I think sustainable and scale solutions are going to require complex, interactive systems. This means that both the intrinsic generation or sources are going to have to evolve, but so will the systems by which we control them, and ultimately the way in which we get people to interact with them.

I think as a result we have a unique opportunity now to work together to build a cleaner, more sustainable way forward to transform our economic competitiveness in the United States, and to spark a new wave of investment in startups and green collar jobs that will focus on this.

So, why do we need to solve this energy problem? I think, of course, there are two key reasons. One is just the geopolitics and national security issues that surround the current way that energy is distributed around the world, and the second is the environment. To mitigate the climate change problems, we clearly need to reduce our carbon footprint.

If we faced either of these problems in isolation, it would be an interesting challenge in its own right. But today we face them with a need to solve them simultaneously, and that more than doubles I think the complexity of addressing them.

Then on top of that we now face an unprecedented – in recent memory – recession, and yet we know that economic evolution will result in the Third World countries having more and more of an appetite for power. And yet as we think about trying to solve the climate change problems, we really find most people are focused on just trying to manage the sources and uses that we have today, without even contemplating the likely doubling of aggregate global demand that will come.

I think that one of the challenges, of course, is that the most accessible fuel that's around today is all carbon based in one form or another, and has the most significant impact on the overall environment.

Some nations have chosen to move toward energy dependence by using their domestic sources of oil or natural gas or even coal, and this may solve for the energy security problem, but it certainly doesn't solve the environmental crisis. And, of course, it does seem that relying strictly on carbon is unsustainable for the very long term in any case.

Craig Mundie

Craig Mundie, Chief Research and Strategy Officer for Microsoft, spoke on “The Future of Energy Conservation” and the launch of Microsoft Hohm at the Edison Electric Institute Annual Conference held in San Francisco, CA, June 24, 2009, excerpted by World-Gen.

At the same time, the most environmentally friendly sources aren't really ready for deployment at scale, either because we have the wrong kind of infrastructure or because they're simply not yet mature enough to be able to meet the demand.

So, it's clear that we need to be on a path to some type of zero carbon energy ecosystem, and I think there's no other way for people to solve the compound problem that geopolitical, national security, environment, and increasing demand problems that we face simultaneously. But I think we're simply not in a position today to completely replace our energy sources right now, to reduce our carbon footprint as much as we would like, and at the same time accommodate future growth and demand.

However, I do think we can take immediate steps now to conserve energy. No matter what your predispositions are about what the future sources of energy are, there's no question that if we can just intrinsically use less, that's just a good thing. And so I think an equal emphasis, perhaps a large emphasis in the short term has to be applied to this question of conservation, and how do we deal with that.

At this time I think if we focus on some conservation issues, while we work to develop through engineering activities and business activities zero carbon sources, we'll be in a better position than we would if we only focused on the new generation capability.

I think many of these zero carbon sources may emerge but are clearly on a 10- to 20-year cycle, and so we have to do something in the interim to improve the situation.

I think one reason for optimism is that forthcoming advances in computing are greatly accelerating our engineering and scientific discovery capabilities. Individual computers – the ones you'll find on your desk or in your cell phone – over the next few years will go through a quantum change in the architecture of the microprocessor.
There has not been a change of this magnitude in the computing environment since the microprocessor was invented about 30 or 40 years ago.


These changes will result in machines that are 50 to 100 times more powerful than the ones you use today that consume less than or equal power, and cost the same or less money than the ones we have today.

So, this kind of step function change represents an opportunity, both in terms of how we use that capability close to where people are, but when we aggregate them into these exoscale computing facilities, what people now call the cloud, we have the ability to gang them up literally by the thousands, potentially even the millions, and to create a computational capability the likes of which the planet has never known.

In the United States the country's preeminence in many fields actually has its roots in the ability to do computing in ways that other people can't. That was certainly true in the nuclear power and nuclear weapons era, but I think it's been true in many, many fields of endeavor. And I think that we have an opportunity in the United States to lead again in the use of these advanced computing capabilities to look for engineering solutions to these problems in ways that we really don't fully expect to find.

I think we clearly need to materially reduce the largest CO2 emitter, which is transportation, and this too will present us a huge challenge. In the United States today transportation accounts for I think roughly about a third of CO2 emissions, and about 29 percent of all energy consumption, and it's two-thirds of all the oil that we consume.
The majority of that consumption is by cars and other vehicles. So, I think what we have to do is plan under the assumption that the conversion to more electric transportation is going to be a 10 or 20 year problem, one where the natural replacement cycle of the vehicles will largely drive the time over which that conversion could take place. So, I think it's clear enough that the first crucial step we have to take across the board is conservation. We clearly can't conserve enough energy to solve the energy and environmental problems, nor can we achieve the reduction of consumption that we need by using just traditional sources of energy. But there are short-term opportunities to achieve significant savings, perhaps as much as 20 to 30 percent, using some of these technologies.

I think software can help consumers to better understand their energy usage, and help them find ways to use less energy with minimal disruption on their lifestyle.

