Tag Archives: energy

#437620 The Trillion-Transistor Chip That Just ...

The history of computer chips is a thrilling tale of extreme miniaturization.

The smaller, the better is a trend that’s given birth to the digital world as we know it. So, why on earth would you want to reverse course and make chips a lot bigger? Well, while there’s no particularly good reason to have a chip the size of an iPad in an iPad, such a chip may prove to be genius for more specific uses, like artificial intelligence or simulations of the physical world.

At least, that’s what Cerebras, the maker of the biggest computer chip in the world, is hoping.

The Cerebras Wafer-Scale Engine is massive any way you slice it. The chip is 8.5 inches to a side and houses 1.2 trillion transistors. The next biggest chip, NVIDIA’s A100 GPU, measures an inch to a side and has a mere 54 billion transistors. The former is new, largely untested and, so far, one-of-a-kind. The latter is well-loved, mass-produced, and has taken over the world of AI and supercomputing in the last decade.

So can Goliath flip the script on David? Cerebras is on a mission to find out.

Big Chips Beyond AI
When Cerebras first came out of stealth last year, the company said it could significantly speed up the training of deep learning models.

Since then, the WSE has made its way into a handful of supercomputing labs, where the company’s customers are putting it through its paces. One of those labs, the National Energy Technology Laboratory, is looking to see what it can do beyond AI.

So, in a recent trial, researchers pitted the chip—which is housed in an all-in-one system about the size of a dorm room mini-fridge called the CS-1—against a supercomputer in a fluid dynamics simulation. Simulating the movement of fluids is a common supercomputer application useful for solving complex problems like weather forecasting and airplane wing design.

The trial was described in a preprint paper written by a team led by Cerebras’s Michael James and NETL’s Dirk Van Essendelft and presented at the supercomputing conference SC20 this week. The team said the CS-1 completed a simulation of combustion in a power plant roughly 200 times faster than it took the Joule 2.0 supercomputer to do a similar task.

The CS-1 was actually faster-than-real-time. As Cerebrus wrote in a blog post, “It can tell you what is going to happen in the future faster than the laws of physics produce the same result.”

The researchers said the CS-1’s performance couldn’t be matched by any number of CPUs and GPUs. And CEO and cofounder Andrew Feldman told VentureBeat that would be true “no matter how large the supercomputer is.” At a point, scaling a supercomputer like Joule no longer produces better results in this kind of problem. That’s why Joule’s simulation speed peaked at 16,384 cores, a fraction of its total 86,400 cores.

A comparison of the two machines drives the point home. Joule is the 81st fastest supercomputer in the world, takes up dozens of server racks, consumes up to 450 kilowatts of power, and required tens of millions of dollars to build. The CS-1, by comparison, fits in a third of a server rack, consumes 20 kilowatts of power, and sells for a few million dollars.

While the task is niche (but useful) and the problem well-suited to the CS-1, it’s still a pretty stunning result. So how’d they pull it off? It’s all in the design.

Cut the Commute
Computer chips begin life on a big piece of silicon called a wafer. Multiple chips are etched onto the same wafer and then the wafer is cut into individual chips. While the WSE is also etched onto a silicon wafer, the wafer is left intact as a single, operating unit. This wafer-scale chip contains almost 400,000 processing cores. Each core is connected to its own dedicated memory and its four neighboring cores.

Putting that many cores on a single chip and giving them their own memory is why the WSE is bigger; it’s also why, in this case, it’s better.

Most large-scale computing tasks depend on massively parallel processing. Researchers distribute the task among hundreds or thousands of chips. The chips need to work in concert, so they’re in constant communication, shuttling information back and forth. A similar process takes place within each chip, as information moves between processor cores, which are doing the calculations, and shared memory to store the results.

It’s a little like an old-timey company that does all its business on paper.

The company uses couriers to send and collect documents from other branches and archives across town. The couriers know the best routes through the city, but the trips take some minimum amount of time determined by the distance between the branches and archives, the courier’s top speed, and how many other couriers are on the road. In short, distance and traffic slow things down.

Now, imagine the company builds a brand new gleaming skyscraper. Every branch is moved into the new building and every worker gets a small filing cabinet in their office to store documents. Now any document they need can be stored and retrieved in the time it takes to step across the office or down the hall to their neighbor’s office. The information commute has all but disappeared. Everything’s in the same house.

