Live from the Summit: Intro to Red Hat Enterprise Linux OpenStack Platform

oxanham-RHEL-OSP-sessionIf you’ve been anywhere near a tech conference, blog, or the word “cloud” in the last year, you’ve probably heard about OpenStack. It’s one of the most exciting projects right now in the open source community, and with good reason. But what does it do, and what is Red Hat’s involvement in the project?

Rhys Oxenham, OpenStack field product manager for Red Hat, answered these questions and more in a standing-room only afternoon session on Tuesday.

THE WHAT

“In the most basic sense,” he said, “OpenStack is an open source cloud operating system. It has a number of sub-projects and building blocks that let you create an Infrastructure-as-a-Service (IaaS) cloud.”

OpenStack’s architecture has 9 components, such as identity, image service, object store, orchestration, and more, and each serves a uniquely separate purpose. They combine to offer “a framework for how you would implement a cloud, relying on plugins and drivers to let you configure the cloud you want to choose,” said Oxenham.

Developed by a community of contributors, both individual and corporate, OpenStack offers an innovative alternative to propriety cloud services. The community produces a new release every 6 months, and the timing for Oxenham’s session is perfect—the next release, codename “Icehouse,” is due out on April 17.

The current release, Havana, had 13,700+ code commits, 920 individual contributors, 150+ organizations, and 400+ new features. “It’s growing exponentially,” said Oxenham.

THE WHY

Traditional IT has its share of issues, as Oxenham explained. “There’s too much data, and service requests are too large,” he said. “Applications weren’t written to cope with this kind of demand, and performance is slipping as a result.”

“Public clouds are setting the benchmark for how IT can deliver to users, but not all organizations are ready yet,” said Oxenham. “With OpenStack, you get public cloud-like capabilities behind the privacy of your own internal firewall.”

Red Hat got involved with OpenStack 10 months after the project launched in July 2010 and has been the “No.1 corporate contributor for the last 2 releases,” Oxenham said. Red Hat Enterprise Linux OpenStack Platform offers all of the benefits of the community in an enterprise-grade product, “built specifically for, and tightly integrated with, [Red Hat Enterprise Linux], with a focus on code maturity, stability, and security.”

THE WHOA

The afternoon session promised a live demo of Red Hat Enterprise Linux OpenStack Platform, and Oxenham delivered. After logging in as an unprivileged user, Oxenham walked the crowd through a tour of the dashboard and a step-by-step guide to creating a workflow for launching an instance. A moment of suspense…and the instance was successfully launched (we had all the faith in the world, Rhys).

After briefly reviewing the security group rule options, his virtual machine (VM) became active, and he showed the overview, log, and console views of a Red Hat Enterprise Linux 6 VM running on top of OpenStack.

In just under 8 minutes, Oxenham showed how to use Red Hat Enterprise Linux OpenStack Platform to launch a successful VM instance—a fitting conclusion to a solid introduction.

More information

 

Event: Red Hat Summit 2014
Date: 3:40 p.m., Tues April 15, 2014
Type: Session
Track: Cloud readiness
Title: Introduction to Red Hat Enterprise Linux OpenStack Platform
Speaker: Rhys Oxenham (Red Hat)

Live from the Summit: Best practices for PaaS, OpenStack, & cloud adoption

13883319464_9c866cf43e_zDefining a cloud strategy for your organization comes with a lot of questions, but once you’ve answered the why, you need the how. Enter best practices for Platform-as-a-Service (PaaS), OpenStack, and more with industry expert David S. Linthicum and Red Hat cloud evangelist Gordon Haff.

After covering emerging standards in cloud adoption, Linthicum discussed 3 important questions to ask yourself as you look for solutions to fit your IT needs:

1. What is open and extensible?
2. What is cost-effective?
3. What meets your damn requirements?

That led him to best practice #1: Open your mind. “Coming in with a pure-born view of everything you’re going to leverage and ‘damn if it doesn’t meet your requirements’ is a dangerous approach,” said Linthicum. An open mind is essential for a mix-and-match environment, where you have to find what just works. “What impresses me about Red Hat is that it works and plays with other stuff really well,” he said.

The audience comprised a variety of verticals—when asked, hands shot up representing transportation, healthcare, tech, retail, finance, and more. That certainly speaks to the broad appeal of open cloud solutions, but Linthicum brought up an interesting and often overlooked trend: “The larger the industry gets, the more likely they are to fail because they don’t like to share [IT services or knowledge],” he warned.

