CoreOS CEO: We’re not out to replace Docker, just its containers

There was a major shakeup in the world of container-based computing this week when operating system provider CoreOS decided to get into the container space with a new open source project called Rocket. It’s a container runtime environment as well as a set of specifications for how App Containers — what CoreOS calls its container images — are built and function. But the bigger news industry-wide was the suggestion from CoreOS that it built Rocket because developer darling Docker isn’t living up to expectations.

CoreOS Co-founder and CEO Alex Polvi came on the Structure Show podcast this week to clarify that message and to explain the rationale behind Rocket and everything CoreOS does. If you’re interested in the future of containers, distributed systems and even cloud computing, both business-wise and technologically, it’s a must-listen interview. Here are some highlights, but there’s a lot more good stuff.

[soundcloud url=”” params=”color=ff5500&auto_play=false&hide_related=false&show_comments=true&show_user=true&show_reposts=false” width=”100%” height=”166″ iframe=”true” /]

Download This Episode

Subscribe in iTunes

The Structure Show RSS Feed

We’re fine with Docker, really!

If there’s one point that Polvi really wants to get across, it’s that CoreOS didn’t build Rocket because it doesn’t like Docker — either the technology or the company. He called that notion — expressed by the media, as well as, in numerous fora, Docker founder and CTO Solomon Hykes — “fundamentally flawed.”

The rationale behind Rocket is simple, Polvi explained. Docker is turning into more of a platform, adding in features around cluster management, networking and booting cloud servers, and CoreOS wanted to make sure that the original, simple container component didn’t get lost to the world as that happens. In fact, he says he’s fine with the idea of a Docker platform:

[blockquote person=”” attribution=””]”That’s a fine product, the private cloud is an open territory right now still. So the Docker platform is a product that needs to exist. We just want the simple composable building block to also exist for people that have their own platforms or they’re trying to build their own platform to use as a reusable component.”[/blockquote]

Although, below the surface, it might not be the mutual respect society the companies would like everyone to think it is. Later, while comparing Docker’s move away from containers to VMware’s move away from virtual machines, Polvi noted, “There is a debate as to whether the technology warrants another company like VMware to emerge.”

CoreOS CEO Alex Polvi

CoreOS CEO Alex Polvi

We build what we have to

When you consider the CoreOS business strategy, the reasons for Rocket begin to look a little more clear. Polvi calls the CoreOS lineup of technologies, which also includes a database, registry service, cluster management and other pieces, “a platform for platform builders.” It’s building the “primitives” that people need to build next-generation distributed systems and platforms, as opposed to actually building the platforms (think Heroku or CoreOS partner Deis) where people ultimately deploy applications.

“We are never trying to just take somebody else’s solution and build it,” Polvi said. “We’re trying to fill in the white space and build something that’s technically sound in an area we think is an open problem.”

He contrasts this with Docker, which he says is now becoming more akin to cluster (and container) management plays such as Mesosphere and the Kubernetes project, or VMware. Those technologies might use containers and let users move them around and manage them, but they’re far more about the management aspect than about the containers, or any other pieces of infrastructure, themselves.

Kubernetes works levels above the container, which isn't mentioned on this diagram from Microsoft.

Kubernetes works levels above the container, which isn’t mentioned on this diagram from Microsoft.

In fact, despite the fact that CoreOS has its own cluster-management tool, called Fleet, Polvi said the company actually contributes quite a bit to the Google-led Kubernetes project because it really likes the technology and the trajectory the project is on.

“Docker was a similar thing early on,” he added. “We used it for a year, we collaborated heavily with that community, but then it became clear they were on a trajectory that was no longer what we needed — and what a lot of people needed, not just us.”

Still, Polvi noted, technically, there’s no reason why Docker containers and Rocket can’t coexist provided Docker is willing to work within CoreOS’s container specifications or collaborate with CoreOS to develop a standard container format.

Structure 2010: Sebastian Stadil – CEO, Scalr; William “Skip” Bacon – VP of Products and CTO, Virtual Instruments; Michael A.Jackson – Co-Founder, President, and COO, Adaptive Computing; Jagan Jagannathan – Founder and CTO, Xangati; Alex Polvi – CEO and Co-Founder, Cloudkick; Javier Soltero – CTO for Management Products, SpringSource

A younger Polvi (far left) talking cloud a Structure 2010.

