text
stringlengths
0
599
What Are Architectural Decision Records (ADRs)
Architectural decision records (ADRs) document important architectural decisions made along with their context and consequences. They were first introduced by Michael Nygard in a 2011’s blog post.
An ADR usually consists of a short text file describing a specific architecture decision. You can write them in plain text, AsciiDoc/Markdown format, or use a wiki page template.
Benefits of Using ADRs
Let’s see some of the main benefits of using ADRs:
They capture essential information about the software that can be reviewed at any time.
They can act as an effective means to document the software architecture. The ADR highlights why a particular decision was made. It also describes trade-off analysis, for instance, choosing performance over scalability.
If the developers want to understand why a particular technology was chosen for a given project, they can find the explanation in the ADR. This reduces internal debates. Developers have the critical thinking, so it’s crucial to let them understand the whys. This leads to more trust within the teams.
They help new team members to get familiar with the architectural way of thinking. Newcomers will understand what aspects play the most significant roles in the infrastructure. They will get an idea of what has been done in the past.
If someone wants to propose a new technology, it may turn out that the topic has already been discussed and rejected. The reason is in the ADR. So, they can invest their time in searching for a new solution.
Maintain transparency inside and outside of the teams. Anyone interested in the topic can read about the decision process. This way, different groups can learn from each other.
ADR Structure
ADRs should be concise and simply written.
Title
The title contains a short phrase describing the architectural decisions.
Example:
Programming language and framework for our frontend.
Status
The status can be:
Proposed (under review)
Accepted (approved and ready for implementation)
Superseded (superseded by another decision)
Context
The context explains why we need to make a decision. It also describes the alternatives along with the pros and cons.
Example:
We need to choose a programming language for our frontend. It needs to be suitable for web and mobile applications. Our UI needs a lot of detailed customization.
Decision
The decision describes the justification for why the particular solution was accepted. It has more emphasis on the why rather than the how.
Example (This is just an illustration of a technical decision; it’s OK if you don’t agree with it):
We will use Flutter as a frontend framework.
We considered the following frameworks:
Flutter
React Native
Argument
Flutter — fast, provides many UI components out-of-the-box that speeds up the development without needing to implement the components ourselves. It has a built-in material design. It is well-documented even though it is a young framework.
React Native — we have to implement most of the components ourselves. The documentation is not very detailed. The overall performance is not as good as Flutter’s.
Consequences
The consequences section contains information about the overall impact of an architectural decision. Every decision has trade-offs. That’s why it’s crucial to include the analysis to provide a clear picture.
Example:
It has a learning curve: Developers must learn Dart programming language to use the Flutter framework.
When to Write ADRs
You might be asking yourself: should I always write ADRs? Not necessarily, but here are some points that can help you decide if it’s worth creating an ADR:
Google Cloud Storage
This documents defines the approach to select Google services based on specific organization requirements.
| Author | Status | Version | Date | |:--|:--|:--|:--| | Jane Doe | Accepted | 0.1 | 11-06-2023 |
Context
Google Cloud Storage (GCS) is a service for storing your objects in Google Cloud. An object is an immutable piece of data consisting of a file of any format. You store objects in containers called buckets. All buckets are associated with a project, and you can group your projects under an organization. GCS provides several storage classes that enable a consumer to choose the right balance of access frequency against cost. More frequently accessed storage has a higher storage cost but lower access charge, while less frequently accessed data has a lower storage charge and higher access charges.
Decision
Need for an immutable compliant storage on the cloud to store regulatory related files and encrypt using customer managed encryption key (CMEK).
The storage solution must be simple, scalable and can be accessed through APIs.
Usecases: - Media content storage and delivery. Geo-redundant storage with the highest level of availability. - Backups and archives. - durable storage for static websites. - Integrated repository for analytics and ML.
GCS Object storage solution has a unique position with application transformation. GCS offers API level access on data storage for different storage requirements. GCS is tightly integrated with GCE, GKE, database services and git.
It is recommend GCS Object storage as one of the storage solution options.
Consequences
Pros or Capabilities:
Key Functional Capabilities for Google Cloud Storage can be found here.
Cons or Service Limitations:
The resource locations Organization Policy Service constraint controls the ability to create regional resources. When regional resource are created, testing should be carried out for new policy on non-production projects and folders.