The elemental sulfur (S0) recovery was evaluated in the presence of nitrate in two development models of simultaneous desulfurization and denitrification (SDD) process. At the loading rates of 0.9kg S/(m3·day) for sulfide and 0.4kg N/(m3·day) for nitrate, S0 conversion rate was 91.1% in denitrifying sulfide removal (DSR) model which was higher than in integrated simultaneous desulfurization and denitrification (ISDD) model (25.6%). A comprehensive analysis of functional diversity, structure and metabolic potential of microbial communities was examined in two models by using functional gene array (GeoChip 2.0). GeoChip data indicated that diversity indices, community structure, and abundance of functional genes were distinct between two models. Diversity indices (Simpson's diversity index (1/D) and Shannon–Weaver index (H′)) of all detected genes showed that with elevated influent loading rate, the functional diversity decreased in ISDD model but increased in DSR model. In contrast to ISDD model, the overall abundance of dsr genes was lower in DSR model, while some functional genes targeting from nitrate-reducing sulfide-oxidizing bacteria (NR-SOB), such as Thiobacillus denitrificans, Sulfurimonas denitrificans, and Paracoccus pantotrophus were more abundant in DSR model which were highly associated with the change of S0 conversion rate obtained in two models. The results obtained in this study provide additional insights into the microbial metabolic mechanisms involved in ISDD and DSR models, which in turn will improve the overall performance of SDD process.