Cerebras’s megachip is a bit like that skyscraper. The way it shuttles information—aided further by its specially tailored compiling software—is far more efficient compared to a traditional supercomputer that needs to network a ton of traditional chips.

Simulating the World as It Unfolds
It’s worth noting the chip can only handle problems small enough to fit on the wafer. But such problems may have quite practical applications because of the machine’s ability to do high-fidelity simulation in real-time. The authors note, for example, the machine should in theory be able to accurately simulate the air flow around a helicopter trying to land on a flight deck and semi-automate the process—something not possible with traditional chips.

Another opportunity, they note, would be to use a simulation as input to train a neural network also residing on the chip. In an intriguing and related example, a Caltech machine learning technique recently proved to be 1,000 times faster at solving the same kind of partial differential equations at play here to simulate fluid dynamics.

They also note that improvements in the chip (and others like it, should they arrive) will push back the limits of what can be accomplished. Already, Cerebras has teased the release of its next-generation chip, which will have 2.6 trillion transistors, 850,00 cores, and more than double the memory.

Of course, it still remains to be seen whether wafer-scale computing really takes off. The idea has been around for decades, but Cerebras is the first to pursue it seriously. Clearly, they believe they’ve solved the problem in a way that’s useful and economical.

Other new architectures are also being pursued in the lab. Memristor-based neuromorphic chips, for example, mimic the brain by putting processing and memory into individual transistor-like components. And of course, quantum computers are in a separate lane, but tackle similar problems.

It could be that one of these technologies eventually rises to rule them all. Or, and this seems just as likely, computing may splinter into a bizarre quilt of radical chips, all stitched together to make the most of each depending on the situation.

Image credit: Cerebras Continue reading

Posted in Human Robots

#437608 Video Friday: Agility Robotics Raises ...

Video Friday is your weekly selection of awesome robotics videos, collected by your Automaton bloggers. We’ll also be posting a weekly calendar of upcoming robotics events for the next few months; here’s what we have so far (send us your events!):

IROS 2020 – October 25-29, 2020 – [Online]
ROS World 2020 – November 12, 2020 – [Online]
CYBATHLON 2020 – November 13-14, 2020 – [Online]
ICSR 2020 – November 14-16, 2020 – Golden, Colo., USA
Let us know if you have suggestions for next week, and enjoy today’s videos.

Digit is now in full commercial production and we’re excited to announce a $20M funding rounding round co-led by DCVC and Playground Global!

Digits for everyone!

[ Agility Robotics ]

A flexible rover that has both ability to travel long distances and rappel down hard-to-reach areas of scientific interest has undergone a field test in the Mojave Desert in California to showcase its versatility. Composed of two Axel robots, DuAxel is designed to explore crater walls, pits, scarps, vents and other extreme terrain on the moon, Mars and beyond.

This technology demonstration developed at NASA’s Jet Propulsion Laboratory in Southern California showcases the robot’s ability to split in two and send one of its halves — a two-wheeled Axle robot — over an otherwise inaccessible slope, using a tether as support and to supply power.

The rappelling Axel can then autonomously seek out areas to study, safely overcome slopes and rocky obstacles, and then return to dock with its other half before driving to another destination. Although the rover doesn’t yet have a mission, key technologies are being developed that might, one day, help us explore the rocky planets and moons throughout the solar system.

[ JPL ]

A rectangular robot as tiny as a few human hairs can travel throughout a colon by doing back flips, Purdue University engineers have demonstrated in live animal models. Why the back flips? Because the goal is to use these robots to transport drugs in humans, whose colons and other organs have rough terrain. Side flips work, too. Why a back-flipping robot to transport drugs? Getting a drug directly to its target site could remove side effects, such as hair loss or stomach bleeding, that the drug may otherwise cause by interacting with other organs along the way.

[ Purdue ]

This video shows the latest results in the whole-body locomotion control of the humanoid robot iCub achieved by the Dynamic Interaction Control line at IIT-Istituto Italiano di Tecnologia in Genova (Italy). In particular, the iCub now keeps the balance while walking and receiving pushes from an external user. The implemented control algorithms also ensure the robot to remain compliant during locomotion and human-robot interaction, a fundamental property to lower the possibility to harm humans that share the robot surrounding environment.

