Computational Sustainability — Algorithmic Challenges for Sustainability

For the past nearly twenty years, the term "sustainable development" has been coined to capture the ability for the needs of the present generation to be met without compromising the ability of future generations to meet their own needs — or more succinctly, sustainability. When one first thinks of computation and sustainability, the natural inclination is to think of Information Technology as an industry consumer, rather than as source of moderation. While it is without a doubt that the IT sector is placing ever increasing energy needs on society, computational models and algorithms also have the potential to play a significant role in preventing catastrophic overconsumption. There is a tremendous range of settings and domains in which these algorithmic questions arise, but there is one fundamental commonality to them: these problems typically are present at a scale for which sophisticated modeling and algorithmic tools are necessary, since the computational resources required by naive approaches are too great to be of any practical value.

This workshop aims to bring together algorithmic theoreticians, applied computational scientists, and domain experts to explore directions in which the theory of algorithms can have a substantial impact on the scope and scale at which these problems can be addressed. There will be survey talks by Susanne Albers (energy-efficient algorithms and green computing), Dan Bienstock (optimization models and algorithms in energy planning), Miro Dudik (species distribution modeling and MaxEnt), and Adam Wierman (energy modeling and algorithms for data centers). There will be a presentation from the NYC BikeShare program that will highlight implementation issues in the logistics of this program.