BusinessGreen

Santa Clara, Calif. -- Who would have thought the Japanese underworld would get involved in the fight against climate change?

Micron Technology earlier this year bought a company called Numonyx, which specializes in a fast and super energy-efficient type of computer memory called phase change memory (PCM) that has been delayed for years. Early PCM adopters have included computing companies, who use it in internal systems to speed up their prototype development process, said Ed Doller, who runs the advanced memory group at Micron, during a meeting at the Flash Memory Summit taking place in Santa Clara this week.

Another major early application, he added, was in gaming devices, specifically, "Pachinko machines," he said. A compulsive staple in the entertainment world of Japan, Pachinko machines have had long connections to shady characters.

Over the next 18 months, expect to see a few more PCM applications. Some companies are looking at inserting the technology into phones or DVD players for rapid boot-up. (Side note: Micron formed a solar joint venture earlier this year, so expect to see the name pop up more frequently in green circles.)

Memory and data storage will likely be one of the major topics in green IT over the next few years. That is, after better air conditioning systems. Storage can account for a large percentage of the power consumed by data centers. Startups like Schooner Information Technology and Nimbus Data Systems have started to market all-in-one storage devices based around flash drives and their own software. Meanwhile, Lyric Semiconductor, Fusion I-O and Sandforce have come up with components that boost the performance of the actual flash drives and chips.

In notebooks, meanwhile, a major notebook manufacturer will introduce a notebook that will not accommodate a regular hard drive, according to Micron Technology vice president Dean Klein. It will only come with a solid-state flash drive. Right now, most of the major manufacturers sell notebooks in which consumers can order a flash drive instead of a hard drive, but an all-flash notebook would be novel.

Who will do it? Apple is a candidate. The company already offers solid-state drives on its notebooks: it could just kill off the version of the MacBook Air with a regular drive and claim to have accomplished something significant. The obligatory obsequiousness would be deafening. Then again, Lenovo caters to the corporate types that would get the most use out of a notebook with an extended battery life.

Conventional flash memory will likely be the technology that starts to erode the dominance of drives. Alternatives may not be needed in large numbers or quantities until the second half of the decade. And the leading one is PCM, which has a long and interesting history. (And if history is any guide, a strong chance exists that it will never replace flash altogether.) PCM stores data differently than flash or even hard drives. It is made from a material similar to the stuff DVDs from which are made. To write data to it, heat is applied to a memory cell. When the cell cools, the bit re-solidifies into one of two crystalline structures, depending on how fast the cooling takes place. The two different crystalline structures exhibit different levels of resistance to electrical current. Those differing levels of resistance are ultimately read as '1s' or '0s' by a computer.

Stan Ovshinsky is the original inventor of phase change. Ovshinksy is the celebrated yet controversial inventor who played a major role in amorphous silicon solar panels and nickel metal hydride batteries (he is also the founder of Energy Conversion Devices).

PCM has been heralded as the next big thing since the early 1970s. Gordon Moore himself predicted in Electronics Magazine that computer users might see it in that decade.

The memory, though, actually only started coming out recently. (Don't feel bad for Moore, though. The same issue of the magazine included an article titled, "The Big Gamble in Home Video Recorders.")

Another PCM side note: Brian Harrison, the new CEO of Solyndra, used to run Numonyx.

Home Depot is going after more than hall lamps with its latest LED light bulb announcement.

The DIY giant has started to sell a line of LED downlights -- those interior lights that look like small floodlights -- for homes. The bulb was designed and will be produced by Cree, but sold under Home Depot's EcoSmart brand. Home Depot will also sell Philips-made LED bulbs and already announced it was selling Lighting Science LED bulbs under the EcoSmart brand back in May. 

The Cree/EcoSmart bulbs cost $49.95, last for 35,000 hours and emit about the same amount of light as a 65-watt incandescent downlight. It will screw into standard ceiling fixtures. Cree says the bulb over its lifetime -- which will be 32 years if you use it about three hours a day -- will save consumers $300. The bulbs are available on the Home Depot site now and will be in stores in the fall.

