Thursday, December 5, 2013
Speaker: Mira Mezini
Venue: TU Wien
Objects in the sense of object-oriented languages are often characterized as active data that accept request-messages and autonomously answer them following an individual plan prescribed (mostly) by their type. These “active” objects served well the needs to model reality in software for the last two decades, but do not match as well the requirements of a new computing reality dominated by processing big data and event streams from heterogeneous provenience. This scenario demands for abstracting over data source and representation as well as over how computations on the latter are processed.
In this talk, I will present ongoing research at the Software Technology Group at Technische Universität Darmstadt aiming at empowering the object-oriented programming model to better support the needs of distributed, asynchronous computations that are reactive in the sense that they autonomously and incrementally maintain their results as entities since these results depend on change. The work is part of the PACE (Programming Abstractions for Applications in Cloud Environments) project funded by the ERC Advanced Grant with the same name.