A quick thought on the cloud

We also asked Polvi about the world of cloud computing, where he used to work after Rackspace acquired his last startup, CloudKick, and where many CoreOS workloads will likely run. Maybe old allegiances just die hard, but Polvi thinks Rackspace is actually in a pretty good position as bigger cloud providers such as Amazon Web Services, Google and Microsoft continue to drive down prices.

“Now, because of the competitive pressure of the cloud providers, compute on infrastructure will go asymptotically to free over time, as well,” he said. “If you think about it, what’s left after the hard parts of software are free and the compute itself is relatively free, or free enough? … I think it’s service, that’s how you do it. You help people use all this stuff.”

After years of touting its cloud computing tech, Joyent open sources it

After repositioning itself behind its years-old container-based approach to cloud infrastructure, Joyent has now open sourced the code underlying its distributed cloud infrastructure and storage systems. For a company long heralded as a tech leader, its time to shine is now or never.

VMworld 2014 – Highlights & Analysis

VMworld, the flagship annual event hosted by VMware, is bigger and better with each passing year. This year’s event was one of the most anticipated conferences in the industry, withcustomers and partners expecting major announcements related to hybrid cloud and end-user computing, VMware didn’t miss the opportunity to underscore its commitment to enterprise through continued investments in core infrastructure, hybrid cloud and end-user computing.

In the last two years, VMware has consolidated its position by focusing on what really matters to enterprise customers. With the acquisition of Nicira, AirWatch, Desktone and, most recently, CloudVolumes; VMware has a strong value proposition for customers. Despite the confusing portfolio and complex technology stack, VMware should be appreciated for simplifying their message based on Software Defined Data Center (SDDC), Hybrid Cloud and End-User Computing (EUC). In every conversation with customers, partners, press and analysts, VMware’s leadership team consistently positioned these three offerings as the core pillars of enterprise IT. Based on its own track record and credibility of the companies acquired, VMware has a unique distinction of having the “best-of-breed” technology stack.

Source: VMware

Source: VMware

The growing interest in Linux container technology and the momentum around OpenStack is a threat to VMware’s business. Docker offers a lightweight alternative to hypervisor-based virtualization that appeals to developers and IT administrators. OpenStack was created to check the growing dominance of Amazon Web Services on the public cloud and VMware on the private cloud. Though VMware found a backdoor to the OpenStack foundation through the Nicira acquisition, other members of the foundation always questioned its intentions. VMware attempted to address these two threats through the announcements made at VMworld 2014. It is also leveraging its investments made in hybrid cloud and end-user computing by bringing them closer.

Here is an analysis of the key announcements made at VMworld 2014.

Emphasis on Software Defined Data Center (SDDC)
SDDC is an important piece of VMware’s strategy for sustaining existing business and acquiring new customers. Having conquered the hypervisor market through vSphere, VMware is aggressively moving into storage and network virtualization. It’s trying to repeat history with vSAN that virtualizes storage and NSX, its Software Defined Networking (SDN) offering. With enhancements made to Layer 2 VPN, firewall and load balancing, NSX 6.1, which was announced at VMworld, is one of the most advanced SDN offerings in the market. This puts VMware in the league of mature networking players like Juniper, HP and Brocade. The micro segmentation use-case of NSX transforms VMware into a strong network security company.

The other important announcement related to SDDC came in the form of the vRealize Suite, a platform that focuses on automation, operations and business. There is also a SaaS-based version of this called as vRealize Air. Though vRealize is more of a consolidation and rebranding of existing vCenter and vCloud family of products, the fundamental difference is in the support for 3rd party public clouds including AWS, Azure and Google Cloud. VMware is positioning this as a cloud management platform that extends the vCloud suite to manage OpenStack, AWS, KVM and, of course, its own hybrid cloud, vCloud Air. The SaaS version of vRealize competes with established cloud management platforms like RightScale, Scalr and Dell Cloud Manager.

The flagship product, vCloud Suite is upgraded to 5.8 with improvements in business continuity, disaster recovery and the ability to run Apache Hadoop 2 distributions with YARN cluster resource management.

