Rammed, or What RAM(3)S Taught Us

RAM(3)S (Real-time Analysis of Massive MultiMedia Streams) is a framework that acts as a middleware software layer between multimedia stream analysis techniques and Big Data streaming platforms, so as to facilitate the implementation of the former on top of the latter. Indeed, the use of Big Data platforms can give way to the efficient management and analysis of large data amounts, but they require the user to concentrate on issues related to distributed computing, since their services are often too raw. The use of RAM(3)S greatly simplifies deploying non-parallel techniques to platforms like Apache Storm or Apache Flink, a fact that is demonstrated by the four different use cases we describe here. We detail the lessons we learned from exploiting RAM(3)S to implement the detailed use cases.