There’s nothing new about enterprises using multiple clouds. For over a decade, since the first business units selected a SaaS solution or swiped a credit card for AWS services, enterprises have relied on multiple clouds to run their workloads. Today’s changes are more cultural than technological. What’s new is the idea of enterprise senior IT leadership embracing a strategic plan to leverage multiple clouds. Without a holistic plan and comprehensive understanding of how each cloud provider and technology can be used to its advantage, the business as a whole is often inefficient, ineffective, and operates with only bits and pieces of the available information.
Research consistently shows more than 70% of organizations are using more than one cloud technology to serve internal and external users. SaaS, PaaS, IaaS and DRaaS solutions are leveraged for their individual business needs driven by their individual business unit projects. And while an IO datacenters blog shares industry statistics that only 10% of enterprises have a multi-cloud strategy today, by 2019 that number is expected to grow to 70%. Embracing the reality will allow for better strategic vision, better decision-making capabilities and better economies of scale across the board.
What other realities should the enterprise be prepared for as they embrace the new reality?
1. Empowered Business Units
Business units are feeling more empowered to deliver technology solutions directly to their customers as the underpinning technology becomes more accessible to the entire business. The result for the customer is often a better user experience and faster integration of feedback and new features. Applications are being developed to run on clouds with specific tools and features to enable this integration. While this empowerment often results in a better user experience, it can quickly cause chaos and sprawl for the enterprise.If each business unit is choosing its own cloud platform, the enterprise loses out on scale, efficiency and the business unit may not have internal expertise to manage security and compliance standards which places the enterprise at risk.
The reality of an enterprise who embraces a multi-cloud strategy is a balance between the tension of central control and business unit empowerment. The end user, and the business unit, are both used to having the control to self-deploy and provision what is needed, when. The central enterprise role is to maintain a menu of cloud options that business units can easily order and self-manage. Central IT maintains visibility and management of the larger footprints, while each unit operates within its own scope of knowledge, empowered to deliver to the client without enough rope to put the business at risk.
2. IT’s Evolving Role
The role of IT has been transitioning for many years, from one of order taker and provisioner of technology to one of a strategic consultant and advisor. This shift was driven in part by the increase of outside technology companies and consultants selling directly into business units, propagating the multiple cloud reality in the first place. What’s exciting today is the ability to embrace that evolving role of internal IT into its own consultative advising group for the rest of the business.
As we mentioned before, even with business units delivering their own applications and services to end users, the role of central IT is not decreasing; it’s becoming more critical. Central IT needs to manage the forest as a whole, not the individual trees. The model most often discussed in the industry is one in which central IT helps advise business units with suggested use cases, workload characteristics and risk profiles assigned to each cloud to help guide the business. For example, your organization may choose to pre-approve two or three clouds with unique service profiles. Business units may choose a hyperscale cloud provider, like AWS or Azure, for newer applications still in development and loosely coupled “mode 2” applications that can take advantage of native tools, the economies of scale and variable commitment levels. Mission-critical and tightly-coupled Mode 1 applications may be best served in a managed private cloud for close management and control on reliable, consistent infrastructure. Compliance, security and data residency requirements may push applications into a cloud option regardless of the use case.
Central IT should embrace the opportunity a multi-cloud strategy affords to serve as a consultant and key advisor to all of the business units, which the enterprise will benefit from in the long-run.
3. Increased Complexity Tools like container orchestrators and serverless technologies are evolving and integrating into production environments. These tools are designed to help simplify application deployment and delivery, however, by their nature they increase complexity of the enterprise IT environment. As new applications are developed using the latest and greatest tools and incumbent applications with high business value are re-architected to take advantage of micro-service architecture tenants; the overall IT environment suddenly has more moving pieces, parts and systems to consider. Complexity isn’t a bad thing, the house often looks messiest when it’s in the process of a spring cleaning, but the key to success is understanding that during this time you’re going to need to be more aware of the complexity that’s being layered in.
Central IT should know what new tools are being integrated, how they map to existing security policies, how they impact other processes and procedures and how they shift your application dependency model. Coordination, collaboration and information sharing will be key to managing the flow of constantly changing information and empower the teams to recognize gaps early and swarm to react to problems. A key to mastering this reality is to not assume that new technologies designed to simplify application deployment and delivery will result in simplification of your overall environment.
4. Rethink Recovery Considerations
With each business unit directly managing their individual applications and central IT managing the larger multi-cloud strategy and footprint, it can be easy for important responsibilities to get lost in the shuffle. A comprehensive disaster recovery and resiliency plan is one such critical responsibility that often gets missed, but can’t afford to be forgotten.Individual business units and application owners can and should consider resiliency for their services. This will help maintain resiliency for small, microfailures of individuals systems. However, if the business relies on the individual owners to protect each individual application, the level of risk is high.
Complex multi-cloud environments require complex networking considerations, constantly updated application dependency mappings and comprehensive testing of multiple disaster scenarios (including both security disasters, infrastructure disasters and natural disasters). Many businesses overlook the ongoing maintenance required to have an always reliable disaster recovery plan. That maintenance needs to be managed even in a single-cloud or homogeneous environment, but it becomes even more critical in a multi-cloud environment, as the level of expertise required to manage such problems increases exponentially with each unique system and service.
Your company’s disaster recovery plan is the last thing you want to worry about when a disaster actually hits.
As more functionality moves to the internet cloud every provider and user is developing their own definition. Industry experts and researchers are struggling to formulate a standard set of terms to describe all the different functions.
Mode 1 and Mode 2 are no different. The best way to navigate the rough waters of change is to keep your eye on the facts buried under the hype. Code still runs on CPUs, packets still need to be routed, and the laws of physics haven’t changed. What does change is what these activities look like.