There's no perfect way to implement OpenStack in your company, but there are general principles that can guide you toward a successful deployment.
One of the advantages of open-source software, such as OpenStack, is the ability to download it, test it in action, and gain practical insights without lengthy interactions with vendor companies or the necessity of prolonged internal approvals for a pilot project between your company and the vendor.
But what happens when it’s time to do more than just try out the project? How will you transition from source code to a production environment? How can you overcome organizational barriers to the adoption of new and transformative technologies? Where do you start? What will you do next?
Certainly, there's much to learn from the experiences of those who have already deployed OpenStack. To better understand the patterns of OpenStack adoption, I spoke with several teams that have successfully introduced this system within their companies.
MercadoLibre: The Dictate of Necessity and the Race to Keep Up
If the need is strong enough, then deploying a flexible cloud infrastructure can be almost as simple as 'build it and they will come.' In many ways, this is the experience that Alejandro Comisario, Maximiliano Venesio, and Leandro Reox had at MercadoLibre, the largest e-commerce company in Latin America and the eighth largest in the world.
In 2011, when the development department began the journey of decomposing its then-monolithic system into a platform of loosely connected services integrated via APIs, the infrastructure team faced a sharp increase in requests to their small team that needed to be fulfilled.
"The shift happened very quickly," says Alejandro Comisario, Technical Director of Cloud Services at MercadoLibre. "We literally realized overnight that we couldn't continue working at that pace without the help of some system."
Alejandro Comisario, Maximiliano Venesio, and Leandro Reox — the entire MercadoLibre team at the time, began looking for technologies that would allow them to eliminate the manual steps involved in providing infrastructure to their developers.
The team set themselves more challenging goals, formulating objectives not just for immediate tasks but also for the overarching goals of the company: reducing the time to deliver production-ready virtual machines to users from 2 hours to 10 seconds and removing human intervention from this process.
When they found OpenStack, it became clear that this was exactly what they were looking for. MercadoLibre's rapidly evolving culture allowed the team to move quickly in creating an OpenStack environment, despite the relative immaturity of the project at that time.
— It became clear that the OpenStack approach — research, diving into the code, and testing functionality and scalability — matched MercadoLibre's approach, — says Leandro Reox. — We were able to dive right into the project, define a set of tests for our OpenStack installation, and start testing.
Their initial testing on the second release of OpenStack identified several issues that prevented them from going into production; however, the transition from the Bexar release to the Cactus release occurred at just the right time. Further testing of the Cactus release provided confidence that the cloud was ready for commercial operation.
The launch into commercial operation and the developers' understanding of the ability to obtain infrastructure as quickly as they could consume it determined the success of the implementation.
— The entire company craved such a system and the functionality it provides, — notes Maximiliano Venesio, senior infrastructure engineer at MercadoLibre.
Nevertheless, the team was cautious in managing developers' expectations. They needed to ensure that developers understood that existing applications would not work on the new private cloud without modifications.
We had to ensure that our developers were prepared to write stateless applications for the cloud, said Alejandro Comisario. This was a huge cultural shift for them. In some cases, we had to teach developers that simply storing their data in the instance was not enough. Developers needed to adjust their mindset.
The team was attentive in training the developers and recommended best practices for creating cloud-ready applications. They sent emails, conducted lunches with informal training, and organized formal workshops, as well as providing proper documentation of the cloud environment. The result of their efforts is that MercadoLibre developers now develop cloud applications as comfortably as they did traditional applications for the company's virtualized environments.
The automation they achieved with the private cloud paid off, allowing MercadoLibre to scale its infrastructure dramatically. What began as a three-person infrastructure team supporting 250 developers, 100 servers and 1000 virtual machines, has grown into a team of 10 supporting over 500 developers, 2000 servers, and 12,000 VMs.
Workday: Building a Business Case for OpenStack
For the team at Workday, operating in the SaaS space, the decision to embrace OpenStack was more strategic than operational.
Workday's journey towards implementing a private cloud began in 2013 when company leadership agreed to invest in a wide initiative for a Software-Defined Data Center (SDDC). The hope for this initiative was to achieve greater automation, drive innovation, and enhance data center efficiency.
Workday crafted its vision for a private cloud among the infrastructure, engineering, and operations groups within the company, and an agreement was reached to initiate a research initiative. Workday hired Carmine Remi as the Director of Cloud Solutions to lead the changes.
Rimi's first task at Workday was to expand the initial business case to encompass a larger part of the company.
The cornerstone of the business case was to enhance flexibility in the use of SDDC. This increased flexibility would help the company achieve its desires for continuous software deployment with zero downtime. The SDDC API was intended to enable application development teams and the Workday platform to innovate in ways previously unavailable.
The efficiency of equipment was also considered in the business case. Workday sets ambitious goals for increasing the utilization rate of existing equipment and data center resources.
— We found that we already have middleware technology that can take advantage of a private cloud. This middleware software has been used for deploying dev/test environments in public clouds. With a private cloud, we could expand this software to create a hybrid cloud solution. By using a hybrid cloud strategy, Workday can migrate workloads between public and private clouds, maximizing equipment utilization and achieving cost savings for the business.
