About five years ago I shared some pointers to materials, including a synthesis of the state of the art and the research agenda, describing work going on primarily within the exascale computing community on the resilience of very large scale systems; while much of this is focused on computation (very large numbers of processors), it is also highly relevant to storage systems essential for large scale data management and digital preservation. Recently, a new article has been published by a group that includes a number of the authors of the earlier reports, looking at the progress that has beenmade over the past five years. This will be of interest to CNI-announce readers interested in getting a sense of the progress that has (or has not) been made over the last half decade.
This is available at
http://superfri.org/superfri/article/view/14
(abstract and pointer to PDF of the article).
Clifford Lynch
Director, CNI
