“Science budgets have been significantly dropping over the last decade,” Khanna said. “Here’s a way that people can do science projects less expensively. This new web site will show people how to move forward.”

Typically, scientists rent supercomputer time by the hour. A single simulation can cost more than 5,000 hours at $1 per hour on the National Science Foundation’s TeraGrid computing infrastructure. “For the same cost, you can build your own supercomputer and it works just as well if not better,” Khanna said. “Plus, you can use it over and over again, indefinitely.” The cost for his initial Playstation grid was $4,000.

The guide is freely available to the public under an open source license.

The Cluster Workshop project is partially funded by the National Science Foundation and was first announced and demonstrated at the 2nd Annual Georgia Tech, Sony/Toshiba/IBM Workshop on Software and Applications for the Cell/B.E. Processor.

“This opens up a huge door to partnerships with industry and other universities,” said Khanna, noting that the UMass Dartmouth College of Engineering has an interest and focus in simulation sciences. Tyco Electronics (through the UMass Dartmouth Advanced Technology and Manufacturing Center in Fall River), Sony, Terra Soft Solutions and IBM are among the companies already involved with this effort. The scientists are seeking input from industry members and researchers to determine future project direction.

“We hope to continue to bring supercomputing to a broader audience by providing tools that simplify the use of these systems,” said Poulin, who specializes in distributed pattern recognition and artificial intelligence.

For the full guide, how-to and screenshots head over to:
http://www.ps3cluster.org/

Provided by University of Massachusetts Dartmouth

http://www.physorg.com/news148749271.html

Tsubame was ranked 29th-fastest supercomputer in the world in the latest Top 500 ranking with a speed of 77.48T Flops (floating point operations per second) on the industry-standard Linpack benchmark.

While its position is relatively good, that’s not what makes it so special. The interesting thing about Tsubame is that it doesn’t rely on the raw processing power of CPUs (central processing units) alone to get its work done. Tsubame includes hundreds of graphics processors of the same type used in consumer PCs, working alongside CPUs in a mixed environment that some say is a model for future supercomputers serving disciplines like material chemistry.

Graphics processors (GPUs) are very good at quickly performing the same computation on large amounts of data, so they can make short work of some problems in areas such as molecular dynamics, physics simulations and image processing.

“I think in the vast majority of the interesting problems in the future, the problems that affect humanity where the impact comes from nature … requires the ability to manipulate and compute on a very large data set,” said Jen-Hsun Huang, CEO of Nvidia, who spoke at the university this week. Tsubame uses 680 of Nvidia’s Tesla graphics cards.

Just how much of a difference do the GPUs make? Takayuki Aoki, a professor of material chemistry at the university, said that simulations that used to take three months now take 10 hours on Tsubame.

Tsubame itself – once you move past the air-conditioners – is split across several rooms in two floors of the building and is largely made up of rack-mounted Sun x4600 systems. There are 655 of these in all, each of which has 16 AMD Opteron CPU cores inside it, and Clearspeed CSX600 accelerator boards.

The graphics chips are contained in 170 Nvidia Tesla S1070 rack-mount units that have been slotted in between the Sun systems. Each of the 1U Nvidia systems has four GPUs inside, each of which has 240 processing cores for a total of 960 cores per system.

The Tesla systems were added to Tsubame over the course of about a week while the computer was operating.

“People thought we were crazy,” said Satoshi Matsuoka, director of the Global Scientific Information and Computing Center at the university. “This is a ¥1 billion (US$11 million) supercomputer consuming a megawatt of power, but we proved technically that it was possible.”

The result is what university staff call version 1.2 of the Tsubame supercomputer.

“I think we should have been able to achieve 85 [T Flops], but we ran out of time so it was 77 [T Flops],” said Matsuoka of the benchmarks performed on the system. At 85T Flops it would have risen a couple of places in the Top 500 and been ranked fastest in Japan.

There’s always next time: A new Top 500 list is due out in June 2009, and Tokyo Institute of Technology is also looking further ahead.

“This is not the end of Tsubame, it’s just the beginning of GPU acceleration becoming mainstream,” said Matsuoka. “We believe that in the world there will be supercomputers registering several petaflops in the years to come, and we would like to follow suit.”

