Mr David Jericho1, Mr Gavin Kennedy1, Mr Denis Lujanski1
1AARNet, Brisbane, Australia
Edwards Deming, an American statistician and engineer, once said: “Without data, you’re just another person with an opinion”, and researchers have been calling to the wider community to support their data needs with large scale data solutions that are both easy to use, and cost effective, encouraging wider reuse of the data.
As data generation increases exponentially, so do opportunities to reuse data sets in all fields of research, and consequently the importance of accessible archival data storage grows. Yet the expertise, let alone the financial capacity, to run the infrastructure for the required data management plans continues to remain outside the focus of a researcher’s discipline.
AARNet has been deploying a large scale secure frozen data archival storage service, with rapid retrieval, while focusing on a low annualised cost and higher standards compliance and protections of data.
A key desisn criteria of this service is not to use punitive charging models for accessing and retrieving the data, encouraging reuse and as-needed data access.
Our frozen data archival storage service is based on a model developed by CERN, storing over 300 petabytes, growing over 30% per year, and still archiving decades old data sets. The frozen data archival storage service is investigating alternate protocol options, permitting users to use their discipline specific or preferred tool according to their data management plans.
AARNet is piloting this service by running it as an in-house confirmation across their CloudStor bulk data stores before taking it to the wider community in 2020.
Bio to come