The Fallacy of energy efficiency, green data centers and smart grids

[I am spurred to write this long winded post based on press release issued by Verizon today. Verizon claims that a 3% improvement in CO2 efficiency per million dollars in revenue. This sounds all very laudable, until you realize the Verizon’s year on revenue increase from 2007 was almost 5% which means that Verizon’s overall absolute value of CO2 emissions have increased. Efficiency in energy or Co2 emissions, like that cited by Verizon, is one of the enduring myths that surround the entire issue of climate change. While I agree there can be cost savings and energy efficiencies with smart use of ICTs, energy consumption is NOT the problem facing the planet. Greenhouse gas emissions are the number one problem and often energy or CO2 “efficiency” solutions as we have seen from Verizon may mask the real issue of how to reduce overall CO2 production . Far too often people confuse energy efficiency with helping reduce CO2, where in fact the problem is the energy mix. Ensuring that most, if not all, your energy comes from non-fossil fuel sources, will have much more direct impact on CO2 emissions regardless of how efficient or wasteful you are with that energy. If you are genuinely concerned about global climate change, one’s investments in technology and energy solutions should be first and foremost predicated on how they reduce CO2 emissions, and not efficiency .

The paradox of energy efficiency has been demonstrated over and over by many economists and is often referred to the Jevons paradox or Khazoom Brookes postulate where paradoxically energy efficiency increases energy consumption – and if that energy comes from dirty power sources it then increases CO2 emissions. More accurately in a paper by Horace Herring [http://technology.open.ac.uk/eeru/staff/horace/kbpotl.htm] the postulate states that "energy efficiency improvements that, on the broadest considerations, are economically justified at the microlevel, lead to higher levels of energy consumption at the macrolevel.". The paradox applies not only to energy but also to improved efficiency in a variety of markets from labour to aircraft travel as illustrated by Herring:

“Employees are told that they must raise their productivity if they are to improve their job prospects. On the local microlevel this seems absurd, as many shop stewards (and the Luddites) once argued. However on the macrolevel the increased economic output, resulting from higher labour productivity, has lead (in the long term) to a growth in the number of employees; and

The introduction of wide bodied passenger aircraft, to replace smaller aircraft, was forecast to reduce the number of flights. In fact the resulting lower cost per passenger led, in a competitive market, to a large increase in air travel that more than offset the increased size of the aircraft. The raised productivity per aircraft called for more aircraft, not fewer .”
[From Wikipedia http://en.wikipedia.org/wiki/Khazzoom-Brookes_postulate]
“Increased energy efficiency can increase energy consumption by three means. Firstly, increased energy efficiency makes the use of energy relatively cheaper, thus encouraging increased use. Secondly, increased energy efficiency leads to increased economic growth, which pulls up energy use in the whole economy. Thirdly, increased efficiency in any one bottleneck resource multiplies the use of all the companion technologies, products and services that were being restrained by it. One simple example is that suburban development limited by water use can be doubled if the houses adopt water efficiency measures that cut their water demand in half. That way a small efficiency can have large opposite multiplier effect. Similarly cars that use less fuel are likely to cause matching increases in the number of cars and trips and companion travel activities rather than a decrease in energy demand. It appears that these latent multipliers of opposite effects may be generally greater than the linear result of the original effect. As of late 2008 this appears to not have been factored into the general discussion of sustainability and global warming mitigation strategies.”
Green data center PUEs are another classic symptom of the same problem. While at the micro-level improving PUE is laudable the net effect is to reduce the cost of virtualization, clouds and relocating computers to data centers. We are then likely to see new applications and services take advantage of this much lower computational and storage cost, that would not be practical with todays stand alone PCs and servers, resulting in increased demand for more energy consuming data centers. If we do not design zero carbon data centers and networks from the get go, then ultimately we may end up producing more CO2 than we do today, despite a PUE approaching 1.0

The same analysis applies to Smart grids – which are being foisted on an unsuspecting public in the name of being “green”. Today’s Smart’s grid only do load displacement where the biggest beneficiary is the utility in not having to finance a new power plant. There is very little benefit to the consumer other than a modest reduction in the utility bill, which most consumers are likely to ignore in the name of convenience. Rather than load displacement smart grids utilities should be focused on offering customers tools to purchase directly energy from renewable sources on the spot market. Utilities are inherent conflict of interest with respect to smart grids are their responsibility is to make money by selling energy. An smart grid modeled on the Internet as proposed by Dr Randy Katz will probably a much better solution.