Tsubame 2.0, as he dubbed the next upgrade, should be here within the next two years and will boast a sustained performance of at least a petaflop (a petaflop is 1,000 teraflops), he said. The basic design for the machine is still not finalized but it will continue the heterogeneous computing base of mixing CPUs and GPUs, he said.

Source:
http://goodgearguide.com.au/article/270416

I’ve tried to show real-world throughputs (the sources are posted as a comment) rather than theoretical maximums. Time units are nanoseconds (ns, 10-9 seconds), milliseconds (ms, 10-3 seconds), and seconds (s). Throughput units are in megabytes and gigabytes per second. Let’s start with CPU and memory, the north of the northbridge:

Latency and throughput in an Intel Core 2 Duo computer, North Side

The first thing that jumps out is how absurdly fast our processors are. Most simple instructions on the Core 2 take less than a cycle to execute, hence less than a third of a nanosecond at 3.0Ghz. For reference, light only travels ~4 inches (10 cm) in the time taken by a clock cycle. It’s worth keeping this in mind when you’re thinking of optimization – instructions are comically cheap to execute nowadays.

As the CPU works away, it must read from and write to system memory, which it accesses via the L1 and L2 caches. The caches use static RAM, a much faster (and expensive) type of memory than the DRAM memory used as the main system memory. The caches are part of the processor itself and for the pricier memory we get very low latency. One way in which instruction-level optimization is still very relevant is code size. Due to caching, there can be massive performance differences between code that fits wholly into the L1/L2 caches and code that needs to be marshalled into and out of the caches as it executes.

Normally when the CPU needs to touch the contents of a memory region they must either be in the L1/L2 caches already or be brought in from the main system memory. Here we see our first major hit, a massive ~250 cycles of latency that often leads to a stall, when the CPU has no work to do while it waits. To put this into perspective, reading from L1 cache is like grabbing a piece of paper from your desk (3 seconds), L2 cache is picking up a book from a nearby shelf (14 seconds), and main system memory is taking a 4-minute walk down the hall to buy a Twix bar.

The exact latency of main memory is variable and depends on the application and many other factors. For example, it depends on the CAS latency and specifications of the actual RAM stick that is in the computer. It also depends on how successful the processor is at prefetching – guessing which parts of memory will be needed based on the code that is executing and having them brought into the caches ahead of time.

Looking at L1/L2 cache performance versus main memory performance, it is clear how much there is to gain from larger L2 caches and from applications designed to use it well. For a discussion of all things memory, see Ulrich Drepper’s What Every Programmer Should Know About Memory (pdf), a fine paper on the subject.

People refer to the bottleneck between CPU and memory as the von Neumann bottleneck. Now, the front side bus bandwidth, ~10GB/s, actually looks decent. At that rate, you could read all of 8GB of system memory in less than one second or read 100 bytes in 10ns. Sadly this throughput is a theoretical maximum (unlike most others in the diagram) and cannot be achieved due to delays in the main RAM circuitry. Many discrete wait periods are required when accessing memory. The electrical protocol for access calls for delays after a memory row is selected, after a column is selected, before data can be read reliably, and so on. The use of capacitors calls for periodic refreshes of the data stored in memory lest some bits get corrupted, which adds further overhead. Certain consecutive memory accesses may happen more quickly but there are still delays, and more so for random access. Latency is always present.

Down in the southbridge we have a number of other buses (e.g., PCIe, USB) and peripherals connected:

Latency and throughput in an Intel Core 2 Duo computer, South Side

Sadly the southbridge hosts some truly sluggish performers, for even main memory is blazing fast compared to hard drives. Keeping with the office analogy, waiting for a hard drive seek is like leaving the building to roam the earth for one year and three months. This is why so many workloads are dominated by disk I/O and why database performance can drive off a cliff once the in-memory buffers are exhausted. It is also why plentiful RAM (for buffering) and fast hard drives are so important for overall system performance.

