Functional (aka immutable) data structures are used extensively in data management systems.
From distributed systems to data persistence, immutability makes complex programs significantly easier to reason about and implement.
However, immutability also makes many runtime optimizations like tree rebalancing, or adaptive organizations, unreasonably expensive. In this paper, we propose semi-functional data structures, an approach to data structure design that allows limited physical changes that preserve logical equivalence. As we will show, this approach retains many of the desirable properties of functional data structures, while also allowing runtime adaptation. To illustrate semi-functional data structures, we work through the design of a lazy-loading map that we call a Fluid COG Fluid COG is a lock-free data structure that incrementally organizes itself in the background by applying equivalence-preserving structural transformations. Our experimental analysis shows that the resulting map structure is flexible enough to adapt to a variety of performance goals, while remaining competitive with existing structures like the C++ standard template library map.