The model of distributed storage of split data using multidimensional parity algorithms, resistant to partial loss of storage locations, is considered as an alternative way to ensure security, capable of replacing the generally accepted multiple redundancy, bearing the costs of the expansion of physical volumes. In this model, redundant data is generated to recover partial losses of split parts. In this case, the recovery is performed due to the parity files formed during the splitting procedure, in which the main action is the calculation of bitwise parity with summation modulo "two". It is possible to restore a damaged file without distortion with the help of operation between the source files. Comparison with a number of models using checksum codes for recovery shows that the estimated probability of storage failure losses in the case of multidimensional parity codes is significantly less than using other split data storage options. It is possible completely to restore damaged files, using the operation of adding undamaged files bits. Due to the calculated data, when even theoretically non-recoverable losses are achieved, there is always a chance to find combinations of chain recovery. The complexity of alternative methods, including the iterative decoding method, makes the system of distributed storage of split data using multidimensional parity algorithms a more convenient error correction technology.