Its not easy being green
It's not easy ...
Recently, NVIDIA won a place in IBM’s datacenter workhorses. Previously, IBM had collaborated with others to produce the Cell processor, aimed at providing a scalable high performance and low power computing solution.
It’s not enough to be green; you have to be cheap, easy to program and orders of magnitude better at crunching data.
This observation is important. In the rush to be green, you can’t violate the laws of physics or the free market. Innovations need to stand on their merits, or they will only succede with unsustaniable subsidies. Policy and lifestyle switches that don’t deliver a better quality of life will not stand the test of time.
There are pleanty of examples of green technology that have survived this gauntlet. Interface carpets completely reinvented their enviromnetal impact and emerged more profitable and popular because of it. IKEA reduced their fuel consumption and packaging and have reaped large profits from reducing overhead.
The cell processor was a good idea. But without the full package, just being low power didn’t cut it.
Source – GigaOm – Nvidia Shows Off Its Survival Skills With IBM Win
On my way… (part 2/2)
"I'm on my way..."
Last week I wrote about the rise of performance per watt as the key metric for computing, and how parallel processing helped bring this to the forefront for designers. This week, I examine the other driving force behind performance per watt: mobile.
Since Star Trek and Dick Tracy, the goal of technology is freedom and mobility. At the same time that processing power began to plateau for single cores, mobile devices began to come into their own. This trend of focusing on consuming (and now creating) media on the go has become the driving force in technology. Wireless data rates have long since eclipsed what old dialups could offer. Video is now the common language of mobile platforms, and all of that demands massive processing power without limiting the battery life of the device. Over the past several years, desktop growth has remained stagnant, while mobile has grown year-on-year.
That twin need of powerful computing and long uptime requires low power design to be present throughout the design of a system. Everything from the display to the processor must look at how much power is consumed relative to the end user’s benefit. Techniques in semiconductor design that used to be esoteric (shutting off sections that are not in use) have become mainstream. New techniques like Dynamic Voltage and Frequency Scaling (DVFS) are becoming even more popular to keep the focus on maximizing performance per watt.
Combining together powerful mobile processors, like the Snapdragon or Atom, with programming and system techniques like parallel processing, the impact of computing on power consumption can be limited. The same kind of cost-benefit analysis needs to ripple throughout our world to help technology continue to deliver life changing impact without taxing our already strained system.
(For more indepth analysis, please see the briefing on microprocessors in this months’ MIT TechReview.)
The new black …
Raw performance is no longer acceptable in a world of pervasive computing and rising costs of energy
In government, the common saying is that “all politics are local”. For low power semiconductors, it would be “all computing is mobile”. The confluence of mobile computing and parallel processing is key to managing the growing demand for power from computing, which will have a huge impact on the energy needs of the planet. The new watchword is “performance per watt”.
The building block of the internet, personal computing, etc… are processors. In the past, processor design power houses like Intel and AMD focused on pure performance. Processors were rated based on clock frequency. As time went on, the demands on computing began to grow faster than what processor designers could squeeze out of their designs. Software then advanced to take advantage of parallel processing. This allowed slower, smaller processors to accomplish bigger tasks faster than a single high horsepower processor. That was a huge shift in the industry. The impact of parallel processing on the energy grid is tremendous. The economy of scale from ganging up a large number of processors means that a small increase in power for one processor results in a major impact on the grid. This is why low power design for the processors is so important. Saving a few milliwatts adds up when you have thousands of processors active in a data centers. Multiply that across the global data centers of Microsoft, Google, Cisco, … and you see the impact.
At the same time that parallel processing became popular, another phenomenon came on the scene: mobile computing. I’ll continue next week on this.
China and Low Power Semiconductor Design
The growing importance of China for energy engineering came front and center for me over the weekend. A short while back, I authored an article for new low power techniques in semiconductor design, for all markets in general, but the request came specifically for the Chinese semiconductor market. It was published yesterday at 中国电子应用网.
(English translation coming next, and in print, in Power Systems Design).
