Jonathan Bryce, Executive Director of the OpenStack Foundation, kicked off day two of the Summit by tying it back to the first day’s keynote about the challenge of meeting the scalability requirements of projects such as the Square Kilometer Array radio telescope.
The SKA project is expecting to require multi-exaFLOPS — that’s one quintillion (10^18) floating point operations per second — once it’s fully online. With those requirements, Bryce said, we need to abandon the old paradigm of picking winners and losers in the cloud, or seeing the cloud space as zero sum game. Instead, he proposed that we embrace a “positive sum cloud world.”
To do that, Bryce believes the OpenStack community needs to focus on three key principles: keep innovating, collaborate with other technologies and replicate success.
As many know, OpenStack began with just the Nova compute project and the Swift object storage project, but continues to grow to meet user needs. The challenge is to continue the pace of innovation to address opportunities such as NFV.
New innovation, however, doesn’t just mean creating new tech. It means collaborating and integrating with other technologies, such as Kubernetes and Amazon Web Services. To make innovation and collaboration a reality will require the OpenStack community to also focus on replicating success — not only in deploying applications but in fostering cultural change in IT and collaborating with different technologies.
This is all necessary, Bryce said, “because the future is multi-cloud.”
Coming off that introduction, Summit participants were then treated to a number of presentations and demos that illustrated what is current and what is possible for OpenStack in a multi-cloud world.
First up was a presentation and demo by Bryce and Elizabeth Joseph from the OpenStack Infrastructure team, which operates the world’s largest multi-cloud automated continuous integration, or CI, application. This CI system, running across multiple OpenStack-powered public and hosted private clouds, is used to test and review the community release of OpenStack, including 1.7 million test jobs in the past six months. The demo showed the power of aggregating multiple OpenStack cloud resources to scale an application.
A frequent challenge to adopting a multi-cloud strategy is the issue of interoperability. The common perception is that you can’t easily deploy the same application to different OpenStack clouds. To address that perception, a number of OpenStack vendors came together to take on the “Interoperability Challenge.”
Rackspace, along with 15 other OpenStack vendors, took to the stage to do a live demo using the same Ansible playbook to orchestrate deploying an enterprise-type workload to their respective clouds.
In recent years, collaborating with other technologies has chiefly meant integrating with container technologies like Docker and Kubernetes. Jose Avila from Crowdstar presented its use case for leveraging containers on OpenStack. Crowdstar traditionally built their gaming apps on virtual machines. But the need to control costs while extracting higher performance led Crowdstar to use the Ironic project in OpenStack to provision, and use bare metal servers. Next, Crowdstar adopted containers running on bare metal as a way to give control back to developers in terms of packaging software.
AWS has been friend and foe during the OpenStack’s short history. Work has been done to bake AWS compatibility into OpenStack compute, but there has also often been a strong sentiment to regard AWS as a competitor. The growing reality is, however, that customers are leveraging multiple clouds, and many include both OpenStack and AWS.
Madhura Maskasy, co-founder and vice president of product at Platform9, demonstrated a project they open sourced called OpenStack Omni, which modifies OpenStack so the Horizon dashboard and the OpenStack CLI can also be used to provision and manage AWS resources, including compute, network and storage.
The morning keynote ended with a panel discussion on “Containers and OpenStack.” Panelists included Michal Jastrzebski, the Kolla project PTL; Adrian Otto, the Magnum PTL and Antoni Segura Puimedon, the Kuryr PTL. Each of these projects focus on a specific aspect of container integration with OpenStack, including deploying OpenStack control services as containers, managing containers on OpenStack to host workloads and networking containers. If you did not realize it before, containers will be a huge part of OpenStack’s future.
Stay tuned next week for a Summit recap from a Rackspace perspective.