With the new enhancements, SDDC becomes the blueprint for its converged infrastructure, private cloud and hybrid cloud offerings.

EVO:RAIL & EVO:RACK – Software defined data center in a box
As a key stakeholder of VCE, VMware contributed to the Vblock systems portfolio to deliver converged infrastructure, which is considered to be expensive by many customers. Branded as “Hyper-converged infrastructure”, EVO:RAIL and EVO:RACK provide choice for customers to choose from a variety of OEMs certified by VMware. EVO:RAIL targets the enterprises that may run several hundreds of VMs while EVO:RACK is meant for service providers delivering scalable workloads. With this initiative, VMware has officially joined the Open Compute Project founded by Facebook.

Though it may appear that VMware is getting into the hardware business with the EVO family of products, it is only providing reference platforms and certifying those reference platforms from OEM hardware vendors like Dell, Supermicro, NetOne, Inspur and Fujitsu to deliver the building blocks of the converged infrastructure. This is similar to the independent hardware vendor (IHV) program that Microsoft runs to certify Microsoft Windows compatible hardware. This will create a new converged infrastructure ecosystem in the industry offering an affordable choice to customers.

Source: VMware

Source: VMware

With EVO, VMware made new friends and foes in the industry. The aggressive push of SDN has already taken VMware into the territory of Cisco causing friction between the two companies. Cisco is positioning Application Centric Infrastructure (ACI) as the preferred SDN for VCE customers instead of VMware’s NSX. For delivering the EVO family of infrastructure, VMware has partnered with F5 and Cumulus, archrivals of Cisco. This further widens the gap between VMware and Cisco. The other company that may get impacted is Nutanix, which offers converged infrastructure that just closed a fresh round of funding of $140 million. Though the list of partners for EVO will grow in the future, HP is conspicuously missing. Given the push of converged infrastructure, VMware will become a direct competitor to HP.

OpenStack – Applying the embrace and extend philosophy
One of the most surprising announcements at VMworld 2014 was VMware Integrated OpenStack (VIO), VMware’s own distribution of OpenStack. For a long time, VMware executives downplayed the impact of OpenStack on their customer base but with the growing momentum followed by increased investments from Red Hat, HP, Mirantis and IBM forced them to rethink that strategy. Though VMware is a gold member of the OpenStack Foundation, its contribution is predominantly for the networking project codenamed Neutron. VMware made it to the top 10 contributors through the acquisition of Nicira. A quick look at the top contributors of OpenStack’s latest Icehouse release confirms this.



VIO is positioned as an enterprise friendly OpenStack distribution that is designed to run on top of the VMware stack. The compute, storage, networking and management components of OpenStack are tightly integrated with the equivalent building blocks of vSphere and vCenter family of products. Customers can use familiar vCenter tools to manage OpenStack-based infrastructure. Apart from its own distribution, VMware has partnered with HP, Mirantis and Canonical to support their distributions.

Source: VMware

Source: VMware

The official entry of VMware into the crowded OpenStack distribution space is to keep Red Hat at bay. It is the only company that comes close to VMware in the enterprise market. With their own hypervisor (KVM), Linux OS distribution (Red Hat Enterprise Linux), orchestration (CloudForms + ManageIQ), Storage (Gluster & Ceph) amd OpenStack distribution (RHEL OpenStack), Red Hat has all the essential pieces to deliver an end-to-end IaaS strategy. It also happens to be one of the top contributors of OpenStack code.

The industry is skeptical of VMware’s commitment to OpenStack. It is perceived as the ‘embrace, extend and extinguish’ tactic being used by the largest commercial private cloud provider. VMware has not articulated how OpenStack fits into their vision of hybrid cloud and SDDC-in-a-box powered by EVO. Their motivation seems to be to attract customers who are evaluating OpenStack as a potential alternative to vSphere. By embracing OpenStack, VMware can play the card of “best-of-breed” technologies to protect their customer base and securing licenses. Only time will tell how customers and the OpenStack ecosystem respond to this move from VMware.

