A Dynamic Programming Algorithm for Resource Allocation with Data Cache Optimization in Big Data Scenarios

This paper proposes a dynamic programming algorithm for resource allocation with data cache optimization in big data scenarios. The core contribution is an architectural shift that decouples allocation logic from global sorting by maintaining a fixed-size cache window, reducing time complexity from O(n log n) to O(n). The algorithm trades bounded forced fairness for global allocation stability through a space-for-time design, where "space" is realized as a finite set of cached values rather than scaling with input size. The work is presented as an engineering architecture disclosure; formal proof and numerical benchmarks are deliberately omitted, as the architectural skeleton itself constitutes the primary contribution.