Finding shareable solution patterns in cloud architecture

You just completed the data architecture for the system supporting the manufacturing organization as part of a massively complex multicloud migration. You designed a unique approach to normalizing all databases and getting down to a single source of truth for all important data attributes, such as customer, invoice, product, etc.

Feeling very proud of yourself, you hold a Zoom meeting to walk the other sub-teams through the approach, architecture, and use of tools. With arms folded and concerned looks on their faces, they tell you that they also have leveraged a unique approach to getting to a normalized data model for their part of the migration, as well as finding and exposing core single sources of truth. 

Although these situations can’t always be prevented, clearly in this fictitious case, if the two groups were communicating and collaborating, they could have solved this problem in a more repeatable way. Instead, they worked in silos, not bothering to keep the other teams informed of what they were doing, and thus never finding common architecture patterns that could have provided repeatable solutions for hybrid cloud and multicloud deployments and reduced cost and risk.

Many look at this as an easy problem to solve—just get better at communications and collaboration. However, the problems become more commonplace when enterprises have more than one complex cloud architecture project underway. Projects may be siloed by a matter of organization, and teams may not even know that other projects exist. What then?

Core to this issue is that although we’ve gotten much better at solving complex problems when doing cloud architecture, our ability to share those patterns has gotten worse. We can blame the pandemic and the sudden increase in remote workers, but this is more about a lack of knowledge management infrastructure and training, as well as not having a culture of sharing. 

Some look at combining and shifting organizational resources to become more collaborative, but without a change in culture led by better frontline leadership, the issue will only persist.

So, how do you change things? I’ve found a few things that work. First, an enterprise architect could help. We’ve had enterprise architects before, but many were supplanted because their focus was geared toward understanding abstract technology concepts rather than working with specific projects.  

Today’s enterprise architects are charged with driving collaboration, knowledge management, culture, training, and working with all projects (alone or with a team) to ensure that everyone benefits from the work being done within all teams, no matter where they sit within the organization. This also means using efficiency metrics to determine improvements in leveraging repeatable solution patterns to avoid redundant work.

Second, tossing tools at the problem may also help, such as collaboration tools and knowledge sharing infrastructure, but without a parallel change in culture, it won’t be enough. Don’t forget incentives, such as rewards for sharing—even spot bonuses for providing sharable solutions that are useful in other areas of the business, no matter if they are implemented or not. Spending only $100K a year may buy you $3 million to $5 million in value, easy.

You may have guessed that this is larger than just cloud computing architecture. It affects how we build, deploy, and operate everything in IT. The days of reinventing the wheel need to end soon, or else we’ll have no chance of solving the problems coming in 2022 and 2023.