When it comes to simulation, no project is too big for NASA’s supercomputing team. Recent efforts have included modeling of the Milky Way galaxy, examination of the world’s oceans, and simulation of the birth of the universe.
”We solve problems across all areas of NASA -- aerospace, earth science, and space science,” Bryan Biegel, deputy of NASA’s Advanced Supercomputing Division at NASA Ames, told Design News. “And the most powerful tool we can use to advance our knowledge is high-fidelity modeling.”
Indeed, the modeling that takes place at NASA Ames could be stoically described “high fidelity.” Using the agency’s biggest supercomputer, Pleiades, scientists have a stunning 162,496 Intel Xeon processor cores at their disposal, rated at 2.88 quadrillion floating point operations per second. To put it another way, the computer’s speed is such that a man punching in an operation per second on a calculator would take about 90 million years to accomplish what Pleiades can do in a single second.
The speed would be impressive enough by itself, but NASA also demonstrates the results of its simulations on a so-called “hyperwall,” which can read data directly from Pleiades file system over an InfiniBand connection. Hyperwall-2, the most recent embodiment of the technology, is said to be the world’s highest-resolution scientific visualization environment. Consisting of 128 screens, the 23 ft x 10 ft wall of displays is capable of rendering a quarter-billion pixel graphics.
The numbers are, in a word, overwhelming, and so are the results. During Design News’ recent tour inside NASA Ames (thanks to the sponsorship of Littelfuse Inc.), the agency’s supercomputing team demonstrated its stunning computing power. Using the hyperwall, it provided a visual depiction of NASA’s heavy lift launch vehicle, which will one day be “the most powerful rocket that mankind has produced.” The agency also showed us a simulation of the birth of universe, the evolution of the Milky Way, and the heat flow of the world’s oceans.
Check out this video, as we provide a glimpse of those simulations -- part of NASA’s effort to advance the state of human knowledge.
Thanks for covering this; it's really cool stuff, Rob. I have written about some of the super-computer simulations and it's truly amazing. It's good to see NASA remains the innovator it always has been as an agency and continues to demonstrate to the commercial sector some of the latest and greatest technology, as well as show us more of the world and universe around us!
Nice article on the power of computing that man has designed, built, and programmed. One wonders why the human brain continues to process in a uniquely human way that converts an overload of sensory inputs into a coherent existance. And yet, the super computer can beat us in Jeapordy, compute our beginnings, and provide a visual simulation of a heavy vehicle launch. However, one has to keep all this in perspective, the supercomputer has an OFF switch (at least all the Skynet people hope so).
In all this, our brains came about by evolutionary chance mutations? Wonder if the supercomputer could calculate the odds of a human evolutionary path using probabilities?
I had the same reaction. With all the money and technology they have access to, the average sports venue has better looking large scale video. I would have thought they could have gotten the joints between the monitors much smaller.
There are four things shown in this video: 1. Simulation of launch of NASA heavy lift vehicle. 2. Simulation of flow of world's oceans. 3. Simulation of birth of the universe (notice the clock counting down from 13 billion years to one). 4. Flow of air around helicopter rotor.
Yes, it's interesting how machines that humans themselves programmed beat us at our own games in every aspect of the word, GTOLover. This is quite an interesting thought to ponder, although i wouldn't begin to have a scientific answer for it. I suppose religious people (one of whom I am not) would say it has something to do with the soul, or even non-religious people might say it has to do with our emotions (although those can be scientifically proven to have something to do with how our brains work). For now, I suppose, it remains a mystery and subject of debate!
I don't think so... Computers will never as you say it beat us... They may do some operations faster than us, in higher quantity then us and even with fewer mistakes. However I guarantee you the most powerful computer in the world is only as smart as the collection of people that programmed it. If you went through an automata theory class you would know that a computer cannot write code in any language. It can cut and paste code that was already prewritten for it but that is. Everything you see in that screen was calculated by functions that the engineering team inserted in the simulation. If anyone should impress you it's not the supercomputer... It's the people that worked on days end to make it happen the supercomputer is just an aid. As for evolution what can happen, will happen pretty much sums it up. Yes millions of molecules can gather together and form a single cell organism. We know this can happen because it has happened already. The forces that caused this are a matter of religion
Atheist: Chaos theory
Agnostics (ME): Can't prove, Don't care, Thankful to be here bye....
