Unlike other RAID systems, RAID-Z does not have a fixed «strike» (stripe) size, but rather uses the logical block size of the file system. Each record in a RAID-Z array is a full segment record, which provides reliable data storage. RAID-Z uses diagonal distribution of checksums, which guarantees data integrity and allows to restore information in case of failure of one or more discs. This technology allows fast recovery not only of entire discs, but also of individual data blocks. In case of checksum mismatch, the logical error is instantly detected and corrected. Through the use of caching and optimised data access, RAID-Z provides high performance when reading and writing data.