Poster: Dynamic and NM1-Right Selection for the Parallel Iterative Improvement Stable Matching Algorithm

Sequential algorithms for the Stable Matching Problem are often too slow in the context of some large scale applications like switch scheduling. Parallel architectures can offer a notable reduction in complexity. We propose a stable matching algorithm using n 2 processors that converges in O(nlog(n)) average runtime. The algorithm is structurally based on the Parallel Iterative Improvement (PII) Algorithm [3], which successfully finds a stable matching in approximately 90% of cases. We suggest an alternative selection method for pairs in the PII algorithm, called Dynamic Selection, leading to an algorithm that fully converges over 200,000 trials and generally converges much faster. We also suggest the Right-Minimum selection method, leading to an algorithm which converges in approximately 99.9% of cases and can be combined with the PII-SC algorithm's cycle detection and Gale-Shapley Initiation to remove all cycles observed thus far.