SIParCS 2015- Koirala

Jenish Koirala, Claflin University

Pi in the Sky (Storing Big Data on Cloud)

(Slides) (Recorded Talk)

Weather data collection and analysis is vital for earth climate studies. Approached with the problem of collecting and organising weather data at low cost, we decided to use the credit card sized Raspberry Pi (R-Pi) computers to tackle the problem. R-Pi clusters are flexible to a) collect weather data via sensors, b) host the obtained weather data on a cloud, c) organize the weather data on a database and d) host a web server that can provide access to the weather data. This presentation focuses on the implementation of cloud storage on the R-Pi. Owncloud software installed on each node of a R-Pi bramble works to provide a cloud storage service. Each node would participate as Owncloud server (RPOS) in the cloud. Weather data, collected by portable sensors, are transmitted periodically to the RPOS. RPOS concurrently executes our scripts to validate, monitor and replicate the data on the cloud. In order to facilitate structured access to weather data, it was pushed onto a database. The database is synchronized across all RPOS to help web servers access data in realtime.

Mentors: Raghu Kumar and Rich Loft, CISL TDD