"20 million downlights are sold a year. They are increasingly used in new construction and retrofits," said Ty Mitchell, vice president and general manager of LED lighting at Cree.

Declining prices, policy directives and the rising price of electricity has set off a revolution in lighting. Over the next few years, fortunes and opportunities will be won or lost (here is a roadmap to new lighting).

Osram Sylvania back in May promised an LED bulb by August, so expect something from that company soon, although it may not be in collaboration with Home Depot.

The bulb in part grows out of a line of LED bulbs Cree already sells into the commercial market. Cree combines white and red LEDs inside the bulb so the bulb emits a warm, bright light similar to the light that comes out of familiar, but highly inefficient, incandescent bulbs. You won't see red spots interspersed with white-bluish light because of the white/red mix, or even a sparkling rose. It will look like regular light.

The EcoSmart bulb costs less than Cree's earlier commercial versions of this type of bulb that are already on the market. Cree will also sell a version of the EcoSmart bulb to commercial customers.

Most bulb makers tinker with the phosphor to boost the warmth and take some of the harsh glare out of white light LEDs. The white/red approach, however, results in a more energy-efficient bulb that can put out more light while using less power, according to Cree. Cree also has intellectual property on combining whites and red in this way. (Cree's intellectual property portfolio is something nearly all LED makers have to contend with, either by trying to woek around it or paying Cree royalties under a licensing agreement.)

Philips, Osram and others earlier this year unfurled plans for LED bulbs to replace familiar "A" type bulbs. Earlier this month, Home Depot itself began to tout a 40-watt equivalent standard bulb developed by Lighting Science. General Electric has an LED bulb coming that is also based around LEDs from Cree.

Technically speaking, Cree's downlight will emit 575 lumens and consume 10.5 watts. An equivalent incandescent would consume 65 watts and emit 635 lumens. While the LED bulb emits fewer lumens, fewer lumens are wasted due to the optical characteristics of bulb.

"Brightness is not going to be a problem," he said. "It appears brighter to most folks."

An equivalent compact fluorescent downlight might only consume 15 watts, or close to the level of the LED downlight. But CFL downlights contain mercury, don't last as long, and, perhaps most important of all, don't dim, or at least don't dim in most circumstances. Dimmers and downlights are often used in tandem in homes.

A note on lumens. Typically 60-watt equivalent A bulbs emit 800 lumens. The 60-watt equivalent A bulbs announced by Osram and Philips earlier this year will emit, respectively, 810 and 806 lumens.

Downlights emit fewer lumens. The 65-watt equivalent bulb shown here will emit 575 lumens. The parabolic shape and interior reflective surfaces, however, make downlights more efficient so you get more illumination for less wattage.

Despite all the talk about carbon footprints and the rows of compact fluorescent light bulbs at every hardware store in the U.S., consumers have no idea how much energy they use and don't understand the best ways to reduce consumption, according to a new study.

The shift to educating, and empowering, consumers has been critical in the past among utilities and smart grid startups in the home area network space. But the study shows that most efforts to date have left Americans clueless and simply doing less of their regular behavior, without looking at simple changes that could reap bigger gains in efficiency.

"Of course we should be doing everything we can. But if we're going to do just one or two things, we should focus on the big energy-saving behaviors," said lead author Shahzeen Attari, a postdoctoral fellow at Columbia University's Earth Institute and the university's Center for Research on Environmental Decisions, in a statement. "People are still not aware of what the big savers are."

Nearly 20 percent of approximately 500 study participants listed turning off lights as the best way to save energy. Furthermore, most of the people had no idea how much energy a truck uses in comparison to a train or ship, or how much energy a room air conditioner uses versus central AC. Overall, participants were more willing to somewhat curtail their actions rather than to invest in real efficiency, even if the latter would save more energy and money over the long run.