For example, one of the focuses at Microsoft is to be able to help the PC industry design personal computers and software that controls them that will dramatically lower their aggregate energy consumption.


Today, there are some estimates that say if just U.S. personal computer users had machines that universally put themselves to sleep or into a very low power state when people weren't actually actively using them, that it would save about $22 billion a year in electricity costs, and eliminate about 3 percent of the total U.S. CO2 emissions, just from the personal computers that are distributed around the United States.

I think a bigger opportunity though is to use instrumentation and networking at the level that we now enjoy it to get a better handle on the load, and think about ways to control that load.

Clearly, finding the right technical and regulatory and incentive mechanisms to move people forward in this direction will require some significant work, but it's a place where I think the coming together of the technology industry and the electric utility industry really holds significant promise.

I think we should obviously step up our investment in other sources, but not all of these are created equal. We should undoubtedly increase research and investment in alternative and renewable energy sources such as wind and solar, but equally we need to be clear, at least in my mind, that I don't think these are ever likely to be a substitute for today's primary sources, particularly if world demand at least doubles over the next 20 years.


A better short term bet might be to tap into in the United States new things like the proven reserves in natural gas as a way to help bootstrap ourselves forward in a somewhat reduced carbon footprint, but ultimately we're going to have to find very large scale, zero carbon sources. Carbon sequestration may, in fact, be
a possibility, but I think like some of the others it may turn out to be a significant investment, and would certainly be suboptimal if we could find other ways to do this.
Any of these potential solutions are going to require a lot of research and development, significant capital expense, and a long term rollout plan that transitions us from a world that today is dominated by coal and oil to a world that would be based on these other low or zero carbon sources.

All of these possibilities still raise the same fundamental questions about the grid, and our command and control infrastructure.
And while, in fact, I think we have focused today on a command and control capability in the hands of the utilities and generators and distributors, at the end of the day a lot of the consumption occurs in the homes. The largest single consumer of electricity I believe is residential, followed by small commercial. And, of course, it's the big industrial users that have the most sophisticated arrangements for control and cooperation and pricing arrangements, but they don't actually represent the largest aggregate demand.

It really doesn't matter what you believe the long term sources of zero carbon energy may be; whether they're centralized though or fully distributed co-generation I personally believe as a software technologist will actually turn out to make a big difference. And it isn't clear that the country can afford to bet on both of them, given the magnitude of the investment and the different types of technical challenges associated with deploying them. If it turns out you bet on the distributed co-generation strategy, then I think we will look at a number of interesting challenges.
One, of course, because many of these sources like wind and solar, you can't spin them up on demand, they come and go with the wind and the sun, it puts a much bigger burden on the question of storage of energy, and I think storage is not improving, battery technology and other storage systems are not improving at the rate that semiconductor technology and other technologies are, and I think it would bring with it a demand for a distributed control system to effectively control the loads and to be able to aggregate this and ultimately dealing with the economics of it.

As someone who spends a lot of our time looking at the question of the Web, at the scale that it represents today, and knowing the challenges that we face in command and control of that with the cyber security issues, the economics of it, the need for standardizations, it is a daunting technological challenge in the computer hardware and software domain, and one that I think would be quite significant if we were really to try to say we're going to move to this very I'll say low-scale, distributed co-generation model.

However, if you focus on the other strategy, I'm going to cut carbon sources at scale and find an alternative large scale way of generating power, then, in fact, you wouldn't need the same degree of complexity and sophistication that would be required in a fully distributed system. So, to some extent when people talk about the Smart Grid, I always ask them, well, which one of these are you really focused on, and frankly I don't usually get a clear answer. People talk about a Smart Grid, but they have many different views of what that might comprise, and I think that as a country we're going to have to pick a few key scenarios and focus on them, and then we're going to have to bring forward the best people in the world in computing and communications and related software technologies in order to be able to get that to happen. I believe it will be a much more difficult problem than many people are estimating.

I think it's important to remember that at the end of the day, if there were easy solutions to solving this environmental and demand footprint requirement, we'd have done it by now. It just isn't an easy problem.
And when taken together, these challenges can be quite daunting.

It is clear to me that we need to be on a zero carbon source quest, and that the only long term path to solving the world's energy problem, with a planet that will ultimately end up with about 9 billion people on it, from 6.5 billion people, and a reasonable assumption that certainly over several generations all of those people will come to expect and even demand that they have access to energy just as people in the developed countries do.

I think in the short term what we can and must do is increase conservation and improve efficiency, which is an area where, in fact, we believe at Microsoft we can contribute.
Today, we announced that we're releasing a beta version of a new product called Microsoft Hohm. Microsoft Hohm is a free, online, Web scale service that is designed to help consumers identify what their energy consumption patterns look like, and identify and make suggestions for ways to save money and cut consumption.

We are partnering with four utilities:
Puget Sound, Xcel Energy, Seattle City Light and Sacramento Municipal Utility
District. Microsoft is also partnering with Itron Inc. and Landis+Gyr to make it easier for utilities and consumers to automatically access granular energy consumption data. Integration with meter data will make it even easier for utilities to take full advantage of Microsoft Hohm offerings.