标签: supercomputer

超級計算機 南安普敦大學的西蒙·考克斯教授(Professor Simon Cox, University of Southampton) 和他的團隊，在實驗室中透過64 個 Raspberry Pi，建立一個超級計算機，並且用樂高積木堆疊在一起，Raspberry Pi 一起解決一個單一的問題。而該項目已能夠降低成本從數百萬美元到十萬甚至上百美元一台超級計算機，這個 Raspberry Pi 的超級計算機更容易向學校及學生提供便宜且功能一樣的計算能力。

Well, this is such an exciting time. I'm happy to inform you that we succeeded in pushing the guys at Adapteva over their Kickstarter goal . I think we should all give ourselves a hearty pat on the back. Let's just remind ourselves as to what a difference we made. A few weeks ago, my chum Andreas from Adapteva called me to say "Help!" As you may recall, Andreas was the guy who left his job a few years ago and – working in his basement and living off his pension fund – single-handedly invented a new computer architecture. Andreas then designed his own System-on-Chip (SoC) called the Epiphany from the ground up – Including learning how to use all of the EDA tools – then took the device all the way to working silicon and a packaged prototype. When we chatted a few weeks ago, Andreas explained that he and his colleagues were trying to build a personal supercomputer called the Parallella for only $100. This little rascal is to be based on the combination of a Zynq All Programmable SoC from Xilinx and an Epiphany from Adapteva . In order to do this, the folks from Adapteva had launched a Kickstarter project a little over three weeks prior to our conversation. Their goal was to achieve $750,000 in pledges, but – when we talked – they had raised only around $420,000 . Pledges had started to slow down, the deadline (6:00pm Eastern Time on Saturday 27 October) was only a few days away, and things were starting to look grim indeed. A few days ago, I posted my first blog on this topic on EE Times. Following this blog, thing really started to ramp up – largely due to everyone who read the blog pledging themselves and/or spreading the words around via Twitter, Facebook, LinkedIn, and so forth. By morning of October 26, the folks from Adapteva were up to $611,504 , which means that $171,939 had been pledged in just a couple of days! This was amazing, but they still needed to raise another $138,496 for the Kickstarter project to go forward. I must admit that I was starting to become a little worried – and I also saw that Andreas had posted a comment somewhere saying "I don't think we can get anyone to write any more articles" – so I posted a follow-up blog. In that blog, amongst other things, I said "Oooooh, the things I will be able to do with this little beauty ... BUT ONLY IF THE KICKSTARTER PROJECT REACHES ITS GOAL!!! (Sorry ... I didn't mean to shout)." The reason I mention the "shouting" part will become apparent in a moment. I also started sending emails to all of the marketing and public relations folks I know asking them to spread the work. I must admit that by afternoon of October 26 I was nervously checking Kickstarter every ten minutes or so. We were getting closer and closer... And I also started to look at the comments that were coming in on Kickstarter, which is where I saw the following:   Hurray! How exciting! I especially enjoyed the part where Nico says "I like that he starts shouting." (OK, I must admit that I also liked the part where he said "Max is THE guy when it comes to FPGAs/SoCs." Nico is obviously a very intelligent young man with a bright future ahead of him :-) By Friday evening I was practically jumping up and down with excitement when Adapteva achieved their $750,000 goal. Phew! What a relief! When I got up on October 27 morning I checked again. Good Grief! They had continued to charge ahead and were now at something like $820,000. And it went on and on ... by the time the Kickstarter project automatically shut down, the total pledge amount was $898,921 – almost $150,000 over the original target!   I can only imagine the excitement at Adapteva. I received an email from Andreas saying how much he appreciated all of our efforts. I told him to stop talking and to get back to work on my Parallella supercomputer (grin). Now I cannot wait for the little beauty to arrive. Anyway, as I say, I think we all deserve to give ourselves a pat on the back for our efforts here. You never know – maybe Adapteva will one day grow into a big company (maybe even the next Apple :-) in which case we can all say "We were part of that!" So thanks to everyone for helping out!  

