This blog is about using ICTs to develop climate change preparedness solutions built around Energy Internet and autonomous eVehicles
Energy Internet and eVehicles Overview
Governments around the world are wrestling with the challenge of how to prepare society for inevitable climate change. To date most people have been focused on how to reduce Green House Gas emissions, but now there is growing recognition that regardless of what we do to mitigate against climate change the planet is going to be significantly warmer in the coming years with all the attendant problems of more frequent droughts, flooding, sever storms, etc. As such we need to invest in solutions that provide a more robust and resilient infrastructure to withstand this environmental onslaught especially for our electrical and telecommunications systems and at the same time reduce our carbon footprint.
Using autonomous eVehicles for Renewable Energy Transportation and Distribution: http://goo.gl/bXO6x and http://goo.gl/UDz37
Free High Speed Internet to the Home or School Integrated with solar roof top: http://goo.gl/wGjVG
High level architecture of Internet Networks to survive Climate Change: https://goo.gl/24SiUP
Architecture and routing protocols for Energy Internet: http://goo.gl/niWy1g
How to use Green Bond Funds to underwrite costs of new network and energy infrastructure: https://goo.gl/74Bptd
Wednesday, April 20, 2011
Free Lunch: Exploiting Renewable Energy For Computing
In these days of severe financial constraints reducing energy consumption and CO2 emissions will free up millions of dollars for critical research and development and eResearch. As many of the readers of this blog know, this is not a new concept and something I have been blogging about for some time. The Greenstar network (http://www.greenstarnetwork.com/node/6) has actually implemented a working version of “Free Lunch” with nodes located across Canada and as far afield as California, Iceland, Ireland and Spain. The Greenstar network is also actually measuring the CO2 savings (and potential offset revenue) that could result from such an architecture. Some commercial venture are also looking to deploy similar distributed cloud facilities on farms using manure and agriculture waste to power methane generators. Not only this provide additional “Free lunch” for cloud computing it may also help accelerate deployment of fiber to the farm-BSA]
Free Lunch: Exploiting Renewable Energy For Computing
http://www.usenix.org/events/hotos11/tech/final_files/Akoush.pdf
Abstract
This paper argues for “Free Lunch”, a computation architecture
that exploits otherwise wasted renewable energy
by (i) colocating datacentres with these remote energy
sources, (ii) connecting them over a dedicated network,
and (iii) providing a software framework that supports the
seamless execution and migration of virtual machines in
the platform according to power availability. This work
motivates and outlines the architecture and demonstrates
its viability with a case study. Additionally, we discuss the
major technical challenges facing the successful deployment
of Free Lunch and the limiting factors inherent in its
design.
1
Introduction
There is currently strong support for mitigating both individual
and organisational impact on the environment by reducing
fossil fuel consumption and, by extension, decreasing
the associated carbon footprint. Our dependency on
information technology in our daily lives has led to computing
infrastructure becoming a significant consumer of
energy. A study commissioned in Japan in 2006 showed
that digital infrastructure used 4% of total electricity production
[1]. Similarly in the USA and UK it accounts for
3% [2] and 10% [3] of countrywide energy generation respectively.
As information technology is deployed more widely
(and in ever increasing domains), it is likely that the total
energy consumed by this sector will continue to increase
despite gains achieved by technological enhancements.
The total energy consumed by information technology
in Japan for example, grew by 20% over the past
five years [1] even though there has been marked improvement
in computing and communication efficiency over that
period.
In anticipation of this growth, our industry is beginning
to explore renewable energy as an alternative energy
source to fossil fuels in the hope of mitigating our ecological
footprint [4]. For example, British Telecom, which
consumed 0.7% of the total UK energy in 2007, has announced
plans to develop wind farms to generate a quarter
of its electricity requirement by 2016.
However, it is difficult to reliably and efficiently exploit
renewable energy due to its unpredictability, variability
and remote location. Recent studies have shown that incorporating
renewable energy into the electricity grid is an
expensive and inefficient process [5]. A significant amount
of infrastructure investment is necessary to achieve the required
levels of scale that result in a financially viable system
[6]. The electricity grid requires upgrading in order to
collect, store and distribute this type of energy. Additionally,
renewable energy is typically located in remote areas
leading to significant (up to 15%) attenuation losses when
it is transmitted over long distances.
We have embraced the challenge of reducing individual
and organisational energy consumption and are working
towards an optimal digital infrastructure, as part of our research
into computing for the Future of the Planet [7].a
In this paper we outline “Free Lunch”, an architecture
that enables us to exploit otherwise wasted renewable energy
without requiring the expensive investment necessary
to deploy these resources at scale. Our design alleviates
many of the disadvantages associated with incorporating
renewable energy resources into the electrical grid. We locate
datacentres close to renewable energy sites and link
them together with a dedicated communications network.
We migrate workloads seamlessly to locations that have
enough power for them to continue execution. Our proposal
is considered a free lunch because we make use of remote
renewable energy that would otherwise be unutilised.
Additionally, the energy consumed by computations in our
architecture may be offset against an equal amount that
would have been obtained from traditional (fossil fuel)
sources.
In the remainder of this paper we describe our architecture
in detail (Section 2), describe a case study that provides
a broad overview on the viability of this system (Section
3), highlight the pertinent technical challenges (Section
4), and finally outline the limitations of our design
(Section 5).
[…]
------
Green Internet Consultant. Practical solutions to reducing GHG emissions such as free broadband and electric highways. http://green-broadband.blogspot.com/
email: Bill.St.Arnaud@gmail.com
Bill@St-arnaud.org
twitter: BillStArnaud
blog: http://billstarnaud.blogspot.com/
skype: Pocketpro