Managing the Blast Radius as System Components Get Bigger – Storage Field Day Roundtable

As every component of IT infrastructure gets bigger, we need to consider the impact of component faults. The so-called blast radius due to a failure has always been a consideration, and was taken into account with RAID, replication, high-availability architecture, and more. Even a small server has over a dozen cores, a terabyte of memory, and many terabytes of storage. We must also consider the inflation of data, adjusting for the size of applications and data sets, when comparing the impact versus historic systems. The reason components have become so large has more to do with economics of production and sale than with customer demands, and the ratio of storage capacity to data throughput has fallen, making the blast radius a real concern.

The conversation is moderated by Stephen Foskett of Gestalt IT and features David Klee, Enrico Signoretti, Glenn Dekhayser, Andy Banta, Ray Lucchesi, Frederic Van Haren, Richard Kenyan, Gina Rosenthal, and Jim Czuprynski.