[iquerno]

The only use cases I have seen for units larger than 'petabyte' are those representing the maximum allowed file sizes for ZFS, Btrfs and such. I also don't see a point in inventing more prefixes so that statisticians don't have to use scientific notation for large numbers. What use is that? How many people know how much a yottabyte is? If they need to Google the answer, that defeats the point.

1e12 terabytes seems easier to digest than 1 whatever-the-hell-,-I-don't-know-what-this-unit-is-meant-to-represent-byte. Not to mention, easier to read.

[saltcured]

Hmm, why would you mix 1e12 terabytes instead of saying 1e24 bytes? Why do we talk about 200k USD salaries instead of 2e5 USD? Or why isn't a US postage stamp marked as 6e-1 USD?

Also: in the past 25 years, "tera-scale" (TB and TFLOP) went from a prognostication about future high-performance computing into something you find in affordable consumer products. When campus computing centers are now deploying hundreds of petabytes, it seems myopic to think the PB threshold is anything but a signpost flying by the window...

[rjbwork]

>200k USD salaries instead of 2e5 USD?

You mean 2 lakh USD?

:D

[titzer]

$MEGACORP measures internal disk storage capacity in exabytes.

[Spivak]

Yes but translate this statement to the 80s and you might have said the same about giga.