Design Diagrams

Posted Jan 1, 2024

By Sakharam Shinde 3 min read

Introduction

Workload architecture is complex and multidimensional.
Each dimension type focuses on a specific aspect of the system by providing a level of detail that’s specific to that dimension.
Having different types of diagrams allows for a comprehensive understanding of dimensions.
It helps encourage effective communication, problem-solving, and decision-making among stakeholders.

A high-level system diagram serves as a broad overview of a whole workload or of a subsection within a workload.
It includes the main components, their relationships to each other, and the rough order in which data flows through the system.
Arrows show the direction of interaction.
These diagrams are good for reaching a common understanding so that you can start deeper discussions or for stakeholder communication.

A block diagram breaks down a workload into its major functional blocks.
The blocks are usually technology agnostic.
They refer to the functionality that’s being performed instead of a specific component.
For example, a block diagram might reference a “messaging bus” instead of a specific message-bus technology.
This type of diagram can help explain a system’s structure, data flow, and processing flow without distracting the audience with fine details.

A component diagram works like a block diagram but replaces generic functionality blocks with specific technologies.
It presents a detailed view with the goal of communicating the system’s individual technology components and their relationships, such as client/server.
These diagrams are a sort of visual bill of materials for the scope of the diagram.

A deployment diagram focuses on the deployment of infrastructure, commercial off-the-shelf (COTS) software, and custom code across the workload.
It shows how the software and code are distributed across the hosting infrastructure.

A data-flow diagram (DFD) illustrates how data moves through a system, which is useful when you’re modeling data-centric systems.
In a diagram like this, it’s a good idea to note if data is moved in batches or in real time to remove ambiguity.

A sequence diagram depicts the communication exchanges between workload components over time.
It illustrates client/server relationships and their synchronous or asynchronous nature.
It also highlights dependencies in these exchanges and evaluates fault scenarios within them.

A user-flow diagram focuses on a scoped interaction between workloads, users, or actors and the workload.
It’s helpful for clarifying and visualizing functional requirements across various ways that a user and the user’s data interact with the system.

An entity-relationship diagram (ERD) is a modeling diagram that represents the structure of a database or another storage system.
It shows the relationship between entities (such as tables) through industry-standard attributes and association symbology.

A network diagram illustrates the solution from the perspective of the network that it runs on or interacts with.
These diagrams are useful in visualizing the workload’s network segmentation, network points of failure, and key network transitions such as internet egress and ingress points.
Network diagrams usually have a life past implementation. Network diagrams are often used in audits and incident response.

A state diagram is a specialized visualization.
It shows the state that a flow (or an individual component) is in.
It also shows how the flow transitions between states in response to conditions or events.

Although it’s not an architecture diagram specifically, a flowchart is another way to bring clarity to a design.
Flowcharts are often useful when they represent complex workflows or logic.
You can use them to help refine requirements and to help drive implementation choices.
Flowcharts are useful to include in workload incident response plans to help highlight key decision points and their associated actions or notification channels.

This post is licensed under CC BY 4.0 by the author.