While the “sustained” disk throughput is real in the sense that it is actually achieved by the disk in real-world situations, it does not tell the whole story. The bane of disk performance are seeks, which involve moving the read/write heads across the platter to the right track and then waiting for the platter to spin around to the right position so that the desired sector can be read. Disk RPMs refer to the speed of rotation of the platters: the faster the RPMs, the less time you wait on average for the rotation to give you the desired sector, hence higher RPMs mean faster disks. A cool place to read about the impact of seeks is the paper where a couple of Stanford grad students describe the Anatomy of a Large-Scale Hypertextual Web Search Engine (pdf).

When the disk is reading one large continuous file it achieves greater sustained read speeds due to the lack of seeks. Filesystem defragmentation aims to keep files in continuous chunks on the disk to minimize seeks and boost throughput. When it comes to how fast a computer feels, sustained throughput is less important than seek times and the number of random I/O operations (reads/writes) that a disk can do per time unit. Solid state disks can make for a great option here.

Hard drive caches also help performance. Their tiny size – a 16MB cache in a 750GB drive covers only 0.002% of the disk – suggest they’re useless, but in reality their contribution is allowing a disk to queue up writes and then perform them in one bunch, thereby allowing the disk to plan the order of the writes in a way that – surprise – minimizes seeks. Reads can also be grouped in this way for performance, and both the OS and the drive firmware engage in these optimizations.

Finally, the diagram has various real-world throughputs for networking and other buses. Firewire is shown for reference but is not available natively in the Intel X48 chipset. It’s fun to think of the Internet as a computer bus. The latency to a fast website (say, google.com) is about 45ms, comparable to hard drive seek latency. In fact, while hard drives are 5 orders of magnitude removed from main memory, they’re in the same magnitude as the Internet. Residential bandwidth still lags behind that of sustained hard drive reads, but the ‘network is the computer’ in a pretty literal sense now. What happens when the Internet is faster than a hard drive?

I hope this diagram is useful. It’s fascinating for me to look at all these numbers together and see how far we’ve come. Sources are posted as a comment. I posted a full diagram showing both north and south bridges here if you’re interested

http://duartes.org/gustavo/blog/post/what-your-computer-does-while-you-wait

At the heart of the new Tesla personal supercomputer are three or four Nvidia Tesla C1060 computing processors, which appear similar to a high-performance Nvidia graphics card, but without any video output ports. Each Tesla C1060 has 240 streaming processor cores running at 1.296 GHz, 4 GB of 800 MHz 512-bit GDDR3 memory and a PCI Express x16 system interface. While typically using only 160-watts of power, each card is capable of 933 GFlops of single precision floating point performance or 78 GFlops of double precision floating point performance.

While the Tesla C1060 computing processors are powerful, they have a massively-parallel architecture that may have trouble with serial computing modes. The Tesla personal supercomputer also features a powerful Intel or AMD quad-core processor, which is another important component of the system, especially when dealing with these serial computing modes. The Tesla personal supercomputer includes at least 4 GB of system memory per included Tesla C1060 card and at least a 1200- to 1350-watt power supply. System noise is rated at less than 45 dbA and the supported operating systems include Windows XP, Red Hat and SUSE.

It is pretty clear that the Tesla personal supercomputer is not designed for PC gaming, but rather for highly computational research and professional work. Ideal types of applications for this system would likely include the processing of large sets of consistent data, such as transcoding a DVD or studying seismic activity. The GPU-based Tesla Personal Supercomputer is now available from retail HPC OEMs, system builders and resellers, including Dell, Asus, Western Scientific and Microway. Prices vary depending on configuration, but expect to pay around $10,000 for your own personal supercomputer.

http://www.tomshardware.com/news/Nvidia-Tesla-Supercomputer,6616.html

“The mind has an amazing ability to integrate ambiguous information across the senses, and it can effortlessly create the categories of time, space, object, and interrelationship from the sensory data,” says Dharmendra Modha, the IBM scientist who is heading the collaboration.

“There are no computers that can even remotely approach the remarkable feats the mind performs,” he said.

“The key idea of cognitive computing is to engineer mind-like intelligent machines by reverse engineering the structure, dynamics, function and behaviour of the brain.”

‘Perfect storm’

IBM will join five US universities in an ambitious effort to integrate what is known from real biological systems with the results of supercomputer simulations of neurons. The team will then aim to produce for the first time an electronic system that behaves as the simulations do.

