Back-of-the-envelope estimation

Back of the envelope (BOE) estimation is a rough calculation method used in system design to quickly estimate the size, capacity, or performance of a system. It is based on using rough and approximate numbers and simple calculations to get a rough idea of the order of magnitude of the result. BOE estimation can be used to quickly identify whether a proposed solution is viable or to compare the feasibility of different options.

For example, in designing a distributed storage system, BOE estimation could be used to estimate the amount of storage capacity required by considering the total number of users, the average size of each user’s data, and the expected growth rate of the data over time. Similarly, in designing a distributed computing system, BOE estimation could be used to estimate the number of servers required by considering the number of requests per second, the average response time, and the number of concurrent users.

It is important to note that BOE estimation is not meant to be an accurate calculation and should be followed by more detailed analysis and testing. It is just a tool to get a rough idea of the feasibility of a solution and to identify potential problems early on in the design process.

It’s also important to note that BOE estimations are done in the early stages of the design process, and the results are usually used to decide whether to move forward or not. If the results are promising, more detailed analysis and modeling are done to make sure that the design is sound, and the solution can be scaled.

An example of using BOE estimation in system design is designing a distributed storage system for a social media platform.

The social media platform has 1 billion users and each user uploads an average of 200 photos per year. The photos are stored in a resolution of 2 megapixels, and it’s expected to grow at a rate of 10% per year.

To estimate the storage capacity needed for the photos, the following calculation could be done:

1 billion users * 200 photos per user per year = 200 billion photos 200 billion photos * 2 megapixels per photo = 400 petabytes 10% growth per year * 400 petabytes = 40 petabytes of additional storage capacity needed per year

This BOE estimation shows that the storage capacity needed for the photos is around 400 petabytes and it will grow by 40 petabytes per year. This rough calculation helps the designer to understand that it’s a huge amount of storage capacity and it can be used to decide whether to move forward with the proposed solution or to consider other options.