In this study, the microbial community characteristics in continuous lab-scale anaerobic reactors were correlated to reactor functionality using the microbial resource management (MRM) approach. Two molecular techniques, denaturing gradient gel electrophoresis (DGGE) and terminal-restriction fragment length polymorphism (T-RFLP), were applied to analyze the bacterial and archaeal communities, and the results obtained have been compared. Clustering analyses showed a similar discrimination of samples with DGGE and T-RFLP data, with a clear separation between the meso- and thermophilic communities. Both techniques indicate that bacterial and mesophilic communities were richer and more even than archaeal and thermophilic communities, respectively. Remarkably, the community composition was highly dynamic for both Bacteria and Archaea, with a rate of change between 30% and 75% per 18 days, also in stable performing periods. A hypothesis to explain the latter in the context of the converging metabolism in anaerobic processes is proposed. Finally, a more even and diverse bacterial community was found to be statistically representative for a well-functioning reactor as evidenced by a low Ripley index and high biogas production.