As much as possible, in every aspect in our lives aim for zero carbon as opposed to energy efficient solutions. A good example is this initiative in the UK to build zero carbon homes. Once you are zero carbon you can be as wasteful or efficient with energy as your heart desires. – BSA]


Building zero carbon homes
http://www.communities.gov.uk/documents/planningandbuilding/pdf/1101177.pdf

See also:
Building a zero carbon Internet
http://www.slideshare.net/bstarn/building-a-zero-carbon-internet


“Smart Grid”: R&D for an Intelligent 21st Century
Electrical Energy Distribution Infrastructure
Randy H. Katz
UC Berkeley

http://www.cra.org/ccc/docs/init/Energy_Grid.pdf

Technology Opportunities: The Smart Grid, Pushing Intelligence to the Edges, to
Integrate a Diversity of Energy Sources and Loads
There is much that can be done near-term to exploit information technology to improve the
reliability, visibility, and controllability of the existing grid and to support loads that reduce their
energy demand in response to price signals, through such technologies as synchronized phasors
and intelligent metering. Longer-term and more transforming, the Internet suggests alternative
organizing principles for a 21st Century Smart Grid. It succeeded by pushing intelligence to the
edges while hiding the diversity of underlying technologies through a well-defined interface.
Any device can be a source or sink of routable traffic and intelligent endpoints adapt their
behavior to what the infrastructure can deliver in accordance with localized utility functions.
Radical proposals to replace existing infrastructures, given their wide deployment, high capital
costs, and well-understood technologies, are unlikely to succeed. Here, too, the Internet offers a
model — of infrastructural co-existence and service displacement. The early network was
deployed on top of the telephone network. It provided a more resilient set of organizing
principles, became its own infrastructure, and eventually the roles reversed: services such Voice
over IP (VoIP) telephony are recent additions, having been added over time. The same approach
can yield a new architecture for local energy generation and distribution that leverages the
existing energy grid, but achieves new levels of efficiency and robustness, similar to how the
Internet has improved the phone network.
1 For the most current version of this essay, as well as related essays, visit http://www.cra.org/ccc/initiatives
Imagine a system built on packetized energy: store energy where it is generated, “route” it to
where it is needed. The existing infrastructure is generally unable to store energy for later use,
yielding a centralized system with crude mechanisms to adapt load (e.g., regional exchanges,
peaker plants, curtailment) and provisioned so that users lack the means to present an easier-tomanage
load to the infrastructure. New environmentally friendly energy storage technologies are
needed, with capacity/cost metrics suitable for deployment in homes, buildings, and throughout
the transmission and generation system, including renewable/intermittent energy sources.
Combining intelligent communication protocols with energy transmission in a common
architecture makes possible distributed control and demand response to pricing signals.
Such an infrastructure design would permit a shift from peak/worst case to the average case
“with sufficient headroom,” analogous to statistical multiplexing in packet networks. The key is
to use this headroom as an input for controlling generation, storage, and loads. Standardized
intelligent “interfaces,” at the level of homes or even individual appliances, allow independent
powered operation, distributed generation, and energy exchange. The architecture should allow
aggregation to plug into the regional grid, the neighborhood peer-to-peer grid, or the facility grid
to use localized storage and control to smooth load, adapt demand, and engage in exchange.

Verizon Achieves 3% Improvement in CO2 Efficiency
Verizon's rate of 64 metric tons of CO2 emissions per million dollars in revenue in 2008 represents a year-over-year improvement of 3 percent, according to figures cited by the company. Overall, Verizon's "carbon intensity" is approximately nine times below the U.S. average, as reported by the U.S. Energy Information Administration.

Verizon cited a number of ongoing initiatives that should see its energy efficiency improve even more in the years ahead. In 2008, Verizon:
• Became the first telecommunications company to require that the network equipment it purchases be 20 percent more energy efficient.
• Launched an initiative to limit engine idling, which cut fuel consumption by more than 1 million gallons (equivalent to taking some 1,600 cars off the road for a year.
• Electronically delivered approximately 100 million bills to customers.
• Collected more than 1.1 million cell phones, an increase of 6 percent from 2007, through HopeLine, a Verizon Wireless program that collects, refurbishes and sells used phones to benefit domestic violence prevention programs.
• Recycled 35.1 million pounds of telecommunications equipment. (equivalent to the weight of more than 11,500 Toyota Prius sedans.)
• Focused on purchasing goods containing recycled material. As a result, 60 percent of all goods purchased by the company contained recycled material.

http://www.verizon.com