Author Archives: Android
University of Wollongong (UOW) researchers have mimicked the supercoiling properties of DNA to develop a new type of artificial muscle for use in miniature robot applications. Their research is published today in Science Robotics. Continue reading
In order to effectively navigate real-world environments, legged robots should be able to move swiftly and freely while maintaining their balance. This is particularly true for humanoid robots, robots with two legs and a human-like body structure. Continue reading
A team of researchers affiliated with Seoul National University, Harvard University and Hankook Tire and Technology Co. Ltd., has developed a tire based on an origami design that allows for changing the shape of a tire while a vehicle is moving. In their paper published in the journal Science Robotics, the group describes their new tire design and how well it worked when tested. Continue reading
The world’s biggest AI chip just doubled its specs—without adding an inch.
The Cerebras Systems Wafer Scale Engine is about the size of a big dinner plate. All that surface area enables a lot more of everything, from processors to memory. The first WSE chip, released in 2019, had an incredible 1.2 trillion transistors and 400,000 processing cores. Its successor doubles everything, except its physical size.
The WSE-2 crams in 2.6 trillion transistors and 850,000 cores on the same dinner plate. Its on-chip memory has increased from 18 gigabytes to 40 gigabytes, and the rate it shuttles information to and from said memory has gone from 9 petabytes per second to 20 petabytes per second.
It’s a beast any way you slice it.
The WSE-2 is manufactured by Taiwan Semiconductor Manufacturing Company (TSMC), and it was a jump from TSMC’s 16-nanometer chipmaking process to its 7-nanometer process—skipping the 10-nanometer node—that enabled most of the WSE-2’s gains.
This required changes to the physical design of the chip, but Cerebras says they also made improvements to each core above and beyond what was needed to make the new process work. The updated mega-chip should be a lot faster and more efficient.
Why Make Giant Computer Chips?
While graphics processing units (GPUs) still reign supreme in artificial intelligence, they weren’t made for AI in particular. Rather, GPUs were first developed and used for graphics-heavy applications like gaming.
They’ve done amazing things for AI and supercomputing, but in the last several years, specialized chips made for AI are on the up and up.
Cerebras is one of the contenders, alongside other up-and-comers like Graphcore and SambaNova and more familiar names like Intel and NVIDIA.
The company likes to compare the WSE-2 to a top AI processor (NVIDIA’s A100) to underscore just how different it is from the competition. The A100 has two percent the number of transistors (54.2 billion) occupying a little under two percent the surface area. It’s much smaller, but the A100’s might is more fully realized when hundreds or thousands of chips are linked together in a larger system.
In contrast, the WSE-2 reduces the cost and complexity of linking all those chips together by jamming as much processing and memory as possible onto a single wafer of silicon. At the same time, removing the need to move data between lots of chips spread out on multiple server racks dramatically increases speed and efficiency.
The chip’s design gives its small, speedy cores their own dedicated memory and facilitates quick communication between cores. And Cerebras’s compiling software works with machine learning models using standard frameworks—like PyTorch and TensorFlow—to make distributing tasks among the chip’s cores fairly painless.
The approach is called wafer-scale computing because the chip is the size of a standard silicon wafer from which many chips are normally cut. Wafer-scale computing has been on the radar for years, but Cerebras is the first to make a commercially viable chip.
The chip comes packaged in a computer system called the CS-2. The system includes cooling and power supply and fits in about a third of a standard server rack.
After the startup announced the chip in 2019, it began working with a growing list of customers. Cerebras counts GlaxoSmithKline, Lawrence Livermore National Lab, and Argonne National (among others) as customers alongside unnamed partners in pharmaceuticals, biotech, manufacturing, and the military. Many applications have been in AI, but not all. Last year, the company said the National Energy Technology Laboratory (NETL) used the chip to outpace a supercomputer in a simulation of fluid dynamics.
Will Wafer-Scale Go Big?
Whether wafer-scale computing catches on remains to be seen.
Cerebras says their chip significantly speeds up machine learning tasks, and testimony from early customers—some of which claim pretty big gains—supports this. But there aren’t yet independent head-to-head comparisons. Neither Cerebras nor most other AI hardware startups, for example, took part in a recent MLperf benchmark test of AI systems. (The top systems nearly all used NVIDIA GPUs to accelerate their algorithms.)
According to IEEE Spectrum, Cerebras says they’d rather let interested buyers test the system on their own specific neural networks as opposed to selling them on a more general and potentially less applicable benchmark. This isn’t an uncommon approach AI analyst Karl Freund said, “Everybody runs their own models that they developed for their own business. That’s the only thing that matters to buyers.”
