Reply #76: I had a lot of help... my team had about 7 developers

[lapfog_1]

and something like 2 QA plus another 3 or 4 SysAdmins, plus the 24-7 operations staff.

The one cardinal rule was "Thall shall NOT lose any data!!!", so we had a number of things to ensure that... we developed the RAID disk technology (I had the pleasure of funding Garth Gibson's group at UC Berkeley). And when data arrived at the disk, two immediate copies were written to tape (in two different buildings). At a later point... after the data had not been looked at for a while, the disk copy was released (a UNIX HSM, one of the first in the industry and the one that was used as a model for most future UNIX HSMs). Also, another copy of the data was sent to the East coast so that if the "big one" was to hit California, we would still have a copy. Which was a good thing... I was there for the Loma Prieta quake.

The data that I was most proud to archive was the MTPE data... all that stuff that you can compare to current data to see things like Global Warming and El Nino events.