How to plan the architectural attributes?

Last week I was talking to a former colleague, he is running a SaaS startup now, and our conversation went around SRE, production, and reliability. We talked a lot about architectural attributes that relate or impact the incidents, from detection to recovery, and on to prevention. I thought to put some of my thoughts around architecture into a blog post since that was the core of our conversation. The next post I will cover more details on architecture for fault tolerance.

The architectural attributes can be sliced and diced in different ways, I will probably write another slicing and decision making in another post.

At the end of the day, any attribute and quality of a software system are going to impact the users. But some attributes are more concerns of customers and users of the system while another group is more concerning those building and maintaining it (developers, support, QA, etc.).

There is always a push and pull between different stakeholders on the importance and investment that should go into these two categories. For example, usability, availability, and responsiveness might be the product owner’s focus while a technical stakeholder would also think, and advocate about investment in extensibility, supportability, and so on. Each stakeholder advocates for the attributes they better relate to.

Leave a Comment