贯穿整个IC实现流程的集成化低功耗设计技术
时间：2010-04-08 09:09:24来源：中国电子应用网[原创] 评论 0 条
作者：Rob Knoth 高级技术产品经理
微捷码设计自动化有限公司 美国加州圣荷塞
www.magma-da.com降低功耗是现代芯片设计最具挑战性需求之一。采用单点工具流程时，往往只有到了设计流程后期阶段才会去考虑降低功耗的需求，从而经常导致大量问题和延时。微捷码设计自动化有限公司高级技术产品经理Rob Knoth向我们解释了‘为何功率优化应是完整设计流程必不可少的集成组件’。
起初，低功耗设计技术主要是用于移动产品。而今插在电源插座上的产品数不胜数，它们都在不断吸收着电流，全世界大量功率都浪费在了这些产品上。今天，政府正积极要求电子企业遵从更严格的要求来帮助降低全球功耗。低功耗设计与每个人息息相关。现在低功耗需求无处不在，而且变得更具挑战性。
低功耗设计，不论是动态功耗还是静态功耗，均要求设计流程各个阶段时序、功耗和面积间复杂的折衷权衡。这些需求相互间联系密切，要想解决这些需求，低功耗分析和优化引擎必须集成并运用于从RTL规格到GDSII输出的整个流程中。由于芯片尺寸还在持续增长，因此这一流程必须是可缩放的，否则它将会对设计师工作效率造成限制。
New hope from the desert?
The past two weeks have seen new life for Desertec.
Desertec is an ambitious plan to harvest solar energy from the deserts of North Africa and pipe it to European consumers via an improved transmission network. I first heard of them several months ago, and then it was radio silence. (side note, they need help leveraging social media. None of their recent announcements were sent on their Twitter feed)
One of the premier companies in photovoltaic panels, First Solar, joined the consortium this past week (article in NYT’s Green Inc). First Solar is a leader in driving down the cost of solar energy. They were the first to break the important $1/W barrier, the threshold for “affordable” energy. (side note: behemoth Applied Materials is in hot pursuit of this same cost point – article in MIT TechReview). Having a credible partner like First Solar on board adds important momentum behind this effort. This week the Italian power company Enel, and the French builder of energy-efficient construction materials Saint Gobain, also joined the Desertec project.
What I like about the Desertec Project is its scope. It is a truly global project. Harnessing a resource where it is most plentiful and engineering a low impact method to provide it where it is needed is a massive undertaking. To make it successful, it needs to drive down the cost and drive up the reliability of harvesting solar energy. First Solar is an important part of this. The scale helps with managing the costs of raw materials, but the challenges of an installation this massive and widespread will need innovation to be pulled off. There is a lot that the oil industry can bring to the table here. These companies have perfected the art of deploying complex systems in difficult global environments. Desertec could be different with benefits for the local communities. There is a big difference between bringing oil out of the ground and putting it in a car or light socket. The DC energy harvested by solar could be immediately used to raise the standards of living across North Africa. Basic examples are refrigeration, lighting, and powering cell stations for better communication.
The connection between the generation and the consumption is huge. This presents a lot of challenges that are needed for rebuilding our aging electrical grid. Lossless transmission lines, low-cost construction and deployment, and remote monitoring are all technologies that will make Desertec and the smart grid a reality.
While this large-scale system is an amazing challenge, it is important that it does not remove the focus on local, distributed generation. In a coming post, I’ll describe some of the recent successes here. In particular, the grassroots work by 1BOG (one block off the grid).
The scope of Desertec is key to its success and the solution to many of the problems we face globally. I look forward to following them and seeing this story unfold.
-Rob
March 22, 2010
Above Lake Tahoe, heading to Boston.
Be The Change
This week, Bloom Energy unveiled their Bloom Box on 60 Minutes, followed by a press event at their campus in Sunnyvale. There is a lot of hype for Bloom to live up to, but looking at their 10 year stealth development, the secret beta program at Ebay, and founder K.R.Sridhar’s plans for future developments, I have a lot of faith.
Greentech Media did a great interview with K.R.Sridhar at the event this past week.
One of the quotes from the media circus really made an impact on me.
“Silicon Valley is learning some hard and important skills, mainly making stuff again,” said Mr. Doerr, a partner at the venture-capital firm Kleiner Perkins Caufield & Byers and a Bloom Energy board member.
This hits at the core of what I believe is needed in our society in general, and in the valley in particular. We’ve spent a lot of time and energy on infinite permutations of social networking. Morphing from a culture of pure software and app development to produce a system is a big change. What companies like Bloom Energy, Tesla, and Better Place have done is dream big and build big. Tackling the hard problems earns you respect and (hopefully) reward.
