BigData and Disaster Management Applications on Clouds

Innovative Solutions for Deployment of BigData and Disaster Management Applications on Clouds

Funded by Department of Industry, Australia-India Strategic Research Fund Round 7
Partners:	–
Duration:	2014 – 2017

Effective response to crises and disaster events such as floods, fires, hurricanes, tsunamis, and man made disasters is dependent on the availability of historical data as well as on the effective real-time integration and utilisation of data streaming from multiple sources including social media feeds and text messaging on mobile devices (i.e. text messaging, location, etc.). Timely analysis of social media data can help rescue teams, medics, and relief workers in (i) sending early warning to people, (ii) saving lives, (iii) coordinating rescue and medical operations, and (iv) reducing the harm to infrastructure. For example, during the 2010 Queensland flooding in Australia, Queensland Police (QP) analysed messages posted on social media by people in affected regions to understand the situation on the ground and appropriately coordinate search and rescue operations. QP also used social media as a clearing house for disaster-related information about aid, centres, and other resources available to those affected. Twitter was used during the Mumbai terrorist attacks (2010) for reporting road closures, death toll, and unsafe locations as they were happening.

Timely acquisition and processing of data from different sources and extraction of accurate information plays an important role in coordinating disaster prevention and management. However, there is a pitfall; the growing ubiquity of social media and mobile devices means there are more sources of outbound traffic, which ultimately results in the creation of a tsunami of data, beginning shortly after the onset of disaster events. This data tsunami phenomenon is being described as a new grand challenge in computing: The ‘BigData’ problem, which is defined as the practice of collecting complex data sets so large that it becomes difficult to analyse and interpret manually or using onhand data management applications (e.g., Microsoft Excel).

Hence, it is clear that there is an immediate need to leverage efficient and dynamically scalable ICT infrastructure to analyse BigData streams from mobile devices and social media in a timely and scalable manner to establish accurate Situation Awareness (SA) during disaster events. Inadequate SA in disasters has been identified as one of the primary factors in human errors, with grave consequences such as deaths and loss of infrastructure.

This project is to achieve this by leveraging cloud computing systems to engineer and host next generation Big-Data applications. The project  develops efficient techniques and a software framework called Cloud4BigData, which will enable QoS-based autonomic provisioning of BigData applications (such as disaster management) over cloud resources. Next step is to develop an innovative approach for diffusion and analysis (detection, collection, storage, processing, extraction, and reporting) of data from multiple sources, including mobile devices and social media, whereby mobile devices can interact with each other via an ad hoc network and with cloud-based resources. Finally, this project validates Cloud4BigData by developing application demonstrators for disaster, crowd, and traffic management and deploy them on private and public clouds as SaaS applications.