The longer-term goal is to create a system with the level of complexity of a cat’s brain.

Prof Modha says that the time is right for such a cross-disciplinary project because three disparate pursuits are coming together in what he calls a “perfect storm”. We are going not just for a homerun, but for a homerun with the bases loaded – Dharmendra Modha, IBM Almaden Research Center

Neuroscientists working with simple animals have learned much about the inner workings of neurons and the synapses that connect them, resulting in “wiring diagrams” for simple brains.

Supercomputing, in turn, can simulate brains up to the complexity of small mammals, using the knowledge from the biological research. Modha led a team that last year used the BlueGene supercomputer to simulate a mouse’s brain, comprising 55m neurons and some half a trillion synapses.

“But the real challenge is then to manifest what will be learned from future simulations into real electronic devices – nanotechnology,” Prof Modha said.

Technology has only recently reached a stage in which structures can be produced that match the density of neurons and synapses from real brains – around 10 billion in each square centimetre.

Networking

Researchers have been using bits of computer code called neural networks that seek to represent connections of neurons. They can be programmed to solve a particular problem – behaviour that appears to be the same as learning. But this approach is fundamentally different.

“The issue with neural networks and artificial intelligence is that they seek to engineer limited cognitive functionalities one at a time. They start with an objective and devise an algorithm to achieve it,” Prof Modha says.

“We are attempting a 180 degree shift in perspective: seeking an algorithm first, problems second. We are investigating core micro- and macro-circuits of the brain that can be used for a wide variety of functionalities.”

The problem is not in the organisation of existing neuron-like circuitry, however; the adaptability of brains lies in their ability to tune synapses, the connections between the neurons.

Synaptic connections form, break, and are strengthened or weakened depending on the signals that pass through them. Making a nano-scale material that can fit that description is one of the major goals of the project.

“The brain is much less a neural network than a synaptic network,” Modha says.

First thought

The fundamental shift toward putting the problem-solving before the problem makes the potential applications for such devices practically limitless.

Free from the constraints of explicitly programmed function, computers could gather together disparate information, weigh it based on experience, form memory independently and arguably begin to solve problems in a way that has so far been the preserve of what we call “thinking”.

“It’s an interesting effort, and modelling computers after the human brain is promising,” says Christian Keysers, director of the neuroimaging centre at University Medical Centre Groningen. However, he warns that the funding so far is likely to be inadequate for such an large-scale project.

That the effort requires the expertise of such a variety of disciplines means that the project is unprecedented in its scope, and Dr Modha admits that the goals are more than ambitious.

“We are going not just for a homerun, but for a homerun with the bases loaded,” he says.

Story from BBC NEWS:

http://news.bbc.co.uk/go/pr/fr/-/2/hi/science/nature/7740484.stm

Last year the company had a 16-qubit quantum computer that founder and CTO Geordie Rose claimed was the most powerful quantum computer ever built and the first ever to run commercially-relevant applications. This year it has 28 qubits, it can recognise photos of famous landmarks – and you might soon be able to use it over the Web.

That’s far ahead of most other quantum computing developments and D-Wave has managed it by using semiconductor manufacturing techniques and existing chip fabs instead of optical circuits, quantum dots, laser containment or other approaches requiring exotic manufacturing techniques. D-Wave is also working on the other half of the problem; the programming tools for writing applications that take advantage of what quantum computing promises to deliver.

Rose defines a quantum computer as “a machine that harnesses the language of nature at the most fundamental level to gain, in some cases, extremely impressive performance gains over conventional computers. Computers are constrained by the laws of physics; what you can do with information is no more than the laws of physics, when you operate at classical level. On a quantum computer, information processing is done on devices that obey the laws of quantum mechanics. These things can be very small and very cold, and they can be built out of exotic materials.”

