The article considers two data processing algorithms in distributed systems: P-BFT and Raft. Both algorithms are designed to ensure data consistency and avoid conflicts in distributed environments. However, they have different approaches to solving these problems. The P-BFT algorithm operates at the bottom layer of data access and uses the principles of distributed consensus to achieve high transaction processing speed. Raft, in turn, operates at the top layer of data access and provides guarantees for data integrity and system state consistency. The article discusses the advantages and disadvantages of both algorithms and the choice between them depends on the requirements and characteristics of a particular system. Recommendations are given on the use of hybrid algorithms for more efficient operation of the system.