Practices of DevOps Track

Chris Brown
Chris Brown
Track Chair
DevOps is a cultural and professional movement that grew directly from the collective experience of practitioners. It is a reaction to the rise of online consumption and of customers’ expectations of increasing pace of development and delivery. IT, once a tool of back office efficiency, has become the front office. Devops integrates development and operations for safe, measurable and continuous delivery of business value.

We will have practitioners sharing how they are adopting practices like continuous deployment, production instrumentation, and telemetry to deliver higher-quality, more valuable software into users’ hands more quickly – whether those users are in the cloud or serviced with a hybrid cloud / enterprise infrastructure.

Breakout Sessions

Mike Brittain
Mike Brittain
Director of

Advanced Topics in Continuous Deployment

Continuous Delivery changes the fundamental processes involved with building software and launching new features. This talk will focus on the impact of Continuous Delivery on software release cycles, managing versions of software and database schemas, the structure of your engineering team, and risks involved with Continuous Delivery and how to manage them. Using Etsy as a case-study, Mike will discuss each of these aspects with takeaways that you can apply to your own teams.
Sebastian Holst
Sebastian Holst
EVP & Chief
Strategy Officer

Application Analytics: What Every Application Stakeholder Should Know

This session will provide practical guidance as to when, how, and why application analytics should be incorporated into your development process and lifecycle management arsenal. The gold standard for any class of analytics is “actionable insight;” how much smarter, faster, or efficient can we become when the right people get the right information in the right format at the right time?

Accurate, timely, and relevant application usage analysis is a “no brainer” for any team committed to agile feedback driven design and development principles, but knowing what telemetry is actionable and actually delivering that data safely and efficiently to the right application stakeholders has been, more often than not, considered to be too difficult, risky, or expensive to implement.

This session will use specific examples and demonstrations to quantify the value of application analytics and demystify the steps required to implement an effective application analytics implementation. The session will:
  • Identify well-understood application analytics use cases for Usability, Quality, and Effectiveness.
  • Provide a common set of Application Analytics requirements and the functional and architectural implications that follow.
  • Offer implementation tips that relating to IDE integration, proper collection of business and functional requirements, as well as design and planning decisions to ensure a healthy evolution (maturation) of application analytics usage over time.
Rob Cummings
Rob Cummings

Bootstrapping Continuous Delivery in the Enterprise

The established enterprise often struggles to adopt major shifts in the technology landscape. However, for most it has never been more important than right now to do this successfully and quickly. Nordstrom is in the early stages of one such transformation and our challenges around implementing continuous delivery and a devops culture will sound familiar. This talk focuses on why change in an established organization is hard, where we stumbled, and patterns that will help you avoid similar traps.
John Esser
John Esser
Director of

Continuous Delivery at Scale

Doing continuous delivery by itself is hard enough, but what about doing at large scale? We know that everything gets harder at scale. Doing continuous delivery with a single or a few services is pretty straightforward, but what about when you have dozens or hundreds of teams releasing hundreds of services and components into production? How can DevOps work at scale? How does culture, team structure, architecture, and tool chain enable or hinder this? John, Director of Engineering Productivity, will share lessons learned at on scaling continuous delivery.
Mahendra Pingale
Mahendra Pingale
Senior Product Manager
IBM DevOps

DevOps: It’s More Than Just Getting Code into Production

The software industry has fully embraced the notion that adopting DevOps methods will increase communication and integration between the development and operations teams. And when people talk about implementing DevOps, they usually bring up continuous delivery and continuous integration. But paradoxically, this conversation is only about the flow of the code that’s to be put in production. There is usually very little talk about the flow of the artifacts among the various practitioners, and disciplines, on the software development and delivery team. Yet these artifacts embody the core of the collaboration that has to happen in order to get the whole team running well.

But let’s think about the tools we give our practitioners:
  • product owners have requirements management tools
  • testers have test management tools
  • developers prefer to interact with lightweight project and issue tracking software
  • the Program Office uses PPM software
In many ways, these tools serve to separate the various disciplines in software delivery. Since these tools don’t interact, how can the practitioners who use them be expected to collaborate effectively using them?

