when a file is created/compressed/segregated into multiple parts ! data transmission takes place , now when we delete/extract/recompose those files we need to make sure all the data's were present.
Now to uniquely identify all those files we use hashing techniques and the algorithm used here in this case is md5 (which is a hashing algorithm which produces checksum (unique hash)) ! with this we can make sure that our actual file is recovered without any data loss
This explanation is not an accurate one but this is a wrap around those concepts