This series on container orchestrators has covered:
- the general concept of a container orchestrator, their responsibilities, and typical architecture
- an overview of the orchestrators leading the market in 2019q4 (Swarm, Elastic Container Service, Kubernetes), what makes that product different, and who it is best for
Let’s close out the series by connecting back to strategy.
Highlighting the differences between orchestrators and who each orchestrator is likely best for was intentional. Each of these products is optimized to help people solve certain problems well, and that requires a tradeoff of some other potential advantage.
In this series I’ve suggested who each orchestrator is best for based on attributes like overarching technology provider strategy and assets, team size, experience, workload type, and scale.
But those suggestions will only get you maybe a third of the way.
Shall we play a game?
Deciding which of these container orchestrators is right for your organization now and in the future is an exercise in strategy. Only you know the specific customer needs, technological and organizational assets, desired business state, and other information required to build a good, or at least better, strategy.
If you’re thinking of making a significant change to your compute services such as adopting a (new) container orchestrator, I recommend you start by Identifying Undifferentiated, Heavy Lifting in your organization using a technique like Wardley Mapping. Determine for yourself where a container compute platform fits into your overall execution strategy. In particular, I suggest you answer the dual question of, “in what ways does the container compute platform provide value to customers and how visible is that value?”
With Wardley Mapping you can lay out how your system provides value to customers now and work out the steps necessary to evolve your system to a highly competitive target state, rather than just optimizing an existing process built for a previous set of conditions and constraints. Mapping can also help you avoid having your efforts distracted by shiny, ever-changing technology baubles that claim to be unicorns, but are really dragons.
Shall we play a game?
I’ve started prototyping a Wardley Mapping Strategy Game. I’m playing this game to develop strategy for products as well as help customers find better solutions to their own product and service delivery challenges. Clearly this prototype board is in the early days, but it’s already useful and fun.
If you have a problem you’d like to explore solutions with me, hit reply and we’ll set something up.
Stephen
#NoDrama