Docker and Kubernetes – Join them if you can’t beat them
Docker has caused a stir in the industry with its simplified approach to container management. Many positioned Docker as an alternative to traditional virtualization that is dominated by VMware and Microsoft. Mature infrastructure providers like Rackspace and IBM SoftLayer have announced their plans to offer bare metal clouds powered by contemporary Linux distributions like CoreOS and Docker. Bare metal clouds avoid dependencies on hypervisors posing a threat to VMware.

At VMworld, VMware unveiled its plan for containers. Instead of competing with Docker, VMware is partnering with them to bring containers to its virtualization platform. VMware developed Cloud Foundry, an open source PaaS before it spun off Pivotal.  By leveraging Pivotal CF (commercial version) and Cloud Foundry (open source), it can bring container-based application deployment and delivery to its customers. That’s not all. VMware is also partnering with Google to support Kubernetes on private cloud and hybrid cloud platforms. Kubernetes is an open source orchestration layer backed by Google to provision, schedule and manage containers running on any infrastructure layer. With support from Microsoft, Red Hat and IBM, Kubernetes is moving towards becoming the de facto orchestration tool for Docker.  VMware may integrate vSphere APIs with Kubernetes to support orchestration of containers running with the VMs provisioned by vSphere. It can even add support to manage containers through the familiar vCenter environment making it easy and transparent for administrators. Eventually, VMware would unify container and VM manageability through vCenter and the vRealize family.

Though VMware and Docker jointly announced their plans to work together, there is an interesting project brewing within the company called Project Fargo, which brings rapid provisioning of VMs. VMware claims that Project Fargo can speed up provisioning by 30X. This technology when combined with CloudVolumes, takes VDI and DaaS to the next level, by enabling administrators to close and provision running images in just a few seconds. This can be extended to other workloads to bring rapid provisioning of server VMs running on private cloud and hybrid cloud. It will be interesting to see if VMware’s stance on Docker will change when Project Fargo becomes mainstream.

Source: VMware

Source: VMware

VMware executives were careful in the way they positioned Docker by consistently stating that “the best way of delivering containers in through VMs”. This is a defensive move by VMware before their competition steps up their assault against traditional virtualization. Red Hat has been at the forefront of integrating containers with its OS and PaaS causing a threat to VMware. Through its support for Docker and Kubernetes, VMware is playing it safe, by protecting its private cloud running on vCloud and its public cloud investments powered by vCloud Air.

vCloud Air – Rebranded and refreshed hybrid cloud strategy
With the rebranding of vCHS to vCloud Air, VMware wants to enter the top league of public cloud providers dominated by Amazon, Microsoft and Google. While still being positioned as the best public cloud for existing VMware customers, it wants to attract a new set of enterprise customers. VMware announced a plethora of new services such as DevOps as a Service, Disaster Recovery as a Service, DB as a Service, and object storage on vCloud Air. Based on EMC’s ViPR software-defined storage, vCloud Air object storage supports the popular S3 API, including lifecycle management and versioning features to simplify and reduce management overhead with data durability of 11 nines per object. Through its vCloud Air OnDemand, VMware is attempting to bring self-service capability to its hybrid cloud.

Source: VMware

Source: VMware

VMware has expanded the VMware Service Provider Program (VSPP) to vCloud Air Network. This program brings vCloud Air capability to more than 3900 partners of VMware in 100+ countries.  Service Providers will be classified into “IaaS Powered,” “Hybrid Cloud Powered,” and the existing “Horizon DaaS Powered” to highlight their key offerings.

By bringing Pivotal CF, Docker and Kubernetes to vCloud Air, VMware is trying its best to attract developers to its hybrid cloud. It has partnered with enterprise backend company, Kinvey to bring mobile backend capability. By combining it with AirWatch, vCloud Air transforms into a mature enterprise mobile management platform that differentiates itself from competing offerings.

Similar to its recent partnership with SoftBank and NetOne in Japan, VMware is expected to partner with mature providers in the EMEA and APJ regions to expand vCloud Air’s footprint. With its eye on enterprise workloads, VMware is clear that it doesn’t want to compete with Amazon to on-board web scale workloads. Its main competitor in this space is Microsoft, which is moving fast to capture the enterprise market.

Editorial Note on vCloud Air Hybrid Cloud Strategy: By offering DRaaS and other SaaS offerings, VMware can alienate its service providers by competing with them directly. Bill Fathers’ response to this concern was, “This means that now there are 3901 service providers instead of 3900.”. Competing with the channel is always a slippery slope and VMware will have to tread very lightly.

