Microsoft Research Proposes Heating Your House with "Data Furnaces"


A smarter way to keep warm. Image credit Lewis Hine, Library of Congress

Server farms create a lot of heat, and use a lot of electricity to cool the computers. (In 2006, the IT industry used 61 billion kWh) Meanwhile, in much of North America, a lot more energy is used to heat our homes. (154 billion kWh) Scientists at Microsoft Research and the University of Virginia have a clever idea: Distribute those servers and put that waste heat to work in our homes.


click to enlarge

Jie Liu, Michel Goraczko, Sean James, Christian Belady of Microsoft, and Jiakang Lu, Kamin Whitehouse of the University of Virginia write:

In this paper, we argue that the problem of heat generation can be turned into an advantage: computers can be placed directly into buildings to provide low latency
cloud computing for its offices or residents, and the heat that is generated can be used to heat the building. This approach improves quality of service by moving storage and computation closer to the consumer, and simultaneously improves energy efficiency and reduces costs by reusing the electricity and electrical infrastructure that would normally be used for space heating alone.

At 110 to 130 degrees F, the exhaust from computers is too cool for most functions, but perfect for home heating. Gang together 40 to 400 CPUs, collect the heat and you have a Data Furnace.

They calculate that the IT industry could double its size without increasing its carbon footprint or its load on the power grid. They also suggest that it would create a faster and a more secure network. There are few variations on the theme, including low cost, seasonal DF systems that work primarily during the heating season. These would be built with less efficient, recycled servers, "- hundreds of thousands of which are decommissioned from existing datacenters every year to make space for servers with latest hardware technology".

They may not be up to scratch for the latest data centres, but they pump out the heat and could serve many non-critical functions.

For example, many delay-tolerant batch jobs can be performed opportunistically, such as non-real-time web crawling, content indexing, and the processing of large scientific data sets (e.g. astronomical data, genome sequencing operations, and SETI@Home). Type A DFs can also help developing communities, societal services, hobbyists, and other organizations looking for extremely low cost computing resources.

Another proposal is for Ecofriendly Urban DF's serving apartments and businesses. They would run during the heating season, but would still save money overall. (perhaps they could be hooked up to the hot tub) And while the study looked at houses, they suggest that "a similar approach could be used to heat water tanks, office buildings, apartment complexes, vegetable farms, and large campuses with central facilities." Put those old clunkers to work!

Read the Microsoft Research Paper via Gizmag
More on computer cooling
95% Data-Center Cooling Energy Reduction Thanks to Fluid-Submerged Servers
Google Floats A Data Center Patent: Offshore, Ocean-Cooled, Wave-Powered, And Modular

Follow me on Twitter! @lloydalter

Related Content on