FRSS 2018 - 2018 Special Session on Failure Resilience in Scalable Systems (FRSS 2018)

Scalable systems are now ubiquitous in numerous realms ranging from clouds and grids, to data and storage systems, to networks, to supercomputers with millions of cores. With the increased scaling however, there is a corresponding increase in failures and therefore the need to detect, correct, and recover from these faults that inevitably occur. These approaches can range from redundancy and replication to checkpoint-restart, performance attenuation, and autonomous self-organization. Each approach has its advantages and limitations, and must be adopted and tailored for given situations.
This special session will focus on the factors that influence failures in systems at scale, and approaches to address tolerance to failures as systems increase in scale...
Topics
We invite original articles, position papers, and short presentations on research that explores models, methods, measurement, and frameworks related to Failure Resilience in Scalable Systems. Participants are encouraged to submit papers on topics including, but not limited to:
•   Predicting failures in scalable systems
•   Models for relating failures to size in computing, networks, and storage systems
•   Approaches to tolerating failures i.e. built-in resilience in scalable systems
•   Detection mechanisms and their costs in scalable software and hardware systems
•   Approaches to recovery in large scale systems
•   Autonomous self-repairing techniques for scalable systems