During this talk, our speaker will talk about the collaboration challenges in software development and delivery teams, and how to address them with an emerging discipline called “Software Lifecycle Integration.” This discipline recognizes the diversity of tools and practitioners in the software development and deployment lifecycle, and seeks to improve the connection and collaboration among them. He will talk specifically about three collaboration/integration patterns and how to implement them.
Richard Seroter
Richard Seroter
Senior Product Manager
CenturyLink Cloud

How Any Organization Can Transition to DevOps – 10 Practical Strategies Gleaned from a Cloud Startup

Achieving the benefits of a DevOps culture isn’t just for startups, and it certainly isn’t about purchasing a new tool or turning operations staff into developers. Tier 3 – now part of telecommunications powerhouse CenturyLink – competes with numerous public cloud providers with far larger development teams. But its leadership knew that efficiency could be a competitive advantage. To achieve this, the Engineering and Operations teams partnered up to create a more consistent delivery and maintenance experience. This challenge was especially daunting as the user base and number of supported data centers grew exponentially. In this session, we will walk through this journey and discuss how 10 strategies related to people, processes, and tools resulted in a dramatic improvement in deployment efficiency and operational performance. Hear lessons learned, key success criteria, and implementation details that can work for organizations of any size and industry.
Ed Blankenship
Ed Blankenship
Product Marketing Manager

Make Data-Driven, High-Impact Improvements to your Applications

Today, application performance monitoring, usage analytics and log analysis are typically disconnected activities, each introducing specialist tools that layer complexity for a developers. Having integrated telemetry at your fingertips lets you see real data on how your customers are using your applications, correlated to their experience and to the application’s business success. This session will show you how to use real data to improve your application and determine the right next priorities in your application backlog.
Bernard Golden
Bernard Golden

The New Cloud Application Design Paradigm

Cloud computing offers increased agility, easier scalability, and better correlation between resource use and cost; however, attaining the full benefits requires a different approach to application design and operation. This session will discuss the technical and organizational changes that cloud computing requires IT to implement in a world of low friction infrastructure resource access and pay-as-you-go pricing.

Specific topics addressed include:
  • Application architecture changes to support elasticity and cost-effectiveness
  • Supporting vastly higher scale and user load
  • Methods to achieve cost-effectiveness in use of cloud environments
  • Organizational implications in accelerating the application lifecycle
Real-world examples will be offered to illustrate how cloud computing enable and challenge IT organizations as they make the shift to the next generation platform cloud computing represents.
Kevin Hancock
Kevin Hancock
Senior Director Worldwide Field Operations

A Practical Path: From Agile Teams to DevOps in the Cloud

Increasing globalization and distributed teams have caused software managers to be responsible for faster cycle times and increased visibility, all while controlling costs and maintaining compliance. By leveraging emerging Platform-as-a-Service offering, an increasing number of enterprises are scaling Agile practices and moving on through DevOps as a way to speed cycle times. This session will look at how organizations can combine Agile, DevOps and Cloud Development to help individual work groups improve collaboration and release software faster. It will provide a practical path for combining commercial and open source development tools, various processes and external clouds to deliver quality software – from Agile to DevOps. With DevOps the end-game, attendees will learn strategies to improve visibility and collaboration across development and deployment processes, reduce manual and error-prone hand-offs among internal teams and across partners, and improve business agility by connecting business owners, development teams and end-users.
Matt Ray
Matt Ray
Cloud Integration Lead

Principles for Navigating the Future Today

"Specialization is for insects." This Robert Heinlein quote is more appropriate than ever, the pace of change continues to accelerate and the standards of today are the relics of tomorrow. How do we thrive in a world that is constantly changing? This talk will apply nine principles for navigating the 21st century outlined by Joi Ito of the MIT Media Lab. Open Source, Agile and DevOps are fundamental to adapting to change, managing risk through resiliency, and cultivating a culture of networked improvement. We may not know what the future holds, but we can get ready for it today.
Claude Remillard
Claude Remillard
Group Program Manager

Release Management Practices for the Enterprise

Release management is one of the key processes in a DevOps strategy for the enterprise. This session will focus on the release patterns, strategies and considerations in implementing a release pipeline. We will look at the tools, people and process aspects and share adoption do’s and don’ts based on experiences in the field.

