And other lessons on cloud adoption and management.
Managing services, applications and workloads in a cloud environment is different to on premise. Not better or worse, just different. These differences are more marked in public cloud, but apply to private cloud and hybrid cloud as well. So getting people ready for the cloud journey is as important as preparing the strategy and plan and working on the technology.
A move to public cloud especially will most likely change the nature of some people’s jobs – which may have been performed the same way for many years. Some functions will stay the same, some will transform, and some may disappear. (Some may of course be passed onto a service provider.)
For example, some tasks that system admins have been doing for twenty years will need to change with public cloud. Take server outages: traditionally seen as a problem to be investigated or rectified in on premise or even highly virtualised environments. Many monitoring toolsets raise alerts at outages and trigger processes aimed at rectification. But in a public cloud environment, where machines may be switched off anytime when they are not needed to provide a service, this set up needs to be amended.
Even if a workload does cause an issue in this environment, it can be readily destroyed and a new one redeployed in its place. This action can be logged for review in the morning rather than cause a major alert. In this context, the traditional mindset of a server down always equating to a serious issue needs to be updated, along with the related processes.
So people will need to reskill and think differently to ensure successful cloud service delivery. This need may be reduced if much of the management is outsourced to a third party, but there will always be a learning curve of some kind required to ensure the business can make the most of its partner’s or partners’ services and support.
There is often understandable resistance to these changes. Change management is therefore a crucial aspect of any transition to cloud and a key consideration to build into cloud strategy and planning. As part of the transition stage of cloud adoption, we spend time with customers to explain what is to change operationally, from a people and process perspective.
The positive flipside of all this challenging change is the huge opportunity for individuals and organisations alike to be empowered and prosper in the cloud era.
With public cloud in particular comes much potential automation of traditionally manual processes. The same can be said of private cloud or even traditional environments, of course, but these kinds of systems by definition have limits that public cloud does not. Nevertheless, with any cloud environment, automation of processes is an important reason why it offers more benefit to an organisation than legacy infrastructure.
The server destroying and redeploying process described above can in fact be wholly automated, with no manual intervention necessary to maintain the service. This mentality of coding and automating is another mindset shift that people need to make to get the most out of cloud.
More automation means that a single engineer may be able to manage 300-400 virtual machines instead of many fewer. It also means that they can focus less on servers, as such, and more on what they deliver. That is, they can get more involved in higher value activities, such as capacity planning and service management and delivery. As automation progresses, these same engineers may become more strategic and powerful in terms of the scale and importance of what they oversee and control.
This and other essential lessons on cloud adoption and management, learned by Datacom over years working at the ‘cloud face,’ are contained in a new white paper available for download now. If you would like to talk to us about it, or cloud adoption and management in general, then please get in touch at firstname.lastname@example.org.