for conflating big datasets i run about 400 gigs o...
# datascience
j
for conflating big datasets i run about 400 gigs of swapfile with lz4 kernel compression