The exotic material in D-Wave’s quantum chip is niobiumhttp://en.wikipedia.org/wiki/Niobium ; cool it enough and it becomes a superconductor. When ordinary metal conducts electricity, the electronshttp://en.wikipedia.org/wiki/Electron carrying the electric current collide with the imperfections in the metal and you get resistance. When you cool superconducting metal like niobium, the metal’s electrons form Cooper pairs where the motion of one electron is matched by an equal and opposite motion of the paired electron, which stops the electrons hitting the imperfections and generating resistance, which means the electrons flow freely without you needing to pump in extra current. When the Cooper pairs enter the Josephson junctions in the chip – made up of two segments of superconducting niobium linked by a weak insulating barrier – they break up, creating electron-like quasi-particles that can tunnel through the insulator in the junction, effectively conducting the current through the junction.

The niobium is arranged in rings through which the current can flow clockwise, anti-clockwise or in a mixture of both directions – corresponding, according to Rose, to the 0, 1 or superposition of the two values in the quantum bit of information (qubit) that quantum computing is based on. “The chip is a series of metal traces on a silicon substrate; the substrate is the same as you’d use for any semiconductor process but on top are layers of metal interrupted by insulators. This is an entirely metal based magnetic thing where all the information is stored in the direction of the current flows around the metal loop and interruptions.”

The direction of the current converts into a value for the qubit based on whether that qubit has a bias towards one direction (0 or 1), whether neighbouring qubits are running the in same or opposite directions and the energy barrier between the different qubit states. The current chip, Leda, has 28 rings, giving 28 qubits, but they’re not all interconnected to each other, only to a number of ‘neighbours’. The Cooper pair in the niobium are technically bosons so they all exist in the same quantum state, Rose claims, which gives the entire superconductor quantum properties even without interconnecting every qubit. Reducing the number of interconnections simplifies manufacturing and has enabled D-wave to go from 2 qubits in 2002 to 16 in 2007, 28 today – and 512 and 1024 over the next year, if things go well.

Many believe that true quantum computing will enable computations that supercomputers would take hundreds of years to process, enabling real-time weather prediction, custom drug design and cracking encryption. Geordie Rose isn’t promising those kinds of universal applications, at least not immediately.

“The breadth of applications is actually quite narrow. The machine can be thought of most profitably as an analogue computer. It’s not exactly an analogue computer, it’s something novel that has never existed before but conceptually you can think of it as a special purpose chip designed to do one thing well. Ultimately, quantum computers will turn into a lot more than that but when you do the first iteration of a technology, it helps to focus what it does. This particular chip, all it does is problems related to pattern matching. Other applications such as code breaking; this chip is disabled in a way that makes those things not possible to run on it. It is possible that in future we might expand – if this particular project succeeds financially – to include other type of processors that are able to harness nature in way that allows you to do these things. But those are long term things and certainly not our focus right now.”

At the Future in Review conference this year Rose showed an image matching program developed with Google image matching expert, Dr Hartmut Neven, that can distinguish photos of, say, the Taj Mahalhttp://en.wikipedia.org/wiki/Taj_Mahal from photos of Big Ben by comparing the image to a group of images already labelled as the Taj Mahal. The software looks for matching points of interest in the photos, which means solving hard maths problems that Orion is very good at, according to Rose. “Similarity matching between images is a very hard artificial intelligence problem and it turns out, with quantum computers, that their sweet spot is in the technical math that underlies certain hard vision problems and certain hard machine learning problems.”

You can match images and look for patterns on conventional computers, but it takes a lot of time it train the system, says Rose. “The requirement to do very fast search on a large number of images requires that you sacrifice quality. Often what happens in image search is that you can do very well on finding certain types of objects in images by spending a lot of time up front. You can detect faces in photos very quickly if you spend a year using an enormous amount of computing cycles to do that.”

Using Orion won’t necessarily speed up the time it takes to search, but he believes it will produce much better matches to what you’re looking for, and he’s not worried by performance that’s actually slower than conventional computing today. “This is not a demonstration of performance; this is a demonstration that we can do this end to end. We will be able to get a quality of matching on large data sets you simply can’t get with conventional computing, no matter how good your algorithms are. When you are searching for something complicated or unique it’s sometimes hard to describe. This is the first step of a system where you can query not with text but with images; it’s the sweet spot of the next generation of search and what these computers do very well.”