Some time ago I wrote about a guy Andreas Olofsson who left his job, formed a company called Adapteva . Working in his basement and living off his pension fund, he single-handedly invented a new computer architecture. Andreas designed his own System-on-Chip (SoC) from the ground up – Including learning how to use all of the EDA tools – then took the device all the way to working silicon and a packaged prototype... and that's when things really started to get interesting! The chip that Andreas designed is called the Epiphany . This is an array of processor cores, each equipped with its own local memory and a single-precision floating-point engine. Everything is designed so as to offer optimum performance while consuming as little power as possible. Epiphany is extremely scalable – The Epiphany-III (implemented at the 65nm node) boasts an array of 16 processors, while the Epiphany-IV (implemented at the 28nm node) features an array of 64 processors. The end result is that, when operating at peak performance, running at 800MHz, the Epiphany-IV offers 100 Gflops of raw computing power while consuming only 2W. This means that, at 50Gflops/Watt, the Epiphany-IV is 50 to 100X more efficient than anything else out there. Well, I just heard from Andreas. His current project is to create an open source personal supercomputer platform that anyone can buy for only $100, and that can be used to implement the most compute-intensive tasks like embedded and robotic vision, software-defined radios, and ... well, almost anything really. This supercomputer, which is called the Parallella , is based on a combination of the Zynq-7000 All Programmable SoC from Xilinx and the Epiphany from Adapteva as illustrated in the block diagram below.   The Zynq-based Parallella personal supercomputer. Initially there will be two versions of this little beauty—the version equipped with an Epiphany E16 (16 cores) will cost only $100, while the version equipped with an Epiphany E64 (64 cores) will cost only $199. I'm told that, even when running flat out, the Parallella equipped with an Epiphany E64 will consume as little as 5W! The guys and gals at Adapteva are currently using a Zynq evaluation board to extensively prototype the user experience of the Parallella boards. In our chat earlier, Andreas told me: "The user experience of running Ubuntu (one of the more popular flavors of Linux) on the Zynq is fantastic!" The picture below shows Andreas' Zynq evaluation board with an Epiphany daughter card plugged in via one of the FMC connectors.   A Zynq development board with an Epiphany daughter card. Are you familiar with Kickstarter.com ? This is a funding platform for creative projects—everything from films, games, and music to art, design, and technology. If people like a particular project, they can pledge money to make it happen. It's only if the project succeeds in reaching its funding goal that the backers' credit cards are charged—if the project falls short, no one is charged. The point is that Andreas and the folks at Adapteva have set Parallella up as a Kickstarter project. If you are interested, you can click here to learn more and—if you wish—make a pledge. Pledges can be as little as $15 or as much as $10,000 or more. In order to proceed, they need to raise $750,000 by the Kickstarter deadline of Saturday 27 October at 6:00 p.m. Eastern Daylight Time. I personally have every confidence that if they get the money they will succeed. After all, this project is led by the man who single-handedly designed a silicon chip in his basement. It's not often you get a chance to really "make a difference" in this world. I just pledged $99 myself . For this, when the project succeeds, I will receive my own Epiphany E16-based Parallella loaded with all of the development tools required to implement almost any project of my dreams. What say you? Are you with me?  

Before we talk about the incredible video, I have a nugget of knowledge I'd like to share. I just ran across the following tidbit of trivia on the Astrium website at www.astrium.eads.net A molecule of water will on average spend approximately 12 days in the atmosphere, 4,000 years in the oceans, two weeks to one year in soil moisture, and two weeks to 10,000 years in ground water. I have no idea who worked this out (or how... or why...) or even how accurate it is (how would one know?), but for some reason it gave me pause for thought... ...and what I thought was... why is it that numbers hold so much sway over us? Take the fact that, according to estimates by the United Nations, this year (2011) the population of the Earth passed 7 billion. I find this to be absolutely amazing... and absolutely terrifying. When I graduated high school there were only 4 billion people. Back in 1804 there were 1 billion. It took 123 years (until 1927) to reach 2 billion; 33 years (1960) to reach 3 billion; 14 years (1974) to reach 4 billion; 13 years (1987) to reach 5 billion; 12 years (1999) to reach 6 billion; and another 12 years (2011) to reach 7 billion. Currently we are expected to reach 8 billion by 2027 and 9 billion by 2046. The mind boggles... Or how about the fact that (as reported in the November 2011 issue of Scientific American) the human brain is estimated to be able to store 3.4 quadrillion bytes of information and perform 2.2 billion megaflops, all while consuming only 20 watts. By comparison, an iPad 2 can store 64 billion bytes and perform 170 megaflops while consuming only 2.5 watts. Meanwhile, a cat's brain blows the iPad out of the water with the ability to store 98 trillion bytes and perform 41 million megaflops (in the case of dogs, I know from personal experience that the brain of a Bichon Frise stores less data, and performs less processing, than a typical peanut ... and that's on a good day!). But wait, there's more, because the world's most powerful supercomputer (as of today – who knows what tomorrow will being), the K from Fujitsu, computes around four times faster (8.2 billion megaflops) and can store ten times more data (30 quadrillion bytes) than the human brain. On the other hand, we've still got things beat in the power consumption department, because the K requires 9.9 million watts to perform its magic. How did we get onto this stuff? I have no idea how you always seem to manage to lead me off into the weeds this way. What I really wanted to talk about was a really amazing video . It involves humming birds and bees and bats and plants and flowers and is absolutely beautiful and thought-provoking (make sure you have the sound turned up on your computer and run this in full screen mode if you can). On the one hand this is incredibly gorgeous. On the other hand it makes me feel sad because it reminds me of the scene from the science fiction film Soylent Green – the part where the character Solomon "Sol" Roth (played by the late, great Edward G. Robinson) goes to a euthanasia centre to die. Sol is lying on a bed watching beautiful videos of plants and animals and landscapes on a towering screen (nothing beautiful survives in this future world) accompanied by Beethoven's Pastoral (Symphony No. 6), which comes over as being hauntingly beautiful in this instance. This is an incredible scene that always brings tears to my eyes. And, of course, now I come to think about it, Soylent Green is set in the year 2022 when the population has grown to forty million people in New York City alone. Housing is dilapidated and overcrowded; homeless people fill the streets and line fire escapes and stairways; and food is very, very scarce. This film was created back in 1973 – the year before the world's population reached 4 billion; we can only hope that it wasn't prophetic...