Communications of the ACM
Computing For the Future of the Planet
Computing could go a long way toward ensuring a sustainable future for human society and the earth, argue professor Andy Hopper and Andrew Rice of the University of Cambridge Computer Laboratory. Research being conducted at the lab has four areas of concentration, including optimizing the digital infrastructure to achieve maximum energy efficiency, developing a global data collection network for sensing and optimizing humans' resource consumption and environmental impact, forecasting and responding to future events in natural systems, and finding digital alternatives to physical activities.
The complexity of data centers and server farms has grown to such a degree that a large percentage of the digital infrastructure is committed only to coping with faults, and Hopper and Rice argue that software redundancy methods can be used to facilitate uninterrupted service. They also see potential in building data centers and server farms in close proximity to wind turbines and other large-scale renewable energy sources. Hopper and Rice write that their lab intends "to create a detailed model of the planet that is kept up-to-date with information collected from millions of sensors around the globe," which would allow the optimization of energy consumption and other natural resources while also measuring the environmental impact of human activities. Meeting the third challenge of predicting and reacting based on a world model requires optimizing the use of the chosen machine architecture and making sure that intermediate values in a model are not computed to an unnecessarily high degree of accuracy.
Migrating from physical activities to digital alternatives carries substantial costs, and Hopper and Rice conclude that "intelligent choices about which activities we move to a digital world, compounded with our optimal digital infrastructure and other improvements in technology, should tilt the balance more strongly in the favor of computing."
From Electronics Weekly (UK)
View Full Article
Abstracts Copyright © 2009 Information Inc., Bethesda, Maryland, USA 
No entries found