Is there anything the cloud can’t do?
A day doesn’t go by without us hearing about the benefits of the cloud. It can make us healthier, our kids smarter and everyone is happier. And the IT community? It seems to be the biggest benefactor of the cloud. Systems magically get better by migrating to the cloud. Software and services are cheaper, instantly and always available when deployed to the cloud. And software development, it’s easier to develop, deliver and it all seems to just work better and faster on the cloud. Clients are happy and instantly profitable when their businesses are on the cloud.
In reality, the cloud does introduce a revolutionary shift in the way technology is procured, used, managed and how organizations budget and pay for technology. The cloud also introduces a change in roles and responsibilities as well as capabilities. Developers are able to provision resources at a click of a button, testers are able to provision and decommission environments that emulate production in a matter of hours if not minutes, agile practitioners are able to use environments for prototyping of new functions and ideas and present them to users for immediate feedback. And with the cloud relieving operations or Ops of much of the day-to-day tasks of the data center, they’re free to move up the value chain by participating as an active member early in the development process.
While DevOps and agile practitioners are embracing the cloud, its capabilities and their new roles, not all of IT is on board nor embracing the migration to the cloud. It has been my experience and one shared and expressed by several of my colleagues that the “Ops” in DevOps does not understand the “why” in migrating to the cloud, the role they should play, or the value they bring to the equation. And as a result, they’re not fully participating in the manner in which they should: as a full-fledged team member.
If organizations are to extract the real potential of the cloud, they must have Ops (and here I use that to mean operations, maintenance and security) participate as a full-fledged team member on their journey to the cloud. Management needs to communicate not only the “how” and “what," but also just as importantly the “why."
In addition, organizations need to align themselves to facilitate the cloud migration with an organizational structure and supporting culture that embraces the new platform and the new roles and responsibilities it mandates.
The Benefits of the Cloud
The cloud introduces a revolutionary shift in the provisioning, use and the management of resources. And while each organization experiences a unique journey to the cloud and benefits in different ways, there are some core benefits the Cloud provides to each in some degree along this path.
- Variable vs. Capital Expense: Instead of investing upfront in data centers and resources before knowing how they will be used, organizations can now pay for these as they are actually used—and paying only for what they use.
- Global Scalability: Organizations can achieve global scalability in a matter of minutes rather than weeks or months. Global redundancy, lower latency and higher availability are all available at minimal costs.
- Capacity Planning: In the past, because of the lack of speed and flexibility in provisioning new resources, organizations were forced to overprovision to meet pikes and surges in demand. Organizations had to choose between sitting on expensive, idle resources or dealing with limited capacity. The cloud allows for on-demand scaling to meet these demand spikes and reducing capacity as needed.
- Speed and Agility: Resources are just a few clicks away rather than days or weeks. This increase in speed and agility allows developers the ability to test and experiment with unprecedented ease and cost effectiveness.
- Focus on Organizational Differentiators: The infrastructure-as-a-service provider relieves the organization and Ops of the time and money spent in running data centers and instead focusing on delivering real value to the end user.
- Economy of Scale: Through massive economies of scale, you can realize a much lower variable costs than you could achieve on your own. Costs are aggregated across hundreds of thousands of users allowing IaaS providers to pass these savings on to their users.
Ops Left Behind
Cloud adoption really shines when we begin talking about development and delivery, in particular DevOps and the agile methodology.
For development, it means rapid prototyping and experimentation, testing at scale with minimal costs and provisioning testing and production-like environments for realistic load and scalability testing. This can be achieved by a click of a mouse through IaaS user interfaces, through rich, extensive application programming interfaces or through powerful infrastructure servers provided via the IaaS or a multitude of vendors.
For agile practitioners, the cloud provides a dynamic development environment for rapid prototyping and experimentation, allowing new ideas and functionality for users to evaluate and provide feedback. This accelerated feedback loop results in an enhanced user experience and provides real value to the end users at a cadence that fits the organizational and end user needs.
In terms of Ops, there is now a shared responsibility model between the IaaS provider and client Ops. IaaS is responsible for the operations and security “of” the cloud and infrastructure, while Ops is responsible for what is “on” the cloud or the application security and operations.
The cloud frees Ops from the heavy lifting of procuring, provisioning and maintaining servers and running data centers, to focus on the delivery of real value to the end user. Ops is now free to be involved early in the development cycle providing guidance and expertise in terms of best practices concerning operation and maintenance of applications, providing value upfront where it has its greatest impact.
However, while development, testing and agile practitioners are embracing their new roles and platform, the same cannot be said for Ops.
First, despite the fact the benefits of the cloud may be obvious to some, it’s not as clear for others—such as Ops. Secondly, organizations and project team structures are not properly aligned with the objectives and the delivery process. This lack of alignment is especially confusing and misleading to Ops, the roles it should adopt and the value it brings bring to the delivery process.
So how do we bring Ops into the delivery process and help them understand and embrace the cloud and their new role?
First, management must clearly communicate the reason for moving to the cloud and how it benefits the organization. Management must communicate the big picture to Ops, the new role they play and how they will bring value to the organization. They have to first understand the mission and their role in that mission and the value they bring to its success before they’ll be able to embrace it. In the book, "Extreme Ownership: How U.S. Navy SEALs Lead and Win," co-author Leif Babin speaks about the importance of clearly communicating to teams on their role in any mission.
“This is not intuitive and never as obvious to the rank-and-file employees as leaders might assume. Leaders must routinely communicate with their team members to help them understand their role in the overall mission. Frontline leaders and troops can then connect the dots between what they do every day—the day-to-day operations—and how that impacts the company’s strategic goals.”
The mission or the “why” is never as obvious to the rank-and-file as it to management. If Ops isn’t fully embracing their new role and understanding their value add to the delivery process, then it is management’s responsibility to clearly communicate the message. Only then will Ops be able to fully understand where they fit in and the value add expected of them.
Second, management must align the organizational structure to the delivery process, platform and mission. We talk of a service-oriented, user-centric delivery process and teams focused on a single mission—delivery of focused functionality to the end user. However our organizations are not structured to support that design or delivery model.
In fact, they encourage the design and development of systems that match old, waterfall approaches and single, monolithic applications. Given Melvin Conway’s law "Any organization that designs a system (defined more broadly here than just information systems) will inevitably produce a design whose structure is a copy of the organization's communication structure," we need to align our organizations and culture to match our expectations.
Ops needs to be brought out of the glass house and into this DevOps culture as a first-class participant in this delivery model. Management and organizations need to embrace the concept of ownership of an application, service or feature with the adage “if we built it, we run it." And we need Ops engaged early and often if we are to achieve the mission. If we expect to succeed in terms of reliability, performance and sustainability, we need Ops able and ready to accept the lead in the production environment.
The cloud is here to stay, introducing a revolutionary shift in the use, procurement, budgeting and management of IT resources. And with it comes changes in roles and responsibilities for everyone. Ops will need to realign itself with this new paradigm and become a full participant in the DevOps culture, sharing its expertise and understanding at the beginning of the process and ultimately leading the deployment and sustainment of applications in the cloud.
To do so, organizations need to realign themselves in order to facilitate and encourage this delivery model. Most importantly, management must accept extreme ownership of this journey to the cloud, communicate the “why” and help Ops understand the mission and the important contribution they can and must make to its success.