The shift from on-prem infrastructure to the cloud footprint capabilities become more critical. Because of the inherent shift we need enterprise architecture more than ever. To the contrary, what I have observed with large companies moving to the cloud is a perception that because of the fungible footprint, enterprise architecture is no longer needed. This is because some of the traditional technical architecture value propositions have transformed. The Open Group defines some of these value propositions (TOGAF – Value Prop) that are affected by the paradigm shift. For example, a key EA proposal “Reduced time to market and increased IT responsiveness” is directly challenged by the new way of developing and standing up infrastructure.
A more in-depth look would affirm that because the infrastructure can be stood up within minutes instead of months, any business plan can be executed quickly and efficiently and the investment recovered just as readily. Associated with this value proposition is an acute need for capacity planning. While this isn’t exactly trivialized with the addition of the cloud, it is marginalized due to new found capabilities in the cloud, namely autoscaling. The fungible shift in resources has found itself commonplace and almost synonymous capability with cloud computing.
Another value proposition described by the TOGAF is “Better access to information across applications and improved interoperability”. The proposition here is minimized because the data is located (usually) in once standard cloud provider. Even if it is in multiple providers, inherent APIs and other web based services make data integration easy. The IaaS footprint provides capabilities such as out of region replication, or even services like Amazon’s S3, a global fully redundant filesystem accessible to all services internal to Amazon and even outside of the service provider. Other integration capabilities are becoming less about motivating an organization to focus and articulate the integration and more about programming software. I do feel this is another key reason for the perceived extinction of technology architecture.