The results of the study, published in this week's Proceedings of the National Academy of Sciences, were slightly more promising when examining less energy-intensive behaviors. Although people's understanding was generally poorer when the potential for energy or carbon dioxide savings were large, they were more accurate on a smaller scale. For example, most participants were able to guesstimate the savings of swapping out an incandescent with a CFL or adjusting the thermostat in summer.

Those in the study also overrated the savings of many activities, including driving slowly on the highway, recycling glass containers or unplugging chargers when not in use. Even people who described themselves as having a high degree of pro-environmental behavior did not always report engaging in a large number energy-efficient habits and actions.

Although the study had some limitations, including the moderate sample size and a lack of incentives for correct answers, the conclusions are stark. "Many people's concerns about energy are simply not strong enough, relative to their other concerns, to warrant learning about energy conservation," the study authors write.

So for utilities and regulators looking to cut energy consumption through efficiency, how do you get people to care? For one, stop telling people to turn off lights when they leave the room. The conversation must become more sophisticated.

The problem with many of the smaller energy savings actions that are constantly suggested is that they don't offer enough gain for the effort. "We're all very instant gratification animals," said Daniel Moneta from MMB Research, an engineering firm that makes a family of ZigBee smart energy hardware and software for commercial vendors.  "From an economic standpoint, we all should have replaced our old bulbs with CFLs. If you do the math, you'd save money."

Instead of just providing actionable information, campaigns need to include information about relative effectiveness of those actionable items. For example, many people thought that line-drying clothes would save more energy than changing their washer settings, according to the study, but the opposite is true. The study also pointed out that people will often make just one or two changes and think they are doing enough, so it might be a better strategy for utilities and government programs to communicate the best two or three actions to get the most bang for the buck.

Public education campaigns and web portals should not only recommend individual actions, but should also strive to paint a picture for people about which actions, both in the home and as consumers, can save the most money and energy. 

"If we have that number in front of us all of the time, and we look at it in comparison to our Facebook friends," said Moneta, "we can see that one device next to another device has a better [contextual] meaning. I think that will certainly help to motivate customers."

Green fervor, for both types of green -- cash and environment -- has spawned a crowd of smart grid and energy efficiency startups: some viable and successful, many undifferentiated, most doomed.

With more than $30 million in revenue, OPower easily falls into the viable category.  The startup, funded by New Enterprise Associates and MHS Capital, has created a new type of energy business with growing momentum.

OPower is an energy efficiency company focused on customer engagement and behavior modification, currently providing millions of homes with in-home energy data and efficiency advice via paper reports or online.  The platform is described as advanced customer engagement and the firm says that about 85 percent of its customers will cut power consumption by around 3.5 percent. The customized data lets people know how much energy they're using in comparison to their neighbors and then follows it up with a recommended course of action.  

"One fundamental observation about OPower," said Ogi Kavazovic, the firm's Senior Director of Marketing and Strategy, is that the firm is "an energy-efficiency company as opposed to a smart grid company."  He added, "The smart grid is wonderful, but it's important to remember that the ultimate goal is energy efficiency. Once the smart grid update is done, that will become the new normal, but energy efficiency will continue to be the goal."

At a recent VLAB event at Stanford University, OPower's CEO, Dan Yates, said that the company's revenue was on track for more than $30 million this year.  He summed up OPower's business in this manner: "We provide information services to consumers through utility partners." He also said, "Energy efficiency is super boring.  No one cares about it," and continued: "You have to push the info, no one is going to pull it. You have to deliver outcomes, not just deliver raw data."  And in his words, that means that "web portals and in-home displays are not delivering value."

"We have cracked the nut in motivating consumers," said Yates. The rapidly growing startup works with utilities to "suck in their meter and billing data" and match that info with demographics, GIS data and weather data.  Without having any devices in the home, OPower can disaggregate the data, parsing, say, heating from cooling, and provide analysis and actionable recommendations to the consumer to lower their energy usage.

