PNNL

Abstract

Do you remember when a gigabyte disk drive was “a lot” of storage in that by-gone age of the 20th century? Still in our first decade of the 21st century, major supercomputer sites now speak of storage in terms of petabytes, 1015 bytes, or six orders of magnitude increase in capacity over a gigabyte! Unlike our archaic “big” disk drive where all the data was in one place, HPC storage is now distributed across many machines and even across the Internet. Collaborative research engages many scientists who need to find and use each others data, preferably in an automated fashion, which complicates an already muddled problem.