Pattern matching covers a wide range of applications. D-wave has previously demonstrated searching a database of molecules, creating a seating plan with many constraints on who can sit together and solving Sudoku puzzles and commercial applications are the next step. Rose talks about improving the logistics of how jet fuel is distributed and stored, cataloguing stars in images of space, modelling protein folding and counting the number of rocks in a possible landing area on Mars but also solving complex business problems: “What is the ideal business unit in my company to work on this project? I need three people who know C++ and earn less than such and such…”

But Orion isn’t anywhere near ready to go in your data center. It’s going to be staying in D-wave’s headquarters in Burnaby, Canada for the immediate future, because of what Rose calls the “extraordinary” cooling requirements. The Josephson junctions are only microns across; the chip they’re on is 5 millimetres square. But Orion itself is roughly the size of a large domestic refrigerator, and most of the system is taken up by the refrigeration equipment.

“This thing sits inside a shielded room, a big metal room which is almost a magnetic vacuum for certain frequencies of EM radiation. Inside is an insert which is half fridge and half filtering. We run this thing at ten milliKelvin, just 0.01 degrees above absolute zero – and just for a point of reference the temperature of interstellar space is about 2.7 Kelvin. The chip needs to sit in a magnetic vacuum. A lot of the gadgetry inside this is very, very robust filters that filter out every bit of noise you can with current technology, to get the signals on the lines coming in and the ambient magnetic field to very low levels – one nanotesla in three dimensions across the whole chip, which is at or beyond the state of the art for magnetic vacuum technology.”

Rather than moving Orion, D-Wave is developing remote access software. Writing quantum annealing algorithms for solving binary quadratic programs on Orion is very different from classical programming. You can work directly in the system’s machine language, directly choosing the current flowing on the input lines on the chip, but Rose expects that will only appeal to scientists studying the way quantum computing itself behave. Demos like the image matching system are written as problems in what he calls ‘industry standard ways to state combinatorial optimization problems’. A conventional computer converts that for Orion so that a solution corresponds to a pattern of current in the qubits that takes up the minimum amount of energy – the annealing.

“It’s like trying to find the lowest point in a valley when you have a ball and you let the ball go; it can find the lowest point by finding its natural state,” Rose explains. “It’s really easy to learn to use the system at this level, but figuring out how to recast the problem you really care about – say image matching – to use this new capability is very hard. Typically the folks who “get it” at this level have PhD-level discrete math backgrounds applied in an industry setting.”

Rose wants more people than that using Orion; “I’m a big fan of opening things up as much as possible to anyone wants to use them and making them easy to use even if people don’t understand quantum computation.” For the rest of us, D-Wave has produced a compiler that means programmers can state problems in SQL using a new FIND command. “This level of access allows anyone who is an expert database programmer to begin using the system within about 30 minutes,” claims Rose, “most of which is in learning the syntax of the FIND command, which is very similar to SELECT.” Developers who want to prototype applications or get familiar with the interface can try the programming model out with D-Wave’s Web service solver, although this currently runs a software emulation rather than sending commands to real hardware.

D-Wave’s demonstrations have generated plenty of controversy, partly because the company hasn’t published the kind of information that comes out of less commercial concerns in peer-reviewed journals. Critics suggest that what D-wave has is an analog computer that won’t achieve quantum performance; Geordie Rose believes time will show it’s a true quantum computer. “It’s known that there are several universal models of quantum computation, just like there are several universal classical models. The one we picked, adiabatic quantum computing, has significant advantage in that it’s easy to implement with large qubits.”

Using semiconductor manufacturing techniques means D-Wave can create a new version of the processor every month and keep tweaking it to fix any problems. And he’s confident there’s enough demand that we’ll see a usable quantum computer within years, not the decades some predict. “There’s a huge push from business and technology and that push is not going to go away any time. This is not like super high temperature conductors or fusion; this is something that’s going to be pushed until it works.”

Source:

http://www.tomshardware.com/reviews/super-cooled-quantum-computing,1976.html

The artificial intelligence powering the camera would also be able to respond to visual cues such as fights, or violent behaviour.

Scientists say the aim is to allow the camera to react just as a human might, hearing a scream and then swinging around to find the source with the same speed as a person, which is about 300 milliseconds.

