Revisiting OpenStack – Design Considerations
Since the last edition of the Mastering OpenStack series, the OpenStack community has kept the momentum by growing and innovating around its ecosystem. At the time of writing this edition, the OpenStack design has been going through different cycles of improvements by including new projects and providing seamless integration with existing systems to respond to organization demands and custom features. Since the first Austin release back in 2010, the innovation kept opening new opportunities for companies and quickly adopting a stable private cloud setup to stay ahead in the market. The challenges of handling a scalable infrastructure have been felt by big and medium players who joined the OpenStack evolvement. Today, as the first page of this edition is being written, 14 years and 28 releases have already passed through different OpenStack releases, from Austin back in the day to the latest Dalmatian release. That is a full circle back to the beginning of the alphabet. A lot of experiences and insights have been revealed that have boosted the OpenStack community and made it the world’s leading open source cloud platform.
Numbers show how successfully OpenStack has been adopted by mid to large enterprises. As per the annual OpenStack User Survey results in the OpenStack blog in November 2022, it was declared that more than 40 million cores are running in production across 300 deployments empowered by OpenStack.
The OpenInfra community has been committing enormously to hardening the most stable versions of OpenStack software. Starting from the Antelope release that was launched in March 2023 and, more recently, the Dalmatian release in October 2024, organizations have quickly upgraded, and the good news is a more stable and reliable version than ever. The secret sauce of this great achievement is that the community has switched gears on stabilizing basic OpenStack services, including compute, storage, and network services, increasing the cadence of release testing of each service extensively and thus leaving neither possible gaps nor potential vulnerabilities that could raise bugs in deployments in production. On the other side of the story, by going through the map of each OpenStack release, you might notice some extension projects appearing and disappearing. Community contributors did the right thing, discounting non-stable features that would hinder the maturity of the OpenStack private setup. Today, the release is exposing basic services in addition to more projects and features in a stable upstream. The other bright side of the story is that the OpenStack ecosystem has always been one step ahead to adopt new trends of market technologies that include containerization, serverless, big data analysis, and so on.
Today, either starting a new OpenStack adventure or turning a new page for a new update would bring great cost savings, flexibility in IT handling compared to old-school virtualization alternatives, and an efficient solution to hyperscale with minimum concerns.
OpenStack’s state-of-the-art documentation and resources can be found more abundantly than ever on the internet. There are plenty of different options to design and deploy a complete OpenStack environment. On the other hand, the paradox of choices between different ways of designing and running a full ecosystem can be challenging even for experienced administrators, architects, and developers. That could be phrased differently as too much can be too little! The adoption of OpenStack can be a struggle and has always been a challenge, from design phases to deployments and operational days. As stated previously, you can think of many reasons for this, but mainly, being such a versatile software can make it overwhelming!
Throughout this book, we will go through the updated pieces of the OpenStack ecosystem and divide and conquer together with a step-by-step approach to design, deploy, and operate a scalable OpenStack environment that would respond to your needs and requirements.
To address the challenges of the OpenStack ecosystem complexity and allow newcomers to enjoy this cloud journey, we will define a state vision for a private cloud adoption strategy throughout the first chapter. Without a well-crafted basic knowledge of this ecosystem, it would be more difficult to make later decisions on empowering running clusters in production when production days start knocking on the door. A very attractive citation from Robert Waterman, an expert on business management practices, is “A strategy is necessary because the future is unpredictable.”
When it comes to a complex system such as the OpenStack ecosystem, setting up the right resources is required as we cannot predict with 100% accuracy. OpenStack is designed with much more flexibility and a loosely coupled architecture. This way, it is up to us to use those key elements to make capacity decisions considering short- and long-term goals for growth and availability and respond to new workloads based on existing resources that we aim to apply in the near and long term.
Our OpenStack journey will continue in this edition by covering the following topics in this chapter:
- Revisiting and highlighting the latest updates of the OpenStack core ecosystem
- Demystifying the logical architecture based on the latest releases introduced from the Antelope release
- Drafting a first physical design to ensure a seamless deployment in later stages
- Preparing for a large-scale OpenStack environment through capacity planning as part of our cloud strategy
