By Chris Fielding
A lot goes into IT infrastructure transformation. We know that firsthand.
We've transitioned workloads to the private cloud, adopted software-as-a-service (SaaS) solutions, transitioned to DevOps, and more in our own business, as well as assisted other organizations on their journey. In the process, we've changed the way we work and the way we do business.
Each transformation was built on the one that came before it, and required buy-in from the entire organization, especially the IT team, which had to adopt new skills to better manage the changing infrastructure.
My hope is that by illustrating our experience, I'll be able to offer the best approach to infrastructure transformation to help you successfully manage teams, costs and bandwidth.
The simplest transformation we undertook resulted from a business objective to streamline our IT disaster recovery (DR).
Since IT DR is an overhead to the business, we are always looking for ways to simplify and spend fewer resources to achieve the recovery point objectives (RPOs) and recovery time objectives (RTO) our business needs.
The infrastructure supporting some key DR domain tools was approaching end of life, and we used that opportunity to start storage consolidation and virtualization into a private cloud.
The results spoke for themselves: Recovery times improved, and the change was so successful that we had an opportunity to move more systems into the private cloud infrastructure.
As we did, we saw a gradual change in the expertise we required within the IT team. We became less dependent on skills in troubleshooting physical hardware, and started to value logic-based skills, particularly scripting.
But the more radical change was yet to come.
Because we had been so successful so far, when business leaders felt that our foundational toolkit was no longer meeting their needs, they decided that a move to cloud-hosted solution was a risk worth taking.
We moved from a completely in-house-provided Lotus Notes platform to the Google collaboration platform. Our IT organization gave up infrastructure control of our email. As a result, the critical skills associated with managing that infrastructure were no longer necessary. Instead, we now needed a team that could administer and understand the Google collaboration platform as well as work with a third-party vendor to manage the service and the relationship.
Although we now had to adapt to changing roles and skillsets, this successful transformation also empowered employees and changed our business attitude toward risk. As our business users became more open to SaaS and cloud solutions, it opened the door to new possibilities. Our IT organization needed to change.
We began to focus on integration rather than application stacks. Full-scale roles across the whole IT organization changed – and changed quickly.
That speed of change created its own set of problems. We needed to change the way we worked to keep up.
We moved from our trusted and heavily documented waterfall methodology to agile. As teams became more confident with agile, we moved to DevOps. These changes to our IT organization required a big investment in skills training. Fortunately, we had employees who were excited about this opportunity; they grabbed it and ran with it.
We are now in a place where all our teams, whether they're supporting on-premise systems, SaaS or anything in between, use our DevOps methodology. But this comes with additional responsibilities.
Once you open the door to your end users using cloud-type tools and SaaS, you need to stay on top of governance and compliance.
Enterprise architecture is critical in this environment. All of your systems should have the same understanding of each microservice -- down to the individual data items within it. Data ownership becomes an even more hotly debated topic, and many representations of the same data have to be managed through.
Many SaaS providers will sell to the end users, so my IT team has had to take on a consultancy-type approach. The IT team must work even closer with the business owners to understand what they need so they can select the right tools.
Technology is a double-edged sword. By using cloud tools, users are exposed to fantastic "easy-to-use" free tools. It becomes everyone's responsibility to ensure tools are used appropriately and in compliance with company policies.
We've come across a lot of interesting findings as we've undergone infrastructure transformation. We were also reminded that this sort of organizational change requires hard work, patience and resilience.
IT transformation also leads to big changes in the skills of the IT organization. Switching to DevOps means that teams work in a more flexible way and gain marketable skills. The result is that you should have a happier team.
The more you bring your enterprise architects into business conversations in search of cost-effective solutions, the more you should expect to see your business understand the value of a good enterprise architecture, and therefore, continue to invest.
You should also be aware of potential pitfalls.
As you push user traffic away from your data center and into the cloud, you will need to review your network architecture to ensure you can cope with this additional internet traffic. And don't forget the ongoing need for cost management.
While you can't overlook the risk of undertaking infrastructure transformation, if it's done correctly and carefully, the benefits to your organization are worth it.