At Indiana University, OpenInfra Days North America 2024 was an event that brought together the brilliant minds of the open infrastructure community. For my teammate Kevin Jackson and I, this was not just another tech event; it was a long-overdue reunion with friends in the OpenStack community and an exciting opportunity to forge new relationships. The atmosphere was charged with collaboration and learning, with best practices, user journeys, and insightful panel discussions taking center stage. The diversity of perspectives, from both seasoned veterans and newcomers, created a melting pot of ideas, pushing the boundaries of what open infrastructure can achieve.
Inspirational Speakers & Community Heroes
The lineup of speakers was nothing short of impressive. Presentations from organizations like NASA and the University of Alabama were not just informative—they were inspiring. The innovation and challenges discussed by Jonathan Mills of NASA’s Goddard Space Flight Center emphasized the power of open infrastructure to tackle scale. He elaborated on the “Next Generation System” being built at Goddard, which is designed from the start to leverage modern cloud-native methods and best practices to flexibly deploy baremetal, VMs, or containers from a single pool of resources. Another notable talk came from John-Paul Robinson HPC Architect and Manager at the University of Alabama at Birmingham (UAB). He discussed how academia continues to benefit from and contribute to the evolution of OpenStack and open-source projects – by highlighting a CI/CD infrastructure that supports development and testing of services in support of campus HPC operations.
Perhaps one of the most unexpected but influential presentations came from a former government official from Mexico. Federico Christian González Waite, an innovator and public policy expert, explored how Mexico’s regulatory framework is strategically designed to foster the adoption and growth of open source, driving digital transformation, democratization, and inclusion across the nation. His insights into how open infrastructure played a role in government operations showed how far-reaching and adaptable these technologies are, touching sectors beyond the usual commercial and the need into educational circles of government to fuel growth.
VMware to OpenStack Panel
The VMware to OpenStack Panel was moderated by OpenInfra Foundation’s Jimmy McArthur, and was joined on stage by Trilio’s Kevin Jackson, Mohammed Naser from Vexxhost, Maciej Lecki from Red Hat and DJ Dongjoon Min from ZConverter. Although the session was quite short to cover all topics of migrating from VMware to OpenStack, the panel covered a lot of ground. An interesting observation was that 3 out of 4 panelists’ software uses virt-v2v, as their solution or core to their solution, Trilio included. Maciej had a session the following day to walk through virt-v2v in more detail, and talk about the tool in a very approachable, and honest way. Questions ranged from recommendations to people who are considering migrating to OpenStack, with budget and time constraints, to feature discrepancies between VMware and OpenStack today. There was a balanced collective response on approaches, such as being prepared by knowing what you want to migrate, to utilizing partners. Kevin assured the audience that whilst OpenStack and VMware are different platforms, fundamentals exist between them, and that, using the 80-20 rule, 80% of the workloads (low hanging fruit) can probably migrate without issues. It is the remaining 20% that will take up 80% of the time. Move what you can, learn from it, and use that learning for the remaining complex workloads.
On the topic of feature parity/discrepancies, Jimmy said that the VMware to OpenStack SIG are looking at feature parity, to help ease transition from VMware to OpenStack. The group unanimously agreed that DRS (Dynamic Resource Scheduler) and SRM (Site Recovery Manager) are the key features that people expect, or want, to have with OpenStack to ensure continuity of service once migrated. Kevin took a more controversial approach and warned about decisions to make OpenStack like VMware would restrict future development, and reminded him of early days of OpenStack whereby a core group of original supporters and contributors wanted feature parity with Amazon’s EC2 API – where it was decided against copying features of Amazon because it would not allow OpenStack development to determine its own future.
A good example is Eucalyptus and CloudStack, two products that adopted that approach but are now considered legacy cloud products today.
Speaking to many people after the talk, they agreed that caution needs to be applied here. We shouldn’t approach this topic as migrating from VMware to Cheap VMware, because it misses a great opportunity to capitalize on new features and direction of future developments for those that have migrated to cloud from traditional legacy. However, improving the scheduler to allow for flexibility of resource usage (DRS) is welcomed if tackled appropriately, in the manner of cloud expectations of cloud. And, in fact, Masakari project (for HA of instances), would also be something that appeals to VMware admins who are used to relying on infrastructure to provide continuity of service to VMs. Masakari was a presentation by Zain Marvi an OpenStack Architect from Rackspace during Wednesday’s schedule. It is also worth noting that whilst ZConverter were attempting to suggest they might be in a good position to provide SRM features, but as Trilio is already a core backup product for OpenStack since the very early days of OpenStack, that already provides DR capabilities, Trilio is in a prime position to solve this for the community.
The Re-Emergence of Rackspace: A Game-Changer
Amidst all the engaging sessions and networking, one announcement made the biggest waves: the re-emergence of Rackspace. After years of speculation and shifts in strategy, Rackspace is once again positioning itself as a leader in the OpenStack ecosystem. This is not just good news for the OpenStack community, but a game-changer for Rackspace customers and anyone contemplating a move away from legacy platforms like VMware.
Kevin Carter and Don Norton, OpenStack leaders at Rackspace, presented their return during the Keynotes, and Principal Architect James Denton reiterated their divergent history in long-time coming presentations on the future direction of OpenStack and its core to Rackspace going forward. In one of the worst kept secrets of early OpenStack, Rackspace’s journey with OpenStack had two paths: the public cloud, and private cloud. Rackspace’s public OpenStack cloud was made up of some core OpenStack products, but included many internally developed services and features. Their managed private OpenStack cloud product was more aligned to the community. This proved problematic for Rackspace as feature and version gaps become increasingly unmanageable. In a return to OpenStack, Rackspace has a unified cloud – both public and private would use the same core, upstream products and versions, based on OpenStack-Helm, acknowledging the growing trend of having Kubernetes managing the Container services that many OpenStack distros are built upon. It’s worth noting that their current private cloud offering is based on OpenStack-Ansible, the project that they started in 2015. Trilio supports both OpenStack-Ansible and OpenStack-Helm.
The whispers about the next version of OpenStack were prevalent, but the buzz around Rackspace’s return felt monumental. As businesses look for reliable alternatives in their infrastructure journeys, Rackspace’s resurgence signifies new hope for scalable, open-source solutions that align with enterprise needs.
Here is a recent Press Release by Rackspace about their reentry into the community:
Looking Ahead: A Bright Future for Open Infrastructure
For those of us who have been involved in OpenStack from the early days, this year’s OpenInfra Days felt like a rejuvenation. The shared enthusiasm from all participants—whether they were contributors, users, or enthusiasts—indicated that open infrastructure is more relevant than ever.
With organizations like NASA, Bloomberg, UAB, and various governmental bodies relying on Open Infrastructure, the ecosystem continues to expand its influence. And now, with Rackspace back in the fold, the future for cloud infrastructure, especially for those seeking freedom from proprietary systems like VMware, looks even brighter.
Attending OpenInfra Days North America 2024 reaffirmed why Kevin and I are proud to be part of this community. It’s about innovation, collaboration, and ensuring that everyone has the tools they need to succeed in an open, scalable infrastructure landscape. Here’s to the future—and to more reunions and collaborations along the way…Thank you Indianapolis…see you all at the 2025 OpenInfra days in California!
