High availability is widely accepted as an explicit requirement for distributed storage systems. Tolerating correlated failures is a key issue in achieving high availability in today’s wide-area environments. This paper systematically revisits previously proposed techniques for addressing correlated failures. Using several real-world failure traces, we qualitatively answer four important questions regarding how to design systems to tolerate such failures. Based on our results, we identify a set of design principles that system builders can use to tolerate correlated failures. We show how these lessons can be effectively used by incorporating them into IRISSTORE, a distributed read-write storage layer that provides high availability. Our results using IRISSTORE on the PlanetLab over an 8-month period demonstrate its ability to withstand large correlated failures and meet preconfigured availability targets. 1

Abstract. The goal of this study is to propose an architecture for an intelligent sensor data processing middleware. In order to fulfill the ambient assisted living data processing requirements we design a flexible and scalable architecture based on multi-agent model. This architecture allows acquisition, interpretation and aggregation of sensor data-streams. Our system is able to process different sensor data-streams, to adapt to different levels of abstraction, to define different data-processing workflows. An extended operators over data stream language is used to define workflows. Different types of agents (simple sensor agent, logic sensor agent, virtual sensor agent etc) are defined. The designed system is a domain independent multi-agent system which can be instantiated for particular AmI problems. The middleware architecture will use a healthcare system for validation. Key words: ambient intelligence, ambient assisted living, context-aware sensitivity, healthcare systems, multi-agent model. 1. Introduction. The goal of an Ambient Intelligence (AmI) system is to response in an intelligent way to changes of context, providing intelligent services and adaptation to changing after the delivery [5]. These services include the control of the actuators integrated in the application, but also the acquisition of contextual information. AmI integrates a set of concepts from mobile computing, intelligent sensor, artificial intelligence, service oriented architecture and reflexive and adaptive systems.