That brought us to best practice #2: Go hire someone with a brain. “You need someone who can make the appropriate calls so that you’re marching in the right direction,” said Linthicum.

Most cloud-based systems are lacking architecture, and what’s more, solutions architects can get too narrowly focused on their own areas. “Typically, people aren’t going to have a range of skills that lets them be agnostic architects to make the right decision from all available choices,” Linthicum said. Hence, the need for open minds and sharp brains.

CAUGHT IN THE TRAP

One of the common pitfalls organizations make when investing in cloud resources is when that investment ignores training, or proofs of concept, or support. As a result, he described, many clouds are not meeting expectations.

Additionally, customers get caught up in the technology itself sometimes. “They call up and the first thing they want to know is ‘what’s the best out there? Amazon? Google? Red Hat?’ instead of asking ‘what’s the best solution for me?” said Linthicum.

That reinforced the advantage of keeping an open mind and choosing open cloud infrastructures.

13883316664_ff9c1f7d0c_zA FIRESIDE CHAT

The session pivoted from structured presentation to a “fireside chat” of sorts as Red Hat’s Gordon Haff steered the hour into audience interaction, which he found in spades.

One audience member asked about adoption habits or trends with PaaS, which coincided with another question about multi-hypervisor strategies. Linthicum explained that a lot of PaaS use is initially small at first but increasing to more mission-critical apps. “If you look at IDC and Gartner, they show a lot more multi-hypervisor use out there. It’s becoming the norm,” he said. “People aren’t throwing out VMware but they’re initially adopting KVM, RHEV, and Hyper-V for new types of projects so they don’t have to increase their VMware spending.”

Another audience member asked about private/public hybrid infrastructure approaches to PaaS. Haff described the variety of options within the OpenShift portfolio—Red Hat’s offering—including OpenShift Online for public and OpenShift Enterprise for private PaaS. Both got ringing endorsements from Linthicum. “Theirs is pretty much the only one in the industry that just works right now. It’s rock-solid, and I have no problems saying that” he said.

 

More information

 

Event: Red Hat Summit 2014
Date: Tues, April 15, 2014
Type: Session
Title: Best practices for PaaS, OpenStack, & cloud adoption
Speaker: David S. Linthicum (Cloud Technology Partners), Gordon Haff (Red Hat)

Live from the Summit: Why mobile, DevOps, & cloud trends matter

IT transformations are pervading the sessions and discussions at Red Hat Summit 2014. But why does it matter to the typical enterprise? It matters because enterprises are, more and more, becoming technology companies. According to Michael Coté, Research Director for Infrastructure Software at 451 Research:

cote-analyst-session“Increasingly companies like Nike and Starbucks that are relying on in-house software development for new products such as the Fuelband and mobile payments. Starbucks, for example, is estimated to have pulled in $1bn in sales from its mobile app.” For another example, from 451 Yankee Group’s “2013 US Mobile Applications and Cloud Survey (IT Decision-Maker) December,” 61 percent of respondents said that mobile app platforms had a “high” impact on their organizations’ profits. For still another, consider how entire businesses such as Uber simply wouldn’t be possible without mobile application development.

In other words, by taking advantage of cloud capabilities—such as improved developer productivity and faster application creation—top companies are making lots of money. Across verticals, there’s a pull to instrument businesses. Coté notes that “’Cloud’ is opening a new way of delivering software: DevOps. Cloud, in this context (applications-cum-business services) is the enabler, the platform, the underlying physics and grammar that enables good poetry and lyrics.”

So how to get started? According to Coté, start out with planning and segmenting out your applications and workloads. Consider what’s differentiating and custom from the tactical (which will increasingly shift toward SaaS). Get first-hand experience and do proofs-of-concept. Do small things while plodding through big problems as you start your proofs-of-concept. And consider culture and people.

Coté then discussed how there are two ways to think about the size of the cloud market. The first is the infrastructure public cloud which takes out SaaS and counts everything else. We already have a $5.7 billion market and it’s growing to about $20 billion by 2016. The other way to think about the market is cloud enablement technologies—i.e. the software used to run cloud “things” like OpenStack or cloud management software. This has been an equally great market, starting at about $10.6 billion in 2012 growing to over $22 billion in 2016.

