The raw sequencing data is 100 GB, and that's what needs to be stored, as the algorithms deducing the sequence are not perfect and under continuous development.

The raw sequencing data is 100 GB, and that's what needs to be stored, as the algorithms deducing the sequence are not perfect and under continuous development. Moreover, storing a genome in one location on a consumer hard drive is asking for trouble. In practice the genomes are stored backed up in multiple locations and in RAID systems on enterprise hard drives, so your estimate is of by about a factor of 1,000.