Multi-dimensional optical data writing techniques for cloud-scale archival storage
- Patrick Anderson ,
- Erika B. Aranas ,
- Richard Black ,
- Stefano Bucciarelli ,
- Marco Caballero ,
- Pashmina Cameron ,
- Burcu Canakci ,
- Andromachi Chatzieleftheriou ,
- James Clegg ,
- Daniel Cletheroe ,
- Bridgette Cooper ,
- Tim Deegan ,
- Ariel Gomez Diaz ,
- Austin Donnelly ,
- Rokas Drevinskas ,
- Christos Gkantsidis ,
- Istvan Haller ,
- Philip Heard ,
- Teodora Ilieva ,
- Russell Joyce ,
- Sergey Legtchenko ,
- Bruno Magalhaes ,
- Aaron Ogus ,
- Ant Rowstron ,
- Masaaki Sakakura ,
- Nina Schreiner ,
- Adam Smith ,
- Ioan Stefanovici ,
- David Sweeney ,
- Phil Wainman ,
- Charles Whittaker ,
- Hugh Williams ,
- Thomas Winkler ,
- Stefan Winzeck
Published by SPIE
The demand for long-term data storage in the cloud grows continuously into the zettabytes. Operating at such scales requires a fundamental re-thinking of how we build large-scale storage systems to archive data in a sustainable and cost effective manner. In Project Silica, a storage technology for the cloud is being designed and developed from the media up by leveraging the recent progress in ultrafast laser nano-structuring of the transparent media. Together with the advances in reading, decoding and error correction processes, high-density and high-throughput multi-dimensional volumetric optical data writing is achieved, enabling successful end-to-end proof-of-concept demonstrations of the technology. With exceptional media longevity, this could transform archival cloud storage. Here we briefly discuss the development of the technology, key metrics for cost-efficient optical data storage at scale, and successful proof-of-concept demonstrations.