And here's a great statistic from the company: the firm is providing the equivalent of one-third of the U.S. solar industry's output in energy savings -- simply by sending out an actionable set of data once a month to utility customers.

The number of utilities working with OPower is now 37 and Kavazovic sees that accelerating over the next 18 months.  According to Kavazovic, "We're closing deals with two utilities a month."  Greentech Media's Katherine Tweed reported on OPower's contract with AEP Ohio here.

"There's so much dust in the air," according to Kavazovic, as the grid undergoes "the biggest infrastructure upgrade. But there's an even bigger story behind that dust." 

According to Kavazovic, the energy efficiency mandates in this country are becoming more aggressive.  There's more and more talk about devising a set of national standards.  "It remains a $10 billion industry -- and it's not a one-time thing, but rather, an ongoing project."

Starting this year, OPower is guaranteeing the level of energy savings it can provide to customers. According to Kavazovic, "We'll deliver 300 gigawatt-hours at less than 5 cents per kilowatt-hour." That kind of number is applicable to a large utility.

Kavazovic sees OPower as the de facto front end of the smart grid, able to provide millions in savings for end customers and a clear value proposition for the consumer, something that's been a bit hard to define, despite all the smart-grid hype.    

Ogi sums it up this way: "In a nutshell, what we've done is to figure out a way to engage a large proportion of a utility's customer base on a very boring subject, energy data."

And OPower has done it without any hardware in the home -- and in a way that lots of web portal and dashboard companies might consider a competitive challenge.

The firm has 105 employees with 40 or 50 job openings.

Examples of Opower's analysis and energy efficiency improvements:

Lyric Semiconductor has developed a type of chip that won't give you a simple yes-or-no answer -- and that's the beauty of it.

And if the idea one day takes flight, the MIT spinout plans to accomplish things like drastically improving online recommendation engines, mapping an individual's genome cheaply in a matter of minutes, or more accurately detecting fraud and spam on the internet. In the process, power consumption would be slashed back at data centers.

"Yahoo uses a lot of servers for spam," said CEO Ben Vigoda. "We combine physics and computing in a new way. It is not digital. We use electrons and transistors in a different way."

The company's GP5 architecture is essentially based around the principles of probability, according to Vigoda. Digital chips, like the Intel or AMD processor in your laptop, operate by breaking down everything into series of 1s and 0s. Transistors either let electrons pass or they don't. A photo from your digital camera or a song from your MP3 player are really just collections of on/off signals that form a coherent whole because you don't detect the individual signals.

Analog devices, by contrast, mimic natural phenomena: photoresistant chemicals replicate reality on film and speakers vibrate.

Chips based around the GP5 architecture combine the probability of one event (a 50-percent chance of A visiting the Empire State Building) with another (a 40-percent chance of B appearing there) to determine the odds of them meeting (0.56 x 0.4 = 20 percent). The answer isn't definitive in a theoretical sense, but it's right.

What does that mean in practice? To solve a sudoku problem or fold a protein, a digital computer will look at a vast number of all possible combinations. A probability processor would look at the basic rules (e.g., you can't have the same number twice in the same line or these two amino acids can't bond) and build from there. In the end, that leads to less power consumption.

Others are mining the field of green IT through better chips, as well. Yesterday, Smooth-Stone came out of stealth with a processor that directly challenges Intel in the market for low-power server chips. Sanforce, meanwhile, has a controller that makes it more practical to adopt flash -- which can be problematic over time because it stores data in a "violent" manner -- for storage.

Although probability sounds like a somewhat loosey-goosey premise from which to derive answers, computing has already moved in that direction. Google's search engines are based around probability, also known as probabilistic computing. The field of artificial intelligence has largely shifted from rule-based computational systems for robots to probability.

The foundation of the field can be traced back to Thomas Bayes, an 18th-century clergyman who tried to prove the existence of God through equations. (Disclosure: Vigoda told me he became intrigued by probability after reading this 2003 article that, coincidentally, I wrote. What are the odds of that?)