This is super impressive, considering that iCub was only able to crawl and was still tethered not too long ago. Also, it seems to be blinking properly now, so it doesn’t look like it’s always sleepy.

[ IIT ]

This video shows a set of new tests we performed on Bolt. We conducted tests on 5 different scenarios, 1) walking forward/backward 2) uneven surface 3) soft surface 4) push recovery 5) slippage recovery. Thanks to our feedback control based on Model Predictive Control, the robot can perform walking in the presence of all these uncertainties. We will open-source all the codes in a near future.

[ ODRI ]

The title of this video is “Can you throw your robot into a lake?” The title of this video should be, “Can you throw your robot into a lake and drive it out again?”

[ Norlab ]

AeroVironment Successfully Completes Sunglider Solar HAPS Stratospheric Test Flight, Surpassing 60,000 Feet Altitude and Demonstrating Broadband Mobile Connectivity.

[ AeroVironment ]

We present CoVR, a novel robotic interface providing strong kinesthetic feedback (100 N) in a room-scale VR arena. It consists of a physical column mounted on a 2D Cartesian ceiling robot (XY displacements) with the capacity of (1) resisting to body-scaled users actions such as pushing or leaning; (2) acting on the users by pulling or transporting them as well as (3) carrying multiple potentially heavy objects (up to 80kg) that users can freely manipulate or make interact with each other.

[ DeepAI ]

In a new video, personnel from Swiss energy supply company Kraftwerke Oberhasli AG (KWO) explain how they were able to keep employees out of harm’s way by using Flyability’s Elios 2 to collect visual data while building a new dam.

[ Flyability ]

Enjoy our Ascento robot fail compilation! With every failure we experience, we learn more and we can improve our robot for its next iteration, which will come soon… Stay tuned for more!

FYI posting a robot fails video will pretty much guarantee you a spot in Video Friday!

[ Ascento ]

Humans are remarkably good at using chopsticks. The Guinness World Record witnessed a person using chopsticks to pick up 65 M&Ms in just a minute. We aim to collect demonstrations from humans and to teach robot to use chopsticks.

[ UW Personal Robotics Lab ]

A surprising amount of personality from these Yaskawa assembly robots.

[ Yaskawa ]

This paper presents the system design, modeling, and control of the Aerial Robotic Chain Manipulator. This new robot design offers the potential to exert strong forces and moments to the environment, carry and lift significant payloads, and simultaneously navigate through narrow corridors. The presented experimental studies include a valve rotation task, a pick-and-release task, and the verification of load oscillation suppression to demonstrate the stability and performance of the system.

[ ARL ]

Whether animals or plants, whether in the water, on land or in the air, nature provides the model for many technical innovations and inventions. This is summed up in the term bionics, which is a combination of the words ‘biology‘ and ‘electronics’. At Festo, learning from nature has a long history, as our Bionic Learning Network is based on using nature as the source for future technologies like robots, assistance systems or drive solutions.

[ Festo ]

Dogs! Selfies! Thousands of LEGO bricks! This video has it all.

[ LEGO ]

An IROS workshop talk on “Cassie and Mini Cheetah Autonomy” by Maani Ghaffari and Jessy Grizzle from the University of Michigan.

[ Michigan Robotics ]

David Schaefer’s Cozmo robots are back with this mind-blowing dance-off!

What you just saw represents hundreds of hours of work, David tells us: “I wrote over 10,000 lines of code to create the dance performance as I had to translate the beats per minute of the song into motor rotations in order to get the right precision needed to make the moves look sharp. The most challenging move was the SpongeBob SquareDance as any misstep would send the Cozmos crashing into each other. LOL! Fortunately for me, Cozmo robots are pretty resilient.”

[ Life with Cozmo ]

Thanks David!

This week’s GRASP on Robotics seminar is by Sangbae Kim from MIT, on “Robots with Physical Intelligence.”

