There was an interesting article over the weekend in the New York Times calling attention to a paper presented last June at the 2011 Usenix Workshop on Hot Topics in Cloud Computing that proposes using cloud server-generated heat to warm up single-family homes.
The paper, The Data Furnace: Heating Up with Cloud Computing (PDF), was co-authored by Jie Liu, Michel Goraczko, Sean James, and Christian Belady from Microsoft Research and Jiakang Lu and Kamin Whitehouse, from the Computer Science Department at the University of Virginia. The authors propose that instead of creating giant data centers of servers to support cloud computing, why not disperse the servers among homes/businesses and use the substantial heat generated to warm them up during the winter?
During the summer, the Times article says, "... the servers would still run, but the heat generated would be vented to the outside, as harmless as a clothes dryer’s. The researchers suggest that only if the local temperature reached 95 degrees or above would the machines need to be shut down to avoid overheating." The assumption is that a data furnace would not be cooled.
How the concept is envisioned operating is that a cloud service provider would sell data furnaces to homeowners for the cost of a regular oil furnace as well as sell the resultant heat to the homeowners at a price that would be equal to what he or she would normally be paying. The cost of the added electricity to run the data furnace would be paid for by the cloud service provider, as would the maintenance costs.
According to their total cost of ownership cost-benefit analysis (check the paper for their operating assumptions), the authors claim that a data furnace that is "designed correctly" could not only heat a 1,700 square foot house to 21 degrees Celsius (70 degrees Fahrenheit) but result in a total cost of ownership savings of about $300 per server per year to the cloud service provider.
The authors point out that depending on their configuration and availability, the data furnaces could be suitable for "... many delay-tolerant batch jobs [that] 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), " or "Internet television services and on-line video rental services [that] could use pre-fetching based on local programming schedules or video queues of people in the local vicinity. Similarly, location-based services such as map serving, traffic estimation, local navigation, and advertisements for local stores are typically requested by customers from the local region."
The authors say that a number of technical challenges would need to be overcome including ensuring that a data furnace's power and networking needs don't interfere with a home's normal operational requirements; that the physical and IT security of the data furnace could be maintained, and; that any hardware or software failures are able to be handled quickly and as remotely as possible. As the authors point out, "Even at the event of software failure, the system should continue to provide heat until receiving physical services."
There is a nice summary of the authors' presentation at the Usenix Workshop here (PDF) which also includes questions from the Workshop audience concerning the viability of the idea. Presentation slides also can be found here (PDF).
Any opinions on the idea? Would you be tempted replace your existing furnace for a data furnace? If not, what would it take?
Robert N. Charette is a Contributing Editor to IEEE Spectrum and an acknowledged international authority on information technology and systems risk management. A self-described “risk ecologist,” he is interested in the intersections of business, political, technological, and societal risks. Charette is an award-winning author of multiple books and numerous articles on the subjects of risk management, project and program management, innovation, and entrepreneurship. A Life Senior Member of the IEEE, Charette was a recipient of the IEEE Computer Society’s Golden Core Award in 2008.