Lyric's first chip will be an error correction controller for flash memory chips used to eliminate artifacts from data. Vigoda will publicly unveil Lyric's technology for the first time at the Flash Memory Summit taking place in Santa Clara this week.

The company's controller is 30 times smaller than conventional error correction components, consumes 12 times less power and provides more bandwidth. In some applications, the a GP5 chip could provide up to 1000 times better performance than a conventional processors.

Ultimately, Lyric would like to make programmable chips that could serve as co-processors inside servers or storage systems, according to Mira Wilczek. The company, in fact, is working on projects for DARPA, which is also one of the company's investors.

The history of the chip industry is filled with noble ventures for improving performance that flash and burn. The escalating performance and ever-declining prices of conventional digital chips make them tough to beat.

Lyric's strong points? Error correction is a looming hurdle for the data storage industry. Moreover, nearly every major flash vendor has a probability project underway. Thus, the need is clear and the foreign-ness of the concept is eroding. The company can make its chips using conventional silicon processes and works with industry stalwart TSMC.

It will also try to license its technology to large vendors (Samsung and Toshiba are two large flash vendors) in a way that will let them differentiate their products. ARM did something similar and now dominates the market for cell phones and smart grid.

That should increase the company's chances of success.

Is less more?

Smooth-Stone, which came out of stealth mode for $48 million dollars from Battery Ventures and the investment fund of Abu Dhabi, says it can chip away at Intel's titanic dominance in server chips with a low-powered processor.

The chip is based around a design from ARM, the British chip design company. ARM chips power the vast majority of cell phones in the world and enjoy an increasing dominance in smart meters. ARM chips aren't as powerful as Intel's top server processors. Server chips, however, are often too good for their own use. They spend a large chunk of their time in idle, even with virtualization software that effectively allows a single server to perform multiple tasks at once.

The idea is that a lower-powered chip would save money by curbing power without crimping performance. Another plus: the chip is based around a familiar ARM design. Software and support for this architecture is bountiful. Data centers adopting this chip would not find themselves working with an unknown. Intel server chips can also be costly.

Uh oh. Ugly reality time. Many have tried to compete against Intel in computer and server processors: Cyrix, National Semiconductor, Sun Microsystems, AMD, IBM, Rise Semiconductor, Transmeta, Montalvo Systems. The successful ones like IBM and Sun were able to put their chips in their own servers, but Intel still owns 90-plus percent of the market for server chips. AMD has made gains, but has lost more money since its founding back in the late '60s than it has made. (I counted it up once and the numbers have only gotten worse.) The rest of these companies saw their ambitions shrivel up. Most are now gone.

And some -- Montalvo, Cyrix and Transmeta -- played the energy card at a time when Intel was particularly weak. All three burned through a lot more money than $48 million, too. Intel has also come out with a family of low-powered Atom chips that can go into servers.

Most high-end data center customers are not willing to gamble on performance. They would rather pay for more power for servers and scrimp someplace else. Some companies have begun to replace traditional hard-drive storage systems with DRAM and flash memory. It's more expensive but it boosts performance. Good luck convincing Yahoo that a few microseconds of delay on a search won't hurt you.

Smooth-Stone's goal is not impossible. Technically, they have a good argument and some sterling allies. But this is a quest that has left many dead. A more promising approach for a company trying to save power is to avoid a direct confrontation with Intel. Sandforce, for instance, has come up with a controller (a less elaborate microprocessor) for making flash drives more economical. It's an Intel-like market that Intel isn't that interested in. You will see other chip companies soon in this market.

Cell phones are perhaps also a more promising market.

The latest in our occasional series of how the 'game-changing' device is changing organizations' behaviors finds Stanford Medical School putting iPads to work in cutting paper waste and improving patient interactions.

A website claiming to be the first in the U.K. to offer cash for all types of electrical waste has launched this week.