There was a group that rewrote the Chromium DOM itself in a data oriented design (which learns from database design and sort of applies to cache utilization and locality) and got a 6X speedup in some places:
This paper is basically what we do, except we have a big container type, aptly named "Domain.cs". Inside, you will find a public List<T> of every type. We decided to emulate SQL throughout the vertical (i.e. each List is a table) in order to make mapping to SQL a trivial affair. None of our Domain types contains any complex type as a property. Everything can go 1:1 to SQL. We use LINQ (or SQL) to produce projections as appropriate.
There isn't actually just 1 big domain instance either. It's more like one per user session, and then a global instance.
The impact this had on reducing complexity and bugs is incredible. I haven't seen a null ref exception in a long time. Also, being able to dump your entire universe to disk by serializing a single object is really nice.
https://meetingcpp.com/mcpp/slides/2018/Data-oriented%20desi...