Sparse file attribute was introduced to save disk space from storing large sections of meaningless data (sparse data sets) that by default contain large sequence of zeros. When sparse attribute is set to file, NTFS file system allocates disk space only for meaningful data, whereas non-meaningful data is not physically allocated.
When a sparse file is read, the data is composed from allocated meaningful and non-allocated meaningless data in order to present a whole file for a program. Sparse files are transparent for users or applications, i.e., they are processed automatically on the NTFS file system level.