Chapter 7. Adapting Your Organization to Hybrid Cloud
As you adopt a hybrid cloud strategy, you should anticipate the need to make changes to your IT organization. Where traditional, siloed infrastructure stacks rely heavily on IT specialists with deep knowledge of a particular discipline such as networking, storage, and virtualization, your hybrid cloud operations may depend much more on IT generalists.
The move to automation and self-service eliminates much of the time spent satisfying user requests and trouble tickets, allowing your team to shift focus from infrastructure and day-to-day tasks to the applications and services your company relies on.
This chapter explains why organizational changes are necessary, outlines opportunities to take advantage of resulting IT changes, and looks at organizational strategies for DevOps success.
Why Organizational Change Is Necessary
As important as technology has become to the modern enterprise, a successful digital transformation also requires significant changes to both company culture and management structure. For example, Dutch banking giant ING developed an agile approach to organization modeled on the success of companies such as Google, Netflix, and Spotify. In the process, ING discovered that significant changes to the company culture were also needed.
The recent report, Digital Transformation in Financial Services: The Need to Rewire Organizational DNA, discusses the need for firms to adopt a “digital DNA” across the company, a specific set of characteristics that enables them to succeed with digital transformation.
While transforming the culture and organizational structure of your entire company is beyond the scope of this report, I want to say a few things about the impacts and opportunities for the organization and culture of your IT and development teams. These teams can serve as a catalyst for the rest of the company.
The preceding chapters emphasize how thoughtful cloud adoption combined with datacenter transformation are essential to delivering the new applications and services necessary to maintain your company’s competitive footing and enhance customer experience. As you work to accelerate the delivery of new digital applications and services, agile development methods will replace the traditional waterfall approach. The balance of power between developers and IT operations has to evolve to deal with this new reality.
Transforming your datacenters combined with prudent adoption of services from CSPs and public cloud providers has significant impacts on your organization:
Frees up budget and personnel
Reduces reliance on IT specialists
Increases your ability to focus on applications and the application stack
Facilitates a transition to DevOps
In order to take full advantage of these changes you may need to reorganize or flatten your organizational structure.
Reprioritize IT Resources
Traditional IT infrastructure is simply not flexible or scalable enough to adapt to digital needs. IT teams spend far too much time and effort on mundane management tasks that keep the lights on but don’t move the business forward.
A high percentage of the IT budget is dedicated to day-to-day operations, which often consume up to 80% of your IT budget, leaving only 20% for innovation. The goal of datacenter transformation and hybrid cloud is to bring that split closer to 50/50, freeing up both budget and staff time that can be reallocated to new projects.
Reduce Reliance on IT Specialists
Fragmented infrastructure is difficult to automate and also creates a need for infrastructure specialists—individuals with specialized storage, networking, virtualization, and other skills who can troubleshoot the issues that inevitably arise. Hiring these individuals is both difficult and expensive. As your operation grows, it becomes impossible to continue hiring enough experts, making a dependence on superstars an unsustainable practice. In extreme cases, you may have just a few engineers who understand the impact of infrastructure changes or who can implement complex changes.
Datacenter transformation and a correctly designed hybrid cloud strategy with extensive automation reduces your reliance on these specialists, allowing infrastructure tasks to be handled by IT generalists and breaking down barriers that exist in your IT operations. You may find that you no longer have a need—and indeed that it’s counter-productive—to continue to be organized into separate server, virtualization, storage, and networking teams. With less time and attention on infrastructure, your teams focus instead on the application stack, application development, and new service delivery.
The resulting infrastructure is also much more comprehensible and amenable to your Dev team. Software-defined infrastructure enables automation and allows both Ops and Dev teams to program the infrastructure directly. This is sometimes referred to as “infrastructure as code.”
Organizational Changes for DevOps
With a rational hybrid cloud infrastructure in place, you’ll be well positioned to begin bringing Dev and Ops closer together, as illustrated in Figure 7-1, accelerating the transition to DevOps (as described in the previous chapter). However, you should anticipate a number of organizational hurdles as you move forward.
Figure 7-1. DevOps breaks down the organizational barriers between Dev and Ops so the two can coordinate efforts.
First, buy-in at the executive level is essential. As with any transition, DevOps will require some added expenditures and productivity may decline initially until you’ve worked the kinks out.
Second, you’ll need to identify an organizational structure that works for your business needs. That structure may extend beyond IT boundaries:
The key has been adhering to the “end-to-end principle” and working in multidisciplinary teams, or squads, that comprise a mix of marketing specialists, product and commercial specialists, user-experience designers, data analysts, and IT engineers—all focused on solving the client’s needs and united by a common definition of success.
Bart Schlatmann, Chief Operating Officer, ING
There’s no single organizational structure for DevOps success. The website DevOps Topologies investigates potential organizational structures, including nine organization types and seven “anti-types” to avoid.
Finally, no matter what organizational structure you settle on, you’ll have to make investments in training to help people entrenched in either Dev or Ops operate cross-functionally and adjust to their new responsibilities. As with most endeavors, DevOps success hinges on the success or failure of the people involved. Prolonged success cannot rely on continuous heroic efforts, no matter how dedicated the team.
Summary
The transition to a hybrid cloud model will in all likelihood require some significant adjustments to your IT organization.
Key takeaways:
Freeing up IT resources due to automation combined with reduced reliance on IT specialists, introduces opportunities to flatten your organizational structure or reorganize around new business priorities.
The organizational changes to enable successful DevOps may require the formation of cross-functional teams with members from across the company.