Paul Cormier: App dev and operations for evolving IT environments

Talking ’bout an IT evolution (baby)

Some call it digital transformation. Others say it’s the evolution from traditional to modern, or from mode 1 to mode 2. Paul Cormier, Red Hat president of Products and Technologies, sees it as an evolutionary approach to architecture, processes, and platforms. In his keynote at the 2016 Red Hat Summit, he explores how these shifting landscapes affect both infrastructure and applications, as well as developers and operations.

To start, Cormier looks at infrastructure and app development through the lens of architecture, processes, and platforms.

Infrastructure

across-all-architecturesInfrastructure architecture is moving from proprietary to open source development, and from single footprints to a combination of physical, virtual, private, and public resources. These changes necessitate accessible software-based storage and networking, as well as common management and consistent applications. Without all the pieces working in harmony, infrastructure can become inefficient and complex–and that isn’t sustainable.

With all the changes to infrastructure, processes must change to match. More automation, better tools, and common management can help, but streamlined processes must infiltrate the entire organization to make increasingly complex infrastructure sustainable.

Similarly, the host platform must be utterly stable across the entire stack. Platform consistency is where Red Hat Enterprise Linux has made its mark–our customers can use the same foundation across physical, virtual, private, and public environments.

Application development

On the application side, organizations are moving from monolithic apps to microservices. This shift means you can take advantage of new architectures and:

  • Eliminate overhead.
  • Increase usability.
  • Build applications and services more quickly.

And just like with infrastructure, app development processes must advance. The traditional, top-down waterfall approach can’t keep pace with today’s need for speed. Many organizations are looking to agile and DevOps practices so they can work in a more continuous fashion.

We must be ready to support customers on any medium. Without the right platform, modern applications, architectures, and processes cannot reach their full potential.

Customers are leading the charge

Cormier called out some of Red Hat’s most innovative customers as proof that the landscape is changing for the better. He highlighted how:

Betfair, the largest betting exchange in the world, moved to OpenStack and software-defined networking (SDN) to achieved higher transaction volume, with less downtime and reduce their reliance on physical networking equipment and operations.

Target, the 2nd largest U.S. retailer, used DevOps with OpenShift on OpenStack to unify in-store and web shopping for a seamless experience. Open hybrid cloud gives Target flexible, scalable options.

Amadeus, the largest distributor of travel and leisure technology in the world, built on its middleware and OpenStack success, expanding to OpenShift 3 and containers to automate and streamline operations and improve platform availability to create a new cloud computing platform that meets the demand for new services. Amadeus is also an example of customers giving back, as they worked with Red Hat to co-develop features they they shared back into the upstream.

We also see in our customers that the change they need isn’t just about technology–it’s about culture. Culture that makes the open source development model work. It’s what makes our company work. Culture is the core of industry innovation today.

“I can’t emphasize enough the need for a change in culture. Collaboration by people across organizations is critical.”

Open is the new standard

And you don’t have to take our customers’–or Cormier’s–word for it. According to Black Duck Software’s annual Future of Open Source surveys from 2016 and 2015, 78% of enterprises are running on open source and 65% of companies are contributing back to open source communities. Tens of thousands of businesses are part of the 26 million repos on GitHub, making real investments in the future of open source.

This is the same investment that Red Hat has been making for years. Part of our business is making open source consumable for the enterprise, but the other part of that is making enterprise enhancements and innovation flow back upstream into the communities.

“This is the model that leading companies are and will embrace. This is just the beginning. Organizations are realizing now that this is the only way to innovate in software.”

community-to-enterprise

The industry is converging around open source architecture, process, and platform advancements.

The architecture is microservices.
The process is DevOps.
The platform is open hybrid cloud.

Let’s get practical: Application development

But how does all this apply to daily life for IT teams? Cormier walked through practical workflows, starting with development.

Local development

If you’re a developer, you can now:

  1. Choose your services: integration, process automation, data services, and application development platform. (Red Hat JBoss Middleware)
  2. Choose your frameworks.
  3. Establish (or choose) a common, lightweight way to run those environments.

Choosing the container environment is critical. Containers give you the power to combine these choices, and run them in a way that’s consistent with your production environment. Cormier emphasized that Linux–and Red Hat Enterprise Linux, in particular–has been re-architected to serve as the foundation for these containers, providing stability, security, and consistency across any footprint.

Take it to the team lab

Once you have a container with your app or service inside, you have to pull in the rest of your team to test and integrate your work and prepare it for production. This is critical as you move towards microservices architecture, where orchestration is essential.

Production-ready

Now you can take the same container, developed using the same code on the same platform, and move it into production. This is a consistent process. This is the right way to do the Dev side of DevOps.

In production, you’ll integrate storage, networking, and high availability–all things that your containers are ready to connect to. And now your application is real, and its life cycle can begin.

Let’s get practical: Operations

Cormier also addressed some typical types of operations activities. Ops is responsible for the security, performance, and reliability of the infrastructure apps run on. Let’s look at a security issue like Heartbleed.

Responding to these security events is one of the most important things operations does. The process might go like this:

  1. Problem is detected in the base container.
  2. Figure out which containers in your platform are affected.
  3. Create a patch, and recompose the app on a new base image. (The application layer may not change at all.)
  4. This change must be done automatically to avoid human error and to operate quickly at scale.

These changes may take place daily, weekly–much more often than you’d think. There are several methods for pushing changes into production in incremental ways, such as canary and blue-green updates.

This isn’t the only kind of change that happens. Apps are also changing constantly as well, and the more apps you have the more likely you are to have a constant stream of updates. You need to be able to make aggressive, frequent changes and roll them back in a consistent, predictable, and reliable way in your production environment. This is what DevOps allows you to do.

“Once you’ve removed the limits of slow processes and manual paths, you’ll be able to move at the velocity you’ve always wanted.”

Red Hat can help

We’ve grown our operational experience over the years, putting us in a unique position in the industry. With our growing portfolio and focus on modern app development and insfrastructure, Red Hat can help organizations achieve one consistent environment.

Yesterday we announced some additions to our product family: consistent container solutions that stretch from the individual development environment to production for the enterprise.  These solutions include:

Red Hat OpenShift Local – for developers
Red Hat OpenShift Lab – for the team environment
Red Hat OpenShift Container Platform – for production environments  (This is the product formerly known as OpenShift Enterprise by Red Hat.)

Red Hat Cloud Suite extends this functionality with management and cloud capabilities to help our customers fully realize open hybrid cloud, all the way out to the public cloud. And Red Hat Open Innovation Labs helps customers take advantage of these products in their own environments.

Cormier then revealed that everything he had just talked about in theory, would be demonstrated live on stage by a team of Red Hat developers and operations experts. He brought up Burr Sutter and his team. Read that summary to get the full details of their demo.

LK1_0951

Learn more

> See the DevOps demo
> See the illustrated version.
> See the video of the talk.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s