Lightning Sessions

Steve Neely
Steve Neely
R&D Engineer
Rally Software

Continuous delivery? Easy! Just change everything. (Well, maybe it isn’t that easy.)

Continuous Delivery is a hot topic in both the engineering and business worlds. Adopting Continuous Delivery radically changes the cadence and fundamental processes involved in building and shipping software.

This talk focuses on lessons learned by Rally Software as we transitioned from delivering releases every 8-weeks (with Scrum teams), to delivering on-demand with a continuous workflow model (Kanban). The presentation details struggles with build systems, test systems, customer enablement, and internal communication, and speaks to the light at the end of the tunnel -- greater control and flexibility over feature releases, incremental delivery of value, significantly lower risk, and much more.

Don’t miss the chance to learn from our successes and mistakes, and get your teams deploying more frequently. As we focus on both business (product management) and engineering perspectives, this session will explain how to automate and manage simultaneous software release cycles while minimizing unnecessary risks.
Sean Kennedy
Sean Kennedy
OSLC Community Development Leader
IBM Rational

Facilitating Stockdale: Confronting and Overcoming the Brutal Facts of Continuous Improvement

To paraphrase, the "Stockdale Paradox" is: To have unwavering faith that you will prevail in the end, while AT THE SAME TIME, confronting the most brutal facts of your current reality. For product delivery organizations, the Stockdale Paradox can be a simple motto to help focus its energies in the face of aggressive competition, changing customer desires, challenging business climates, and more. For an Agile or Lean product delivery organization, it is also a reminder of the need for true, lifecycle-wide, continuous improvement. No matter how "brutal the facts of your current reality", however, sometimes there is nothing more brutal than trying to implement improvement in an established process - especially one (paradoxically?) that has recently been adopted as part of an "agile transformation" or "lean revolution". In this presentation we'll consider how the Stockdale Paradox applies to Agile and Lean product delivery organizations in general, and to their discipline of continuous improvement specifically. We'll take a look at some common barriers and reflect on some of the necessary characteristics for sustainable continuous improvement.
Jose Luis Soria
Jose Luis Soria
ALM Team Lead

Patterns and anti-patterns for (Continuous) Delivery: proven tips to improve the way you release software

Successfully delivering software means much more than only coding or even testing. As Mary Poppendieck says, "How long would it take your organization to deploy a change that involves just one single line of code? Do you do this on a repeatable, reliable basis?" People have been worried about how to dramatically improve the way they code and test their applications, but it has not been until the popularization of movements such as Continuous Delivery or DevOps when they have begun to pay attention to the way they release. The way your team releases software is getting more and more important, even making the difference between a failed and a successful project. And it is getting to represent a key factor to help differentiate an organization from its competitors.

This session covers useful patterns and practices to improve the way you deploy, release and deliver software, as well as some pitfalls to be avoided, both in Continuous Delivery environments as well as for more traditional release processes. It is not targeted to any specific technology or platform, since it is based on general principles that can be applied no matter the language or technology stack you use, although most of the guidance covered through the session is gathered from the book the presenter has co-authored about the subject, Building a Release Pipeline with Team Foundation Server 2012.

Some of the key topics being covered are: tips for orchestrating a deployment pipeline, practices for deployment and test automation, configuration management for release artifacts, getting actionable feedback and metrics about your release process, and advanced release techniques.

David Atkinson

Automating database deployments in an agile world (From Wednesday BOF)

Everyone is comfortable automating application deployments, but having immense difficulty with the database...
Facebook Twitter LinkedIn LinkedIn
2014 Keynote Videos

All ALM Forum 2015 conference
attendees also receive:

photo photo photo
ALM Forum Logowear
ALM Forum Bag