New algorithms designed to simulate a worldwide network of connected data centers can predict how to best limit carbon emissions from cloud computing.
A team of computer scientists at Trinity College Dublin and IBM Research developed the algorithms, which also carry out the necessary computing and deliver the required data. The researchers used the algorithms, collectively called Stratus, to simulate a scenario inspired by the setup of Amazon's Elastic Compute Cloud (EC2) data center, which has locations in Virginia, California, and Ireland.
The experimental model placed data centers in these locations, and used queries from 34 sources in different parts of Europe, Canada, and the United States as tests. The team factored in carbon emissions, the cost of electricity, and the time needed for computation and data transfers to optimize the workings of the network. The Stratus algorithms, which reduced the EC2 cloud's emissions by 21 percent, routed more requests to the Irish data center, which had faster servers, and used less power and generated fewer emissions.
View Full Article
Abstracts Copyright © 2014 Information Inc., Bethesda, Maryland, USA
No entries found