, points of change are when we expect to have injected issues, so we tend to only look for issues then. This approach alone, however, limits opportunities to reproduce intermittent issues. If we also run our automation periodically, we have additional opportunities to reproduce these types of issues; we call this approach periodic automation.

Paul Grizzaffi
Agile DevOps West Who Owns Quality in Agile?
Slideshow

What do you mean, who owns quality? The quality assurance team, of course—the kings and queens of quality, the masters of the tests, the lords of the sign-off. People often used to look down on quality assurance as less technical, the last to get their hands on the code, and the first to be blamed when things go wrong, but of course, agile adoption has changed the industry. These days we have cross-functional teams and develop test automation. But we also do "Scrummerfall" and have hardening sprints and stressful deadlines. Despite all of that planning, testing still often comes as an afterthought. We talk about teamwork, but generally, we see developers and testers as two different breeds. Join Katy Sherman in an exploration of quality as we try to imagine the future of our industry. What kinds of engineers will populate our teams, what skills will they have, and, most important of all, who will own quality?

Katy Sherman
Agile DevOps West What's That Smell? Tidying Up Our Test Code
Slideshow

We are often reminded by those experienced in writing test automation that code is code. The sentiment being conveyed is that test code should be written with the same care and rigor that production code is written with. However, many people who write test code may not have experience writing production code, so it’s not exactly clear what is meant. And even those who write production code find that there are unique design patterns and code smells that are specific to test code. Join Angie Jones as she presents a smelly test automation code base littered with several bad coding practices and walks through every one of the smells. She'll discuss why each is considered a violation and demonstrate a cleaner approach.

Angie Jones

Pages

CMCrossroads is a TechWell community.

Through conferences, training, consulting, and online resources, TechWell helps you develop and deliver great software every day.