While industrial robots are effective in repetitive, precise kinematic tasks in factories, the design and control of these robots are not suited for physically interactive performance that humans do easily. These tasks require ‘physical intelligence’ through complex dynamic interactions with environments whereas conventional robots are designed primarily for position control. In order to develop a robot with ‘physical intelligence’, we first need a new type of machines that allow dynamic interactions. This talk will discuss how the new design paradigm allows dynamic interactive tasks. As an embodiment of such a robot design paradigm, the latest version of the MIT Cheetah robots and force-feedback teleoperation arms will be presented.

[ GRASP ]

This week’s CMU Ri Seminar is by Kevin Lynch from Northwestern, on “Robotics and Biosystems.”

Research at the Center for Robotics and Biosystems at Northwestern University encompasses bio-inspiration, neuromechanics, human-machine systems, and swarm robotics, among other topics. In this talk I will give an overview of some of our recent work on in-hand manipulation, robot locomotion on yielding ground, and human-robot systems.

[ CMU RI ] Continue reading

Posted in Human Robots

#437471 How Giving Robots a Hybrid, Human-Like ...

Squeezing a lot of computing power into robots without using up too much space or energy is a constant battle for their designers. But a new approach that mimics the structure of the human brain could provide a workaround.

The capabilities of most of today’s mobile robots are fairly rudimentary, but giving them the smarts to do their jobs is still a serious challenge. Controlling a body in a dynamic environment takes a surprising amount of processing power, which requires both real estate for chips and considerable amounts of energy to power them.

As robots get more complex and capable, those demands are only going to increase. Today’s most powerful AI systems run in massive data centers across far more chips than can realistically fit inside a machine on the move. And the slow death of Moore’s Law suggests we can’t rely on conventional processors getting significantly more efficient or compact anytime soon.

That prompted a team from the University of Southern California to resurrect an idea from more than 40 years ago: mimicking the human brain’s division of labor between two complimentary structures. While the cerebrum is responsible for higher cognitive functions like vision, hearing, and thinking, the cerebellum integrates sensory data and governs movement, balance, and posture.

When the idea was first proposed the technology didn’t exist to make it a reality, but in a paper recently published in Science Robotics, the researchers describe a hybrid system that combines analog circuits that control motion and digital circuits that govern perception and decision-making in an inverted pendulum robot.

“Through this cooperation of the cerebrum and the cerebellum, the robot can conduct multiple tasks simultaneously with a much shorter latency and lower power consumption,” write the researchers.

The type of robot the researchers were experimenting with looks essentially like a pole balancing on a pair of wheels. They have a broad range of applications, from hoverboards to warehouse logistics—Boston Dynamics’ recently-unveiled Handle robot operates on the same principles. Keeping them stable is notoriously tough, but the new approach managed to significantly improve all digital control approaches by radically improving the speed and efficiency of computations.

Key to bringing the idea alive was the recent emergence of memristors—electrical components whose resistance relies on previous input, which allows them to combine computing and memory in one place in a way similar to how biological neurons operate.

The researchers used memristors to build an analog circuit that runs an algorithm responsible for integrating data from the robot’s accelerometer and gyroscope, which is crucial for detecting the angle and velocity of its body, and another that controls its motion. One key advantage of this setup is that the signals from the sensors are analog, so it does away with the need for extra circuitry to convert them into digital signals, saving both space and power.

More importantly, though, the analog system is an order of magnitude faster and more energy-efficient than a standard all-digital system, the authors report. This not only lets them slash the power requirements, but also lets them cut the processing loop from 3,000 microseconds to just 6. That significantly improves the robot’s stability, with it taking just one second to settle into a steady state compared to more than three seconds using the digital-only platform.

At the minute this is just a proof of concept. The robot the researchers have built is small and rudimentary, and the algorithms being run on the analog circuit are fairly basic. But the principle is a promising one, and there is currently a huge amount of R&D going into neuromorphic and memristor-based analog computing hardware.

As often turns out to be the case, it seems like we can’t go too far wrong by mimicking the best model of computation we have found so far: our own brains.

Image Credit: Photos Hobby / Unsplash Continue reading

Posted in Human Robots

#437407 Nvidia’s Arm Acquisition Brings the ...

Artificial intelligence and mobile computing have been two of the most disruptive technologies of this century. The unification of the two companies that made them possible could have wide-ranging consequences for the future of computing.