Leverage End-User Computing (EUC) and hybrid cloud investments
According to the Sector Roadmap on Virtual Desktops report unveiled at the Gigaom Structure conference in June, VMware emerged as a leading player in the DaaS market. With the announcements made at VMworld, VMware’s EUC story becomes stronger.

Source: Gigaom Research

Source: Gigaom Research

Acquiring Desktone and AirWatch has been a smart move by VMware to strengthen its market position. It is now moving towards leveraging vCloud Air and vCloud Air Network to deliver DaaS offering to enterprise customers.  Codenamed “Project Meteor”, VMware took the curtains off its partnership with Google and NVIDIA to bring best-in-class user experience to thin clients that only run HTML5 browsers. The joint effort uses Nvidia’s GRID virtual CPU, Nvidia Tegra K1 processors that come with Chromebooks, and VMware’s Blast HTML5 technology to deliver superior user experience. This puts both VMware and Google in a win-win situation by accelerating the adoption of vCloud Air and Chromebooks in the enterprise. Enterprises moving workloads to vCloud Air can use cost-effective Chromebooks to access them. This puts VMware ahead of Citrix in the DaaS market.

Apart from session-based and dedicated desktops, Horizon DaaS now supports delivering hosted applications to remote clients. Horizon Workspace Suite completes the DaaS offering by becoming a unified platform accessible via single sign-on to access and determine policy controls for applications regardless of location and operating system.

Source: VMware

Source: VMware

The latest CloudVolumes acquisition brings layering and application containerization techniques to virtual machines. This will help administrators provision desktops in a few minutes.

In another interesting move, Dell has partnered with VMware to offer Horizon DaaS through its Wyse Datacenter offering. With the acquisition of Quest, Dell has entered the DaaS market with vWorkspace. Though it has a lot of work to do in that space, vWorkspace is an affordable VDI/DaaS solution for small and medium businesses.  Dell is positioning Horizon DaaS for Enterprises while still selling vWorkspace for SMBs. This partnership will benefit VMware more than Dell.

End-user computing is one of the key pillars of VMware’s strategy and it seems to be moving in the right direction. As VMware consolidates its investments made in DaaS, MDM, application virtualization and hybrid cloud, it’s positioned to deliver great value to customers. Going forward, EUC will be the key driver to accelerate the adoption of vCloud Air.

Key takeaways
With crucial announcements, partnerships and product launches, VMworld 2014 is a milestone for VMware. The next two quarters will decide the impact of these announcements on VMware.

One key observation from VMworld is that VMware is, again, becoming a developer company. Though it made a conscious decision of moving its platform assets to Pivotal, market dynamics and competitive pressures have forced VMware to embrace OpenStack, Docker, MBaaS and DevOps. Embracing containers was an especially bitter pill for VMware to swallow since it has opposed containers for such a long time. These new offerings will transform it to a developer platform company. VMware has to be cautious by staying focused on delivering value to enterprises through its infrastructure offerings. The developer platform offerings may distract VMware from delivering the vision of SDDC and hybrid cloud. Instead of dealing with developer related products directly, it is best left to Pivotal, which is making great progress with Cloud Foundry and Big Data platforms.

OpenStack, is a tightrope walk for VMware. It’s under pressure to prove its commitment and sincerity to OpenStack by contributing beyond Neutron. VMware may also be forced to open source certain elements of its stack to increase its contribution to the cause.

Another interesting observation is how little Microsoft was mentioned as a competitor by VMware. Having won the hypervisor battle, VMware is moving up the stack to fight it out with traditional networking companies and public cloud providers. Most of the announcements and partnerships targeted Cisco, Citrix, Amazon and Red Hat.

It’s definitely not a “Winner-takes-all” market anymore. Time will tell if the bets placed by VMware will pay off in the long term.

Seven Things the CIO should consider when adopting a holistic cloud strategy

As conversations about cloud computing continues to focus on IT’s inability at holistic adoption, organizations outside of IT continue their cloud adoption trek outside the prevue of IT. While many of these efforts are considered Shadow IT efforts and frowned upon by the IT organization, they are simply a response to a wider problem.

