In 2018, “hybrid IT” will be the most overused term in the IT industry. That’s not to say that hybrid IT isn’t a real thing. It very much is. Enterprise IT organizations have found themselves in a position where they are surrounded on all sides by both emerging and legacy systems, which creates the conditions for operating an environment that is anything but standardized. Hence the hybrid moniker.
Hybrid IT isn’t a destination
It’s important to recognize that hybrid IT isn’t something that IT organizations aspire to create. It isn’t a destination. There is no plan to achieve a hybrid IT architecture, complete with a party to celebrate its completion when the last container is in place.
Hybrid IT is the opposite—a challenge to the philosophy of standardized architecture that has been the backbone of IT for decades. The question once was “How do I make this new application fit into my standard architecture?” Now the question is “What new architecture do I need to fit the application?”
The expansion of architecture options
To better illustrate the evolution of hybrid IT, it’s important to evaluate the nature of cloud environments. Today, there are ever-increasing options for deploying workloads today in various cloud formations. Hybrid cloud is a combination of (usually multiple) public cloud providers and a private cloud environment, which often is little more than a farm of virtual servers. Over the last decade, resistance to public cloud has faded as providers have addressed security concerns, businesses have demanded access to popular SaaS apps and developers devour IaaS for their devops deployments because the tools are better in the public cloud.
While in theory workloads can move seamlessly between public and private clouds in a hybrid cloud environment (to scale up for cloud bursting, for example), the reality usually falls short. Still, Amazon, Microsoft, and others are making strides in making this easier. Greater portability will increase the demand for multicloud architectures.
Hybrid IT is a superset of hybrid cloud
But hybrid IT is even bigger and more complex than hybrid cloud. Hybrid IT spans these multicloud environments as well as all the legacy systems and services that aren’t sexy, but they became legacy because they just work. And replacing them is a cost and risk that most don’t want to undertake—many will exist in data centers until they are retired. Mainframes have proven impossible to dislodge in financial institutions, government and other sectors because sometimes, you just need the computing power or reliability of big iron.
Hybrid IT, then, is an acknowledgement that we’re caught in a mix of cloud and legacy architectures for the foreseeable future. Perhaps one day everything will run in the cloud, although history tells us that the pendulum is more likely to swing back towards in-sourcing at some point. Yet, even though we’ve stumbled into this condition, it does not mean that we have to stumble with the condition.
The dangers of hybrid IT washing
This is where the warning to beware of hybrid IT washing comes into focus. As the industry has determined that not everything can nor should be run in the cloud, vendors are taking note. As long as legacy and cloud services operate independently, there is a division that increases the management complexity for IT operations and security. Reducing that complexity creates a need for unified hybrid IT management.
Many will promise comprehensive coverage of the hybrid IT environment top to bottom, managing security and regulatory compliance, operational monitoring and release management across your choice of architectures. Many will fall short of that promise—when you pull back the curtain over the capabilities you will find that they do not go sufficiently deep.
This will have adverse impacts on security, operations and development organizations, including examples such as these:
- If IT operations can’t abstract the differences/dependencies among the various computing platforms, devops initiatives will slow to a crawl waiting for code deployment.
- Cloud-based services often need access to legacy systems and networks, with rigid change management processes. This dependency creates a flexibility mismatch that reduces the speed and agility benefits of cloud computing.
- Operational management that is divided into silos defined by computing platform, requiring multiple teams and tools, increases complexity, costs, and errors while slowing operations and frustrating users.
- Services delivered on legacy platforms are often unable to elastically respond to peaks or decline in demand, in the same way that cloud services can.
- IT operations models based on manual management of changes and configurations cannot scale to the pace demanded by the business, leading to security policy and compliance violations.
When looking to implement a cloud strategy mature organizations must consider how that strategy incorporates their legacy estate. They must also resist the temptation to adopt an entirely new hybrid IT management platform just because it claims to support hybrid IT.
Rather than rebuilding management tools and processes to achieve comprehensive coverage, hybrid IT management tools must be selected based on how well they support hybrid IT processes and governance, which in turn must be aligned with business priorities and risk appetite. Building processes that are flexible enough to support both the innovative and proven parts of the IT environment is really the greater challenge, but support from the management tools is critical to scaling those processes in an increasingly velocity-hungry digital business environment. A topic for a future article.