Tape Drive Selective Rewrite Units

The patent divides the 16 data tracks into multiple re-write units (e.g., 4 re-write units with 4 tracks each). When an error is detected in only one track, only the specific re-write unit containing that track is re-written, rather than re-writing all 16 tracks. This segmentation allows selective re-writing of only the defective portions, improving re-write efficiency while maintaining data integrity through targeted error correction.

The patent extracts and identifies only the specific re-write units containing errors from the complete set of re-write units. Using read-while-write, the system detects errors in individual tracks and extracts only the defective re-write units for re-writing. This extraction approach prevents the time waste of re-writing error-free tracks while maintaining fast error detection through immediate read-while-write verification.

The patent implements partial accumulation of re-write units - accumulating units until a full set is available for optimized re-writing, but not storing more than necessary. The buffer memory holds only the re-write units that have been identified as containing errors and are waiting to be accumulated into a full set. This partial accumulation strategy optimizes re-write operations by ensuring full sets are available while minimizing buffer memory requirements by not storing excess units beyond what is needed for the next re-write operation.