What workloads are running on-premise versus public clouds? The applications that tilt furthest towards on-premise infrastructure (and private hosted) are customer-facing and back-office enterprise applications. Coté noted that, while ERP applications aren’t likely to move to more dynamic cloud environments, it is possible to wrap APIs around them and otherwise access them from cloud-enabled workloads.

Recently, 451 also started doing some research on the mainstream (i.e. not Netflix or Etsy) DevOps market. The first question asked was where their production apps reside. Few are running purely on a public cloud. So you’re not “doing it wrong” if you run it in other environments.

What’s the maturity of being able to deploy into production? Coté calls the overall result very encouraging. Almost all deploy more frequently than 30 days. Furthermore, about half still want to drive their deployment cycles down further—an encouraging aspirational trend. (Of course, many of the big-name web firms deploy to production multiple times per day, but this is not really a reasonable near-term goal for mainstream companies that are simply not organized around this sort of pace.)

These mainstream DevOps companies are mostly using testing, performance monitoring and log management, release management, and configuration management tooling. You’re still not seeing a lot of the new “utopic” startup toolchains that are most associated with DevOps. And only about 16 percent are using automation tools compared to using “older ways” of doing builds such as customer-written build tools and golden images. (Among those using automation, continuous integration tools such as Jenkins and Bamboo are the most common but there’s a lot of DIY tools out there too.)

The bottom line? “There’s strong business demand, work to be done as far as the eye can see, and lots of maturing ahead of us.”

 

More information

 

Event: Red Hat Summit 2014
Date: Tues, April 15, 2014
Type: Session
Title: What cloud trends mean for you: An analyst's view
Speaker: Michael Coté (451 Research)

Live from the Summit: Introducing the cloud technical roadmap

summit2014-cloud-roadmap-session

Tim Burke, vice president of Linux and Cloud Engineering at Red Hat, was joined by 6 cloud engineering leaders to share a detailed technical cloud roadmap. This lightning talk featured experts from various cloud engineering technology groups, including systems management, virtualization, OpenStack, containers, cloud management, and identity management.

Burke gave an overview of his 14 years at Red Hat, describing the transition to cloud as similar to the shift from Red Hat Linux to Red Hat Enterprise Linux. “[Early-stage Linux] was for rocket scientists. You had to piece it together yourself.” He described cloud as intimidating to enterprises, “Cloud [today] is still in the rocket scientist camp.” Red Hat’s goal, he explained, was to make cloud consumable for customers—just like what Red Hat helped do for Linux. Communities, companies, and partnerships drive this demand for innovation, integration, stabilization, and delivery.

“It’s all about our upstream focus—it’s how we innovate,” Burke said. And making cloud consumable involves many—maybe even all—Red Hat products.

RED HAT SATELLITE

Bryan Kearney, senior manager in Software Engineering at Red Hat, described how Satellite releases continue to become more open. Enhancements in systems and subscription reporting were included in the most recent release, Red Hat Satellite 5.6. Red Hat Satellite 6 will change more dramatically, and is intended to be a re-imagined revision. Kearney assured the audience that customers will not be rushed to the new version. “It’s a little daunting, so that’s why we’ve done [an] extended beta program and a long life cycle.”

Expanded features in Red Hat Satellite 6 should include:

  • More provisioning
  • Recipe-style configuration management
  • Refined life cycle management
  • Modern administration dashboards
  • Simplified content management
  • Drift management
  • Federated services and management
  • The ability to deploy on VMware, Red Hat Enterprise Virtualization, Amazon EC2, and OpenStack

RED HAT ENTERPRISE VIRTUALIZATION

Andrew Cathrow, the director of product management for Virtualization at Red Hat, described Red Hat Enterprise Virtualization as the open source competitor to VMware vSphere, containing all the features you’d expect from datacenter virtualization. The next release, Red Hat Enterprise Virtualization 3.4, plans continue improvements in:

  • Storage, such as mixed domains, single-disk snapshots, and read-only disks.
  • Networking, such as neutron integration, security groups, and IP address management.
  • Compute, such as scheduler enhancements, high availability reservations, and hotpluggable CPUs.
  • Infrastructure, including support for Red Hat Enterprise Linux 6.