Finally, Rimi's cloud strategy noted that simple stateless workloads and their horizontal scaling would allow Workday to start using its private cloud with less risk and achieve cloud operations maturity naturally.
— We can move forward with the proposed plan and learn to manage the new cloud with a small workload, akin to traditional R&D efforts that allow for experimentation in a safe environment, — suggested Rimi.
With a solid business case, Rimi evaluated several well-known private cloud platforms, including OpenStack, based on a wide range of assessment criteria that included the openness of each platform, ease of use, flexibility, reliability, resilience, availability of support and community, as well as potential. Based on the results of his assessment, Rimi and his team chose OpenStack and began building a commercially viable private cloud.
Having successfully implemented their first viable OpenStack cloud, Workday continues to strive for broader adoption of the new SDDC environment. To achieve this, Rimi employs a multifaceted approach focused on:
- concentrating on cloud-ready workloads, especially on stateless applications in the portfolio
- defining criteria and the migration process
- setting goals for the development for migrating these applications
- engaging and training groups of Workday stakeholders using meetings, demonstrations, videos, and OpenStack trainings
— Our cloud supports various workloads, some in production, others preparing for commercial operation. Ultimately, we want to migrate all workloads, and I expect that we will reach a tipping point where we will see a sudden surge of activity. We are preparing the system in increments each day to be able to handle this level of activity when the time comes.
BestBuy: Breaking the Barriers
The electronics retail chain BestBuy has annual revenues of $43 billion and 140,000 employees, making it the largest of the companies mentioned in the article. Therefore, while the processes used by the infrastructure team at bestbuy.com to prepare a private cloud based on OpenStack are not unique, the flexibility with which they applied these processes is impressive.
To implement their first OpenStack cloud at BestBuy, web solutions director Steve Eastham and chief architect Joel Crabb had to rely on a creative approach to overcome numerous barriers that stood in their way.
The BestBuy OpenStack initiative began in early 2011 as an effort to understand the various business processes associated with the release processes of the e-commerce site bestbuy.com. These efforts revealed significant inefficiencies in the quality assurance processes. The quality review process incurred substantial costs with each major site release, which occurred two to four times a year. A significant portion of these costs was related to manually configuring the environment, reconciling discrepancies, and resolving resource availability issues.
To address these issues, bestbuy.com launched the 'Quality Check on Demand' initiative led by Steve Eastham and Joel Crabb, aimed at identifying and resolving bottlenecks in the bestbuy.com quality check process. Key recommendations from this project included automating quality check processes and providing user groups with self-service tools.
While Steve Eastham and Joel Crabb were able to leverage the perspective of significant quality control costs to justify investments in a private cloud, they quickly ran into a problem: although approval for the project was obtained, funding for the project was not secured. There was no budget allocated for equipment purchases for the project.
Necessity is the mother of invention, and the team adopted a new approach to funding the cloud: they swapped budgets with two developers from another team that had an equipment budget.
With the acquired budget, they intended to purchase the necessary equipment for the project. By reaching out to HP, their equipment supplier at the time, they focused on optimizing the proposal. Through careful negotiations and acceptable reductions in equipment needs, they managed to halve their equipment costs.
In a similar vein, Steve Eastham and Joel Crabb struck a deal with the company's networking team, leveraging the existing core capacity to save on typical costs associated with acquiring new networking equipment.
— We were on pretty thin ice, — said Steve Eastham. — This wasn’t a common practice at BestBuy either at that time or now. We operated below the radar. We could have been reprimanded, but we managed to avoid that.
Overcoming financial difficulties was just the first of many hurdles. At that time, opportunities to find OpenStack experts for the project were virtually non-existent. As a result, they had to build a team from scratch by combining traditional Java developers and system administrators.
— We just put them in one room and said, ‘Figure out how to work with this system,’ — recounts Joel Crabb. — One of the Java developers told us: ‘This is crazy; you can't do this. I don't know what you're talking about.’
We had to merge various styles from two types of teams to achieve the desired outcome – a programmatically controlled, testable, phased development process.
Motivating the team early in the project allowed them to achieve several impressive wins. They quickly replaced the outdated development environment, reduced the number of quality assurance (QA) environments, and through this transformation, established a new way of working for teams and increased application delivery speed.
Their success provided a strong opportunity to request additional resources for their private cloud initiative. This time, they had support from top management within the company.
Steve Eastham and Joel Crabb secured the funding needed to hire additional staff and acquire five new racks of equipment. The first cloud in this wave of projects was an OpenStack environment launching Hadoop clusters for analytics, and it is already in commercial operation.
Conclusion
The stories of MercadoLibre, Workday, and Best Buy contain a number of principles that can guide you along the path to successfully adopting OpenStack: being open to the needs of developers, the business, and other potential users; operating within the established processes of your company; collaborating with other organizations; and being prepared to act outside the rules when necessary. All of these are valuable soft skills that complement the OpenStack cloud.
There is no perfect path for implementing OpenStack in your company – the path to implementation depends on many factors related to both you and your company’s situation.
And while this fact may confuse OpenStack enthusiasts looking to implement their first project, it is still a positive perspective. This means that there are no limits to how far you can go with OpenStack; what you can achieve is only restricted by your creativity and resourcefulness.
Source: habr.com