Hi GTOlover, the answer to your question of why our brain processes information in the unique ways it does, is efficiency. The brain human brain dissipates about 20W of electricity, compared to the 4MW of a typical supercomputer of the type that I think you are comparing. They typically fill a large room compared to the smaller than football size of our brains.
Imagine how well we would have survived having to drag a mobile 4MW power plant while being so top heavy with a room on our shoulders :-)
The way our brain processes information is similar all other life (with brains) allowing a spider with a brain the size of the pointy end of a pin to weave intricate webs and go about its day to day routine.
And even though we like to think we have designed machines that are superior at our own games, there are areas where the best computers lag behind the best human minds. These are creativity and pattern recognition.
Sure there are examples of quasi creativity with computers, but the pure inpirational creativity people achieve (including eg. designing computers) is in a whole different league. A large part of this is the fact that even after years of research we still only have a very rudimentary idea of what creativity really is. Also creativity in regard to the unknown.
In regard to pattern recognition consider you meet someone at the age of say 16 or 17 that is about the same age. and for what ever reason part ways. A human can be walking down a street in a completely unfamiliar area some 50 or more years later and see that person in the crowd with all of the signs of aging dying of hair, addition of glasses new scars, etc etc, and still recognise that person while never actually expecting to see them or be actively looking for them, and that in a see of faces. and all within seconds. It also works for a long lost familiar face in a see of familiar faces. As far as I'm aware while the best computers running the highest speed algorithms can only run through a list of faces and compare them sequentially, and are not particularly tolerant of changes. The person having seen 1000's to 10's of 1000's of faces over a lifetime will know which is a new face and which one is familiar in an instant.
Similar things are at play when you hear a familar tune after years and know after 3 bars what the name is
Obviously not everyone can pull this off, just as not everyone is creative, but the best compared to the best I don't see our pattern recognition abilities being outdone until quantum computers have been played with for a while.
The main reason we excel in these areas is that the human brain is around 100 billion computers running in parallel at around 200Hz and according to the university of Alberta due to the way the neurons are wired the brain performs roughly 20 million billion calculations per second. Of course this isn't the full story, because neurons are analogue computers with maybe 1000 or so levels of sensitvity which magnifies the potential again, but of course not all of the brain is involved in our thinking processes, so suffice to say its processing power (for the right type of problem) is beyond the realm of current (non-quantum) computers.
Never the less, I certainly have a healthy respect for the designers of large computing arrays such as NASA's and their brainchilds, as they are an extention of the mind of researchers and engineers the world over and have revolutionised life in affluent countries by allowing them to concentrate on the aspects of problems better handled by the little grey cells.
Interesting thought, but it discounts the chance that we don't run out of energy or don't destroy ourselves before we finally understand how thinking actually works.
Is it possible for someone to understand their thinking processes to sufficient detail to be able to program a computer with intelligence?
I think so.
Will we as a species survive long enough to achieve this?
I don't know.
a 3rd option is that we build a quantum computer modelled on synapes and the likes and hit on the magic formula of our brains and the thing begins to think for itself and we're left none the wiser as to how thinking works. We may have to ask it (if it wants to tell us). Watch "The Forbin Project" from circa 1973 to understand where I'm coming from.
I remember Bill Gates saying we will never need more than 640k and someone else said we would never fly, and the list goes on. I don't think we have yet reached the level of understanding necessary to say with absolute certainty that this or that will never happen in technology.
The question of whether engineers could have foreseen the shortcut maintenance procedures that led to the crash of American Airlines Flight 191 in 1979 will probably linger for as long as there is an engineering profession.
More than 35 years later, the post-mortem on one of the country’s worst engineering disasters appears to be simple. A contractor asked for a change in an original design. The change was approved by engineers, later resulting in a mammoth structural collapse that killed 114 people and injured 216 more.
If you’re an embedded systems engineer whose analog capabilities are getting a little bit rusty, then you’ll want to take note of an upcoming Design News Continuing Education Center class, “Analog Design for the Digital World,” running Monday, Nov. 17 through Friday, Nov. 21.
Focus on Fundamentals consists of 45-minute on-line classes that cover a host of technologies. You learn without leaving the comfort of your desk. All classes are taught by subject-matter experts and all are archived. So if you can't attend live, attend at your convenience.