RED HAT ENTERPRISE LINUX OPENSTACK PLATFORM

Mark McLoughlin, manager of software engineering and consulting engineer at Red Hat, as well as a leading contributor to OpenStack projects, talked about Red Hat’s involvement and influence on the OpenStack community—and the reciprocal OpenStack influence on Red Hat. “OpenStack is exactly the kind of fast-moving emerging project we like to see. It brings a lot of exciting innovation around.” McLoughlin also claimed that a Red Hat foundation is the best platform for running OpenStack—and a stable base for cloud.

Red Hat Enterprise Linux OpenStack platform releases generally follow OpenStack project releases by 2 months. Red Hat Enterprise Linux OpenStack Platform 5, based on the upcoming Icehouse, is slated to include:

  • Trove
  • Sahara (was Savanna) technology preview
  • Support for Red Hat Enterprise Linux 6.5 and 7 beta

McLoughlin also outlined possible future enhancements, which could include:

  • Smoother upgrades
  • Support for Rabbit MQ
  • Support for active/active databases
  • Scalability and reliability improvements
  • TripleO standardization around management and deployment
  • Virtual networking advances and use cases

DOCKER, CONTAINERS, AND THE ATOMIC PROJECT

Daniel Riek, senior director of Software Engineering at Red Hat, succinctly described the emerging technology Docker: “It’s a big deal for us.”

“We are seeing a changing paradigm,” he said, “…a continuity of how IT deployment has evolved.”

Traditional ecosystems built around runtimes led to stable infrastructure that lacked flexibility. As demands increased, we looked to virtualization. The downside: complex management and security. “We think that application-centric IT and PaaS really changes things. [Containers let us] package apps with their runtime, but without their own OS. Shared services is still a part of the overall OS provided outside the containerized application. It’s the best of both worlds,” said Riek.

The Atomic project further expands this utility, creating Linux that is optimized for the container environment. The Atomic project includes:

  • A pattern for an optimized container host
  • A minimal OS to run containers
  • Minimal patching
  • Atomic updates through OSTree
  • Standardized core shared services
  • Orchestration primitives

OpenShift 3 by Red Hat is built using Docker.

RED HAT CLOUDFORMS

Xavier Lecauchois, principal technical product manager for Cloudforms, came to Red Hat in December 2014 as part of the ManageIQ acquisition. Red Hat CloudForms is the Red Hat cloud management platform. “It provides a single pane of glass [approach] to manage hybrid cloud,” said Lecauchois.

Red Hat CloudForms currently supports:

  • Red Hat Enterprise Linux OpenStack Platform
  • Red Hat Enterprise Virtualization
  • VMware
  • Amazon Web Services (AWS)

Planned feature advancements for Red Hat CloudForms may include:

  • Initial support for Red Hat Enterprise Linux OpenStack Platform, such as inventory, editing, capacity and utilization planning, self-service provisioning, and reporting
  • Improved support for Red Hat virtualization manager.
  • improved support for OpenStack
  • Content management

After much effort, Lecauchois announced that CloudForms should also be open-sourced this year.

IDENTITY MANAGEMENT

Dmitri Pal, senior manager in Software Engineering at Red Hat, discussed identity management improvements in Red Hat Enterprise Linux 7 beta. “We are innovating for SSSD and identity management,” he said. Current and future advancements in identity management may include:

  • Microsoft Active Directory integration—both direct and indirect
  • Free IPA for identity management in Red Hat Enterprise Linux
  • System security services daemon (SSSD)
  • Native two-factor authentication with Kerberos
  • Certificate authority management tools
  • Improved look and feel
  • Better access control
  • Improved Active Directory integration
  • DNSSEC availability
  • SSSD features, SSSD for Docker containers
  • Rich identity information over D-Bus

“What you’ve seen here,“ Tim Burke said in closing, “[is that] cloud means a lot of things. It’s a lot of the technologies put together. This is the most exciting thing about being at Red Hat right now.”

“We are weaving all the pieces together… it’s not just about Red Hat. It’s about how we build it together with you, our partners and our customers.”

More information

 

Event: Red Hat Summit 2014
Date: 10:40 a.m., Tue April 15
Type: Session
Track: Best of
Technical difficulty: 3
Title: Cloud technical roadmap
Speakers: Tim Burke, Bryan Kearney, Andrew Cathrow, Mark McLoughlin, Daniel Riek, Xavier Lecauchois, Dmitri Pal