The improvement of resolution of digital elevation models (DEMs) and the increasing application of the Revised Universal Soil Loss Equation (RUSLE) over large areas have created problems for the efficiency of calculating the LS factor for large data sets. The pretreatment for flat areas, flow accumulation, and slope-length calculation have traditionally been the most time-consuming steps. However, obtaining these features are generally usually considered as separate steps, and calculations still tend to be time-consuming. We developed an integrated method to improve the efficiency of calculating the LS factor. The calculation model contains algorithms for calculating flow direction, flow accumulation, slope length, and the LS factor. We used the Deterministic 8 method to develop flow-direction octrees (FDOTs), flat matrices (FMs) and first-in-first-out queues (FIFOQs) tracing the flow path. These data structures were much more time-efficient for calculating the slope length inside the flats, the flow accumulation, and the slope length linearly by traversing the FDOTs from their leaves to their roots, which can reduce the search scope and data swapping. We evaluated the accuracy and effectiveness of this integrated algorithm by calculating the LS factor for three areas of the Loess Plateau in China and SRTM DEM of China. The results indicated that this tool could substantially improve the efficiency of LS-factor calculations over large areas without reducing accuracy.