-Rob
En route to Bend, OR
PS. For good reading on solving real problems, check out Vivek Wadhwa’s piece on TechCrunch (“What’s better: Saving the world or building another facebook app?”)
Companies in the news
This past week saw some recent activity for cleantech companies…
It was announced on GreenTechMedia’s site that Bloom Energy will go public this week with their fuel cell technology. They have a very good history of rumors, facts, and research on this fuel cell startup that is well worth reading. The technology behind Bloom’s cells is a solid-oxide fuel cell. Their list of current/potential customers is extensive. Everyone from Wal-mart to the CIA. This is a company worth watching.
(NOTE: Between writing this on Friday and today, Bloom changed their home page to a countdown clock. 2 days and counting as of Sunday… tik tok)
Broadband Solar (BSX), a spin-off from Stanford, announced some promising advances for thin-film solar. They are using nanoparticles to enhance the efficiency of existing solar cells. The technology behind this is the field of plasmonics. The key to solar power is a solar cell absorbing the energy from photons. By shaping the surface of the cells, they have been able to increase the opportunity for absorbtion. This technique promises to boost the efficiency of cheaper types of silicon. A great synergy between different fields of study, and people to watch.
MIT Tech Review did an excellent article on this company.
On the other end of the spectrum, Google’s application to sell energy was approved by FERC. It’s not immediately clear what will come from this, but it’s a safe bet that it could be used to kick-start existing utilities into adopting more modern practices. This is exactly what Google’s recent move into broadband Internet is doing. Google makes very big waves when it moves, and this is no exception.
Google’s clean energy initiative.
Venture Beat article on the release.
-Rob
enRoute to San Jose, CA.
Feb 21st, 2010
Show me the money
A few months back, I mentioned the work that Amryis was doing to create an organism that would produce hydrocarbons from sugars (On my radar…). On friday, rumors circulated that Amryis could be looking to go IPO in the second quarter of 2010. (See Greentechmedia’s article).
This is fantastic news, as long as it becomes reality. There is a lot of excitement around the cleantech industry (Solyndra also was an IPO rumor at the end of 2009). And taking one or two of these rising stars and showing a realistic path from “cool idea” to “great company” to “profitable investment” is what we need to rally more of the population. Tax breaks, incentives, stimulus funds are a necessary evil, but for the cleantech industry to really become self-sustaining, it needs to follow the example of the semiconductor and networking industries. Working in the semiconductor industry in San Jose, I’ve seen the results when technology and business are joined together.
Tesla (On my radar …) is another company that has flirted with going public. In a confusing announcement this week, they are potentially planning to delay their only production vehicle (the roadster) in favor of a yet-to-see-the-road sedan (Reuter’s article here). I do not claim to understand the inner workings of the automotive industry, but this seems quite risky…
There will always be companies that don’t succeed in making their technologies production worthy. Economic Darwinism is required to make the system work. But at this point in the life of the cleantech industry, it’s especially important to see some home runs.
There are some unanswered questions here, and I will be following up on this topic in coming posts.
-Rob
Bend, Oregon
Feb 7 2010
Brief notes…
I’m returning from a week in Japan, visiting customers for Magma. Business is always local. Integrated circuit designs vary across the globe, and the ones in Japan are large. However, the other important dimension is low power. The overwhelming theme was the need to integrate even larger systems, all with low power a critical requirement. Working in technical product management for EDA, it’s rewarding to help enable this trend in electronics.
The downside, it keeps one busy. A few notes from the past weeks…
- Innocentive – One of the most eye catching challenges I’ve seen. Lightning as an Alternative Energy Source.
- Venezuela –
Dramatic swings over the past month related to energy. At the start of January, they faced a power crisis. Demand was far outstripping supply, even in a nation rich with oil. Today, the government announced that they would expand production by 17%. This kind of demand growth can either be a big boon for clean energy, or a bigger push toward climate change. - State of the Union - This past week, the president placed the spotlight firmly on renewable energy and its importance in the world economy.
I know there have been questions about whether we can afford such changes in a tough economy. I know that there are those who disagree with the overwhelming scientific evidence on climate change. But here’s the thing — even if you doubt the evidence, providing incentives for energy-efficiency and clean energy are the right thing to do for our future -– because the nation that leads the clean energy economy will be the nation that leads the global economy. And America must be that nation.
