- description of components, supporting infrastructure, and technology stack meeting your requirements. Goal here to clarify overall system design, not to define implementation plans and milestones. It could be used also as a base for project requirements or as preparation step for addressing specific technical concerns (throughput, capacity, scalability, interoperability with subsystems,..) - could be addressed in the Executable Architecture step.

Executable Architecture (EA) - in some cases (https://stackoverflow.com/a/35564742/322679) people think of EA narrowly as a way of generating systems from design or even specific format and tools ( e.g. from UML diagrams using RUP-supporting tools). I Mean in broader term: “executable architecture is a partial implementation of the system, built to demonstrate that the architectural design will be able to support the key functionality and, more im- portantly, to exhibit the right properties in terms of performance, throughput, capacity, reli- ability, scalability, and other “-ilities”” So, delivery could be as described in wiki “the description of a system architecture (including software and/or otherwise) in a formal notation together with the tools (e.g. compilers/translators) that allow the automatic or semi-automatic generation of artifacts” -e.g. CloudFormation you can run in your AWS account to generate all artifacts (e.g. ELBs/EC2s/ACLs/RDS/…) and get environment to deploy application components OR it could be a VM with tools and a local environment where developers could test before integration to QA\Production, but also could be just a proof-of-concept environment demonstrating design meeting your concern (e.g. scalability, integration with specific subsystem,...)