It’s also worth noting the WSE can only handle tasks small enough to fit on its chip. The company says most suitable problems it’s encountered can fit, and the WSE-2 delivers even more space. Still, the size of machine learning algorithms is growing rapidly. Which is perhaps why Cerebras is keen to note that two or even three CS-2’s can fit into a server cabinet.
Ultimately, the WSE-2 doesn’t make sense for smaller tasks in which one or a few GPUs will do the trick. At the moment the chip is being used in large, compute-heavy projects in science and research. Current applications include cancer research and drug discovery, gravity wave detection, and fusion simulation. Cerebras CEO and cofounder Andrew Feldman says it may also be made available to customers with shorter-term, less intensive needs on the cloud.
The market for the chip is niche, but Feldman told HPC Wire it’s bigger than he anticipated in 2015, and it’s still growing as new approaches to AI are continually popping up. “The market is moving unbelievably quickly,” he said.
The increasing competition between AI chips is worth watching. There may end up being several fit-to-purpose approaches or one that rises to the top.
For the moment, at least, it appears there’s some appetite for a generous helping of giant computer chips.
Image Credit: Cerebras Continue reading
Building a computer that can support artificial intelligence at the scale and complexity of the human brain will be a colossal engineering effort. Now researchers at the National Institute of Standards and Technology have outlined how they think we’ll get there.
How, when, and whether we’ll ever create machines that can match our cognitive capabilities is a topic of heated debate among both computer scientists and philosophers. One of the most contentious questions is the extent to which the solution needs to mirror our best example of intelligence so far: the human brain.
Rapid advances in AI powered by deep neural networks—which despite their name operate very differently than the brain—have convinced many that we may be able to achieve “artificial general intelligence” without mimicking the brain’s hardware or software.
Others think we’re still missing fundamental aspects of how intelligence works, and that the best way to fill the gaps is to borrow from nature. For many that means building “neuromorphic” hardware that more closely mimics the architecture and operation of biological brains.
The problem is that the existing computer technology we have at our disposal looks very different from biological information processing systems, and operates on completely different principles. For a start, modern computers are digital and neurons are analog. And although both rely on electrical signals, they come in very different flavors, and the brain also uses a host of chemical signals to carry out processing.
Now though, researchers at NIST think they’ve found a way to combine existing technologies in a way that could mimic the core attributes of the brain. Using their approach, they outline a blueprint for a “neuromorphic supercomputer” that could not only match, but surpass the physical limits of biological systems.
The key to their approach, outlined in Applied Physics Letters, is a combination of electronics and optical technologies. The logic is that electronics are great at computing, while optical systems can transmit information at the speed of light, so combining them is probably the best way to mimic the brain’s excellent computing and communication capabilities.
It’s not a new idea, but so far getting our best electronic and optical hardware to gel has proven incredibly tough. The team thinks they’ve found a potential workaround, dropping the temperature of the system to negative 450 degrees Fahrenheit.
While that might seem to only complicate matters, it actually opens up a host of new hardware possibilities. There are a bunch of high-performance electronic and optical components that only work at these frigid temperatures, like superconducting electronics, single-photon detectors, and silicon LEDs.
The researchers propose using these components to build artificial neurons that operate more like their biological cousins than conventional computer components, firing off electrical impulses, or spikes, rather than shuttling numbers around.
Each neuron has thousands of artificial synapses made from single photon detectors, which pick up optical messages from other neurons. These incoming signals are combined and processed by superconducting circuits, and once they cross a certain threshold a silicon LED is activated, sending an optical impulse to all downstream neurons.
The researchers envisage combining millions of these neurons on 300-millimeter silicon wafers and then stacking the wafers to create a highly interconnected network that mimics the architecture of the brain, with short-range connections dealt with by optical waveguides on each chip and long-range ones dealt with by fiber optic cables.
They acknowledge that the need to cryogenically cool the entire device is a challenge. But they say the improved power efficiency and that of their design should cancel out the cost of this cooling, and a system on the scale of the human brain should require no more power or space than a modern supercomputer. They also point out that there is significant R&D going into cryogenically-cooled quantum computers, which they could likely piggyback off of.
Some of the basic components of the system have already been experimentally demonstrated by the researchers, though they admit there’s still a long way to go to put all the pieces together. While many of these components are compatible with standard electronics fabrication, finding ways to manufacture them cheaply and integrate them will be a mammoth task.
Perhaps more important is the question of what kind of software the machine would run. It’s designed to implement “spiking neural networks” similar to those found in the brain, but our understanding of biological neural networks is still rudimentary, and our ability to mimic them is even worse. While both scientists and tech companies have been experimenting with the approach, it is still far less capable than deep learning.
Given the enormous engineering challenge involved in building a device of this scale, it may be a while before this blueprint makes it off the drawing board. But the proposal is an intriguing new chapter in the hunt for artificial general intelligence.
Image Credit: InspiredImages from Pixabay Continue reading