The IT organization needs to adopt a holistic cloud strategy. However, are CIOs really ready for this approach? Michael Keithley, Creative Artists Agency’s CIO just returned from CIO Magazine’s CIO 100 Symposium which brings together the industry’s best IT leaders. In his blog post, he notes that “(he) was shocked to find that even among this elite group of CIOs there were still a significant amount of CIOs who where resisting cloud.” While that perspective is widely shared, it does not represent all CIOs. There are still a good number of CIOs that have moved to a holistic cloud strategy. The problem is that most organizations are still in a much earlier state of adoption.

In order to develop a holistic cloud strategy, it is important to follow a well-defined process. The four steps are straightforward and fit just about any organization:

  1. Assess: Provide a holistic assessment of the entire IT organization, applications and services that is business focused, not technology focused. For the CIO, they are a business leader that happens to have responsibility for technology. Understand what is differentiating and what is not.
  2. Roadmap: Use the options and recommendations from the assessment to provide a roadmap. The roadmap outlines priority and valuations that ultimately drive the alignment of IT.
  3. Execute: This is where the rubber hits the road. IT organizations will learn more about themselves through action. For many, it is important to start small (read: lower risk) and ramp up quickly.
  4. Re-Assess & Adjust: As the IT organization starts down the path of execution, lessons are learned and adjustments needed. Those adjustments will span technology, organization, process and governance. Continual improvement is a key hallmark to staying in tune with the changing demands.

For many, following this process alone is not enough to develop a holistic cloud strategy. In order to successfully leverage a cloud-based solution, several things need to change that may contradict current norms. Today, cloud is leveraged in many ways from Software as a Service (SaaS) to Infrastructure as a Service (IaaS). However, it is most often a very fractured and disjointed approach to leveraging cloud. Yet, the very applications and services in play require that organizations consider a holistic approach in order to work most effectively.

When considering a holistic cloud strategy, there are a number of things the CIO needs to consider including these six:

  1. Challenge the Status Quo: This is one of the hardest changes as the culture within IT developed over decades. One example is changing the mindset that ‘critical systems may not reside outside your own data center’ is not trivial. On the other hand, leading CIOs are already “getting out of the data center business.” Do not get trapped by the cultural norms and the status quo.
  2. Differentiation: Consider which applications and services are true differentiators for your company. Focus on the applications and services that provide strategic value and shift more common functions (ie: email) to alternative solutions like Microsoft Office 365 or Google Apps.
  3. Align with Business Strategy: Determine how IT can best enable and catapult the company’s business strategy. If IT is interested in making a technology shift, consider if it will bring direct positive value to the business strategy. If it does not, one should ask a number of additional questions determining the true value of the change. With so much demand on IT, focus should be on those changes that bring the highest value and align with the business strategy.
  4. Internal Changes: Moving to cloud changes how organizations, processes and governance models behave. A simple example is how business continuity and disaster recovery processes will need to change in order to accommodate the introduction of cloud-based services. For organizations, cloud presents both an excitement of something new and a fear from loss of control and possible job loss. CIOs need to ensure that this area is well thought out before proceeding.
  5. Vendor Management: Managing a cloud provider is not like every other existing vendor relationship. Vendor management comes into sharp focus with the cloud provider that spans far more than just the terms of the Service Level Agreement (SLA).
  6. Exit Strategy: Think about the end before getting started. Exiting a cloud service can happen for good or bad reasons. Understand what the exit terms are and in what for your data will exist. Exporting a flat file could present a challenge if the data is in a structured database. However, that may be the extent of the provider’s responsibility. When considering alternative providers, recognize that shifting workloads across providers is not necessarily as trivial as it might sound. It is important to think this through before engaging.
  7. Innovation: Actively seek out ways to adopt new solutions and methodologies. For example, understand the value from Devops, OpenStack, Containers and Converged Infrastructure. Each of these may challenge traditional thinking, which is ok.

Those are seven of the top issues that often come up in the process of setting a holistic cloud strategy. Cloud offers the CIO, the IT organization and the company as a whole one of the greatest opportunities today. Cloud is significant, but only the tip of the iceberg. For the CIO and their organization, there are many more opportunities beyond cloud today that are already in the works.