-
Obama's State of The Union - 2010
China – While we’ve argued about global warming, the rest of the world has been busy. China is the prime example.
As China takes the lead on wind turbines, above, and solar panels, President Obama is calling for American industry to step up. Image from Shiho Fukada for The New York Times
They have been growing in all dimensions and this growth requires a huge amount of energy to sustain it. In the past, they have been criticized for being too large of a polluter. They still do get a significant part of their energy from coal and oil, almost 2/3 this year from coal, the balance from nuclear and hydropower, and ~4% from wind solar and biomass. However, they plan for 8% by 2020. What makes this more critical is that their total power generation will pass the US in 2012. What’s really fueling their rise in renewable energy is scale. Many major multinational companies, like Vestas, has located production facilities there, and China became the world’s largest producer of wind turbines. They also have a strong government push to require new wind power farms to be connected to the national grid. Not all is roses, but China’s record of results in renewable energy is impressive.
- Publications. Coming soon, I will be publishing an article in Power Systems Design on the need to consider low power for integrated circuit design flows, and the challenges of scalability to match the growing design sizes. I was recently interviewed by Chip Design Magazine on low power and IC design.
Exciting times.
-Rob
En Route from Tokyo back home…
January 31st, 2010
The growing grid
One of the big issues in the realization of a smart grid in this country is a simple one: access. To effectively build a new power distribution infrastructure will require right of way and access to new areas that are not currently served by the existing electrical grid. Many of the locations for wind farms and solar arrays are in remote areas. Getting a contiguous connection between the generation of power and consumers is important to making the smart grid a reality. There have been issues with fiber optics being laid along railroad right of ways and issues with power lines in Oregon and SW Washington.
A lot of progress has been made on this front in Europe. Recently The Guardian reported on planning from the EU to build a continent-wide supergrid to connect wind farms in the North sea, solar arrays in North Africa, and other power sources with hydropower storage and consumers.
Sun, wind and wave-powered: Europe unites to build renewable energy ‘supergrid’
• North Sea countries plan vast clean energy project • €30bn scheme could offer weather-proof supply
All those involved also have an eye on the future, said Wilkes. “The North Sea grid would be the backbone of the future European electricity supergrid,” he said. This supergrid, which has support from scientists at the commission’s Institute for Energy (IE), and political backing from both the French president, Nicolas Sarkozy, and Gordon Brown, would link huge solar farms in southern Europe – producing electricity either through photovoltaic cells, or by concentrating the sun’s heat to boil water and drive turbines – with marine, geothermal and wind projects elsewhere on the continent. Scientists at the IE have estimated it would require the capture of just 0.3% of the light falling on the Sahara and the deserts of the Middle East to meet all Europe’s energy needs.
The DESERTEC Foundation
This new effort to unite the energy sources and users in Europe is also aligned with another venture: The DESERTEC foundation. Their goal is to build a high voltage DC grid to connect CPV plants in the Sahara and Middle East with consumers in Europe. In the middle of December, the World Bank allocated $5.5B to help jump-start this initiative to get a 1GW section going. This is very encouraging. My only concern is the combination of CPV which needs water to turn to steam in the desert. But, nothing is insurmountable, I wrote about few months back in California…
On the home front, an advance in the smart grid came out of Minnesota. Minnesota Power bought rights to a high voltage direct current line (HVDC) between their customers and wind power in North Dakota. HVDC lines are the backbone of the smart grid due to their superior loss over large distances. Additionally, keeping the power in DC form requires one less conversion. That keeps the second law of thermodynamics happy.
Minnesota Power reaches milestone in renewable energy efforts
with purchase of interstate transmission lineDuluth, Minn.—Minnesota Power, an ALLETE company (NYSE: ALE), has finalized the purchase of a 465-mile transmission line, signaling a new era of renewable energy transport and development for the Company.
The grid is really a massive challenge. It will require the cooperation of private landowners, private industry, government agencies, and NGO’s. Not all news will be good. There will be trial and error, but in the end we will all benefit if it’s done right. The Europeans appear to have a good start, and hopefully we will follow in the States…
-Rob
Bend, OR
January 6th 2010
Rob Knoth
