Lacustrine ecosystems play a critical role in regional and global biochemical cycles. Bacteria are dominant contributors to biogeochemical cycles, participating in most of the decomposition of organic material in freshwater, and are critical for the maintenance of the biodiversity and stability of lacustrine ecosystems. In addition, bacterial biomass is an important component of natural lacustrine systems. Investigating the correlations between environmental variations and microbial communities is expected to reveal how microbes tolerate different types of environmental change and to increase our understanding about microbial ecology and evolution. Such information would also enhance our understanding about microbial adaptations to different environments and their function in these environments. Lake Poyang is the largest lake in China. As one of the few lakes that are still freely connected with the river and not heavily polluted, this lake is important for the maintenance of the unique biota of the Yangtze floodplain ecosystem. However, in contrast to the contamination and the water quality of Lake Poyang, little is known about the bacteria and their ecosystem functions in Lake Poyang. In this study, genomic DNA of the microbial community was extracted from sediment and water collected in May 2011 from Songmenshan Region, Lake Poyang, China. The benthic and planktonic bacterial community structures were examined by 454 pyrosequencing of the v4-v6 16S rRNA gene region. We used OTU-based analyses to identify changes in bacterial diversity and community structure. Chao1 estimator, inverse Simpson index, Shannon index, coverage, and the rarefaction curve were used to describe the richness and diversity of separate samples collected from different environments. The libshuff test was used to describe whether benthic and planktonic bacterial communities have the same structure. In addition, correlation analysis between the abundance of bacterial phyla and their diversity in each environment was performed to reveal whether they were related to each other. Overall, the dataset comprised 5751 sequences that were affiliated to Bacteria;of these, 1454 and 269 OTUs were obtained from the sediment and water column, respectively, indicating that benthic bacterial communities have higher bacterial diversity and richness. The results also revealed different bacterial community structures between the sediment and water column. In benthic bacterial communities, 228 bacterial genera belonging to 20 phyla, dominated by Deltaproteobacteria, Betaproteobacteria, and Verrucomicrobia, were found. A further 116 genera were obtained for bacterioplankton, which belonged to 13 phyla, with Betaproteobacteria, Bacteroidetes, and Actinobacteria being dominant. Significant variations in the relative abundance of bacterial phyla and proteobacterial classes were observed between environments. At the genus level, significant differences were also found, the dominant genera in the benthic bacterial community were Subdivision3 genera incertae sedis, Geobacter and Anaeromyxobacter etc. In the bacterioplankton community, the dominant genera were Acidovorax, Polynucleobacter, Hydrogenophaga, Acinetobacter, and Arcicella etc. Of note, Comamonadaceae, which included Acidovorax and Hydrogenophaga etc, was consistently dominant in the bacterioplankton community. In addition, Anaeromyxobacter was only detected in sediment, while Polynucleobacter, Hydrogenophaga, Acinetobacter, and Arcicella were only detected in the bacterioplankton community. The results demonstrated that the relative abundances of bacterial communities at the phylum level are correlated with their diversity, and that the correlation between the relative abundance and number of unique OTUs was significant for benthic and planktonic bacterial communities.