A software-defined data center reshapes IT operations. It gives admins programmatic control, instead of clicking through a GUI, which presents a significant change in their day-to-day activities.
To effectively deploy and manage a software-defined data center (SDDC), IT staff needs to learn new skills and understand new products. For example, one aim of an SDDC architecture is to eliminate the manual IT tasks that are routine and repeatable, which requires skills to automate deployments and user-support tasks.
Familiarize yourself with these four skills, and their related tool sets, before you implement an SDDC architecture.
Data center automation
One obvious area to study, as mentioned above, is the automation tools that will be at the core of your SDDC architecture. It might be VMware's vRealize Operations, Cisco Unified Computing System Director or even an OpenStack distribution. Whichever tool you choose, it will be the core around which your SDDC architecture is built and will drive and control all other technical components. Understand how the tool is built and operated, as well as its limitations. For example, you will likely need to integrate this core automation tool with other software-defined components for more efficient management.
The software that makes up an SDDC architecture will not be entirely built by one vendor. As mentioned above, you might need to build integration between your core automation tool and the additional software-defined parts -- such as networking and storage -- in your data center. For example, your main automation tool might not have native integration with your backup system. To achieve software-defined backup, you might need to build the integration scripts that allow your core automation tool to manage the backup software.
There will also be parts of your SDDC architecture that are unique to your environment. Perhaps you'll need to integrate SAP HANA provisioning into an SDDC deployment workflow. Or maybe you'll need to update a configuration management system when you provision a new network in your SDDC. You will need to write code to build automation and integrate these various components.
Key SDDC terminology
Before planning your software-defined data center, make sure you know the lingo.
The good news is that you probably won't need to use a heavyweight programming language like C++ or Java. Instead, you will be able to use a more approachable scripting language like PowerShell or Python. To help fill in any gaps, learn at least one coding language and write your own scripts. In addition, the primary users of an SDDC are application developers, so you should spend a little time writing code to help you better understand them and their needs.
Interacting with APIs
Many scripting tasks will include the integration of APIs from multiple products. An API directs an SDDC architecture to take action, complete tasks or to report information. Calls to an API are the SDDC equivalent of clicking in the GUI. For example, your SDDC automation tool may not directly support your converged backup product, so you will need to use some scripting to enable the SDDC tool to create new backup policies in the converged backup.
Develop adjacent skills
Just as virtualization broke down IT silos, SDDC deployment will break those barriers down further and require teams to have even broader skills. SDDC software self-manages many IT products, so there's less of a need to understand those details. But software vendors are less likely to automate the integration with other products, so you must know how the different technologies interact.
For example, if your current skills are in virtualization and storage, then you could add networking or backup and recovery skills. It's even better to understand the business impact of your SDDC. Learn how the technology you configure affects the business' ability to develop new efficiencies.
Enterprise SDDC adoption still in the distance
Learn lessons about SDDC deployment from Shutterstock CIO
Explore the role of hyper-converged infrastructure in an SDDC