Over time, the scientists claim the AI algorithms would learn, picking up key words and phrases it associates with criminal activity.

“The longer artificial intelligence is in the software the more it learns. Later versions will get cleverer as time goes on,” says Dr David Brown, director of the project.

Fuzzy thinking

Brown says the foundation of the technology is a new type of fuzzy logic: “In identifying sound we are looking for the shapes of sound. In the same way, if you close your eyes, you can trace the shape of a physical object and ‘read’ its profile with your hand we are developing shapes of sound so the software recognises them.

“The software will use an artificial intelligence template for the waveform of sound shapes and if the shape isn’t an exact fit, use fuzzy logic to determine what the sound is. For example, different types of glass will all have slightly different waveforms of sound when they smash but they will have the same generic shape which can be read using fuzzy logic.

“It’s a very fast, real-time method of identifying sounds.”

While there are clearly privacy implications inherent in the technology, Brown claims the AI will be trained “to only listen for specific words associated with violence, not full conversations.”

However, is this hardly likely to calm privacy advocates already concerned with the growing number of CCTV camera in the UK, and their potential uses.

At the moment the AI is in research stage, though scientists have been given a three-year grant by the Engineering and Physical Sciences Research Council (EPSRC) to further develop it.

The SpursEngine SE1000 will work in much the same way in the laptops.

The operating system will run on an Intel Core 2 Duo chip and the SpursEngine will be called on to handle processor-intensive tasks, such as processing of high-definition video. This arrangement means the laptop should be capable of some tricks that haven’t been seen on machines until now.

Among them, Toshiba said the two computers will be able to upscale standard-definition video to high definition; transcode in realtime digital TV to MPEG4 so that the resulting files are cut down in size by one-eighth and burn video to DVD in half the time of current machines.

A novel feature is face navigation. Faces that appear in video are recognized and displayed as thumbnail images to create a visual index to the video. Users can find the person or scene they want by glancing at the thumbnails and then click on the respective one to watch that portion of video. The computer can also divide up the scenes in user-shot video so they can be viewed one-by-one and analyze and display the volume or the clip across its entire length so, for example, excitement in a sports event can be more easily found.

Finally, by analyzing images from the computer’s built-in camera it’s possible to control video playback with hand gestures.

The Qosmio G50 is a multimedia laptop and has an 18.4-inch high-definition screen, 500G bytes of hard-disk space, NVidia GeForce 9600M graphics processor, dual digital TV tuners and wireless LAN including 802.11n. It weighs 4.9 kilograms and measures 45 centimeters by 31cms by 4.8cms. Battery life is about 4 hours.

The Qosmio G50 will be cost from ¥290,000 (US$2,700) and the F50, which has a 15-inch screen and 250G byte hard-disk drive, from ¥250,000. Toshiba plans to put the machines on sale overseas but has yet to announce launch details.

http://www.pcworld.com/businesscenter/article/147415/toshiba_unveils_laptop_with_cellderived_chip.html

The prefix “peta” stands for a million billion, also known as a quadrillion. For the Roadrunner supercomputer, operating at petaflop/s performance means the machine can process a million billion calculations each second. In other words, Roadrunner gives scientists the ability to quickly render mountainous problems into mere molehills, or model systems that previously were unthinkably complex.

Late last week and early this week while verifying Roadrunner’s performance, Los Alamos and IBM researchers used three different computational codes to test the machine. Among those codes was one dubbed “PetaVision” by its developers and the research team using it.

PetaVision models the human visual system—mimicking more than 1 billion visual neurons and trillions of synapses. Neurons are nerve cells that process information in the brain. Neurons communicate with each other using synaptic connections, analogous to what transistors are in modern computer chips. Synapses store memories and play a vital role in learning.

Synapses set the scale for computations performed by the brain while undertaking such tasks as locomotion, hearing or vision. Because there are about a quadrillion synapses in the human brain, human cognition is a petaflop/s computational problem.

To date, computers have been unable to match human performance on such visual tasks as flawlessly detecting an oncoming automobile on the highway or distinguishing a friend from a stranger in a crowd of people. Roadrunner is now changing the game.

