Design

This chapter explains the strategies used to meet the requirements identified in the analysis.

Ideally, the design should be the same, regardless of the technological choices made during the implementation phase.

You can re-order the sections as you prefer, but all the sections must be present in the end

Architecture

  • Which architectural style (e.g. layered, object-based, event-based, shared dataspace)? Why? Why not the others?
  • Provide details about the actual architecture (e.g. N-tier, hexagonal, etc.) you are going to adopt. Motivate your choice.
  • Provide a high-level overview of the architecture, possibly with a diagram
  • Describe the responsibilities of each architectural component

UML Components diagrams are welcome here

Infrastructure (mostly applies to distributed systems)

  • Are there infrastructural components that need to be introduced? Which and how many of each?
    • e.g. clients, servers, load balancers, caches, databases, message brokers, queues, workers, proxies, firewalls, CDNs, etc.
  • How do components distribute over the network? Where are they located?
    • e.g. do servers / brokers / databases / etc. sit on the same machine? on the same network? on the same datacenter? on the same continent?
  • How do components find each other?
    • How to name components?
    • e.g. DNS, service discovery, load balancing, etc.

UML deployment diagrams are welcome here

Modelling

Domain driven design (DDD) modelling

  • Which are the bounded contexts of your domain?
  • Which are domain concepts (entities, value objects, aggregates, etc.) for each context?
  • Are there repositories, services, or factories for each/any domain concept?
  • What are the relavant domain events in each context?

Context map diagrams are welcome here

Object-oriented modelling

  • What are the main data types (e.g. classes) of the system?
  • What are the main attributes and methods of each data type?
  • How do data types relate to each other?

UML class diagrams are welcome here

In case of a distributed system

  • How do the domain concepts map to the architectural or infrastuctural components?
    • i.e. which architectural/component is responsible for which domain concept?
    • are there data types which are required onto multiple components? (e.g. messages being exchanged between components)
  • What are the domain concepts or data types which represent the state of the distributed system?
    • e.g. state of a video game on central server, while inputs/representations on clients
    • e.g. where to store messages in an instant-messaging app? for how long?
  • Are there domain concepts or data types which represent messages being exchanged between components?
    • e.g. messages between clients and servers, messages between servers, messages between clients

Interaction

  • How do components communicate? When? What?

  • Which interaction patterns do they enact?

UML sequence diagrams are welcome here

Behaviour

  • How does each component behave individually (e.g., in response to events or messages)?
    • Some components may be stateful, others stateless
  • Which components are in charge of updating the state of the system? When? How?

UML state diagrams or activity diagrams are welcome here

  • Is there any data that needs to be stored?
    • What data? Where? Why?
  • How should persistent data be stored? Why?
    • e.g., relations, documents, key-value, graph, etc.
  • Which components perform queries on the database?
    • When? Which queries? Why?
    • Concurrent read? Concurrent write? Why?
  • Is there any data that needs to be shared between components?
    • Why? What data?