right now now the bud only specifies to use either 1MB or 4MB chunks.

We split the file in equal parts and make a merkle tree out of that. we tried to keep it real simple