Distributed Algorithm for Big Data Analytics in Healthcare

In the last few years, the exponential expansion of sodai media is making available an enormous and continuous stream of data containing invaluable information. A wide range of medicai and healthcare applications, including clinical trials, dis­ease surveillance, personalized medicines and population health management, can be also supported by analyzing these data. Obviously, new approaches and innovative methodologies are needed to investigate data characterized by high velocity, volume and variability. Dedicated services, comprising data sources and middleware, can be considered a dynamic content and then handled by exploiting a Content Delivery Networks (CDNs), an effective solution to manage contents in distributed systems. However, CDNs show limits in dynamic and large systems in which a great amount of data is managed. This paper introduces a distributed and self-organizing algorithm to build a management system for big data analytics in highly dynamic environments like healthcare domain. Thanks to autonomous and locai operations performed by hosts of a distributed system, a logically organized overlay network emerges and the resource discovery operations become faster and efficient. Preliminary experimental results show the effectiveness of our approach.