Abstract
Systems and methods are described for a polymorphic in-memory data store that can dynamically adapt its physical structure on a per-session basis. Some applications may experience performance bottlenecks from rigid, static in-memory data structures that may not be well-suited for certain dynamic workloads. The described technology can analyze real-time indicators, such as incoming data patterns and executed query types, to help determine a suitable memory layout. An adaptive engine can then transition the physical representation of the data between different formats, such as key-value, columnar, document, or vector-indexed structures. This approach may allow the in-memory data representation to align with the computational needs of a specific session's workload, which can improve performance and resource utilization for applications with unpredictable query patterns.
Creative Commons License
This work is licensed under a Creative Commons Attribution 4.0 License.
Recommended Citation
Sukumaran, Abirami, "Dynamic Adaptation of In-Memory Data Structures for Session-Specific Workloads", Technical Disclosure Commons, (August 26, 2026)
https://www.tdcommons.org/dpubs_series/11497