On Saturday, Los Alamos researchers used PetaVision to model more than a billion visual neurons surpassing the scale of 1 quadrillion computations a second (a petaflop/s). On Monday scientists used PetaVision to reach a new computing performance record of 1.144 petaflop/s. The achievement throws open the door to eventually achieving human-like cognitive performance in electronic computers. PetaVision only requires single precision arithmetic, whereas the official LINPACK code used to officially verify Roadrunner’s speed uses double precision arithmetic.

“Roadrunner ushers in a new era for science at Los Alamos National Laboratory,” said Terry Wallace, associate director for Science, Technology and Engineering at Los Alamos. “Just a week after formal introduction of the machine to the world, we are already doing computational tasks that existed only in the realm of imagination a year ago.”

Based on the results of PetaVision’s inaugural trials, Los Alamos researchers believe they can study in real time the entire human visual cortex—arguably a human being’s most important sensory apparatus.

The ability to achieve human levels of cognitive performance on a digital computer could lead to important insights and revolutionary technological applications. Such applications include “smart” cameras that can recognize danger or an autopilot system for automobiles that could take over for incapacitated drivers in complex situations such as navigating dense urban traffic.

Los Alamos National Laboratory’s computation science team working with Roadrunner includes: Craig Rasmussen, Charles Ferenbaugh, Sriram Swaminarayan, Pallab Datta, all of Los Alamos; and Cornell Wright of IBM.

The PetaVision Synthetic Cognition team responsible for the theory and codes run on Roadrunner includes: Luis Bettencourt, Garrett Kenyon, Ilya Nemenman, John George, Steven Brumby, Kevin Sanbonmatsu, and John Galbraith, all of Los Alamos; Steven Zuker of Yale University; and James DiCarlo from Massachusetts Institute of Technology.

The Roadrunner is the world’s first supercomputer to achieve sustained operating performance speeds of one petaflop/s. In partnership with Los Alamos and the National Nuclear Security Administration, Roadrunner was built by IBM and will be housed at Los Alamos National Laboratory, where it will be used to perform calculations that will vastly improve the nation’s ability to certify that the United States nuclear weapons stockpile is reliable without conducting underground nuclear tests. Roadrunner also will be used for science and engineering such as energy research, understanding dark energy and dark matter, materials properties and response, understanding complex neural and biological systems, and biomedical applications.

Roadrunner was built using commercially available hardware, including aspects of commercial game console technologies. Roadrunner has a unique hybrid design comprised of nodes containing two AMD OpteronTM dual-core processors plus four PowerXCell 8iTM processors used as computational accelerators. The accelerators are a special IBM-developed variant of the Cell processors used in the Sony PlayStation® 3. Roadrunner uses a Linux operating system. The project’s total cost is approximately $120 million.

Los Alamos National Laboratory is a multidisciplinary research institution engaged in strategic science on behalf of national security. The Laboratory is operated by a team composed of Bechtel National, the University of California, BWX Technologies, and Washington Group International for the Department of Energy’s National Nuclear Security Administration.

Los Alamos enhances national security by ensuring the safety and reliability of the U.S. nuclear stockpile, developing technologies to reduce threats from weapons of mass destruction, and solving problems related to energy, environment, infrastructure, health and global security concerns.

Source:

http://www.lanl.gov/news/index.php/fuseaction/home.story/story_id/13602

Developed to power gaming systems, the Cell chip has also garnered interest from the supercomputing community owing to its ability to handle large amounts of floating point calculations.

IBM hopes that the chips, which currently power climate modelling and other traditional supercomputing tasks, will also appeal to customers ranging from financial analysis firms to animation studios.

To further spark interest in the new systems, IBM has posted an SDK for developers looking to optimise code for the Cell processor.

IBM sees the high-performance systems eventually replacing large data centres with smaller virtualised systems.

“The QS22 is a technological leap over the physical limitations of traditional processors,” said Jim Comfort, vice president of IBM’s Systems & Technology Group.

IBM plans to release the QS22 BladeCenter server next month.

Source:

http://www.itnews.com.au/News/76055,ibm-touts-supercomputers-for-the-enterprise.aspx