Internet Archive: Petabox

Internet Archive: Petabox
The petabox by the Internet Archive is a machine designed to safely store and process one petabyte of information (a petabyte is a million gigabytes). The goals-- and current design points are: * Low power-- 6kWatts per rack, and 60kWatts for the whole system * High density-- 100 Terabytes per rack * Local computing to process the data-- 800 low-end PC's * Multi-OS possible, linux standard * Colocation friendly-- requires our own rack to get 100TB/rack, or 50TB in a standard rack * Shipping container friendly-- Able to be run in a 20' by 8' by 8' shipping container * Easy Maintainance-- one system administrator per petabyte * Software to automate mirroring with itself * Inexpensive design-- So far about $50k for project management, non-recoverable engineering, and samples * Inexpensive storage-- materials cost less than 50% more cost than the cost of the disks