California-based Nvidia’s graphics processing units (GPUs) have powered the deep learning revolution ever since Google researchers discovered in 2011 that they could run neural networks far more efficiently than conventional CPUs. UK company Arm’s energy-efficient chip designs have dominated the mobile and embedded computing markets for even longer.

Now the two will join forces after the American company announced a $40 billion deal to buy Arm from its Japanese owner, Softbank. In a press release announcing the deal, Nvidia touted its potential to rapidly expand the reach of AI into all areas of our lives.

“In the years ahead, trillions of computers running AI will create a new internet-of-things that is thousands of times larger than today’s internet-of-people,” said Nvidia founder and CEO Jensen Huang. “Uniting NVIDIA’s AI computing capabilities with the vast ecosystem of Arm’s CPU, we can advance computing from the cloud, smartphones, PCs, self-driving cars and robotics, to edge IoT, and expand AI computing to every corner of the globe.”

There are good reasons to believe the hype. The two companies are absolutely dominant in their respective fields—Nvidia’s GPUs support more than 97 percent of AI computing infrastructure offered by big cloud service providers, and Arm’s chips power more than 90 percent of smartphones. And there’s little overlap in their competencies, which means the relationship could be a truly symbiotic one.

“I think the deal “fits like a glove” in that Arm plays in areas that Nvidia does not or isn’t that successful, while NVIDIA plays in many places Arm doesn’t or isn’t that successful,” analyst Patrick Moorhead wrote in Forbes.

One of the most obvious directions would be to expand Nvidia’s AI capabilities to the kind of low-power edge devices that Arm excels in. There’s growing demand for AI in devices like smartphones, wearables, cars, and drones, where transmitting data to the cloud for processing is undesirable either for reasons of privacy or speed.

But there might also be fruitful exchanges in the other direction. Huang told Moorhead a major focus would be bringing Arm’s expertise in energy efficiency to the data center. That’s a big concern for technology companies whose electricity bills and green credentials are taking a battering thanks to the huge amounts of energy required to run millions of computer chips around the clock.

The deal may not be plain sailing, though, most notably due to the two companies’ differing business models. While Nvidia sells ready-made processors, Arm simply creates chip designs and then licenses them to other companies who can then customize them to their particular hardware needs. It operates on an open-licence basis whereby any company with the necessary cash can access its designs.

As a result, its designs are found in products built by hundreds of companies that license its innovations, including Apple, Samsung, Huawei, Qualcomm, and even Nvidia. Some, including two of the company’s co-founders, have raised concerns that the purchase by Nvidia, which competes with many of these other companies, could harm the neutrality that has been central to its success.

It’s possible this could push more companies towards RISC-V, an open-source technology developed by researchers at the University of California at Berkeley that rivals Arm’s and is not owned by any one company. However, there are plenty of reasons why most companies still prefer arm over the less feature-rich open-source option, and it might take a considerable push to convince Arm’s customers to jump ship.

The deal will also have to navigate some thorny political issues. Unions, politicians, and business leaders in the UK have voiced concerns that it could lead to the loss of high-tech jobs, and government sources have suggested conditions could be placed on the deal.

Regulators in other countries could also put a spanner in the works. China is concerned that if Arm becomes US-owned, many of the Chinese companies that rely on its technology could become victims of export restrictions as the China-US trade war drags on. South Korea is also wary that the deal could create a new technology juggernaut that could dent Samsung’s growth in similar areas.

Nvidia has made commitments to keep Arm’s headquarters in the UK, which it says should lessen concerns around jobs and export restrictions. It’s also pledged to open a new world-class technology center in Cambridge and build a state-of-the-art AI supercomputer powered by Arm’s chips there. Whether the deal goes through still hangs in the balance, but of it does it could spur a whole new wave of AI innovation.

Image Credit: Nvidia Continue reading

Posted in Human Robots

#437395 Microsoft Had a Crazy Idea to Put ...

A little over two years ago, a shipping container-sized cylinder bearing Microsoft’s name and logo was lowered onto the ocean floor off the northern coast of Scotland. Inside were 864 servers, and their submersion was part of the second phase of the software giant’s Project Natick. Launched in 2015, the project’s purpose is to determine the feasibility of underwater data centers powered by offshore renewable energy.

A couple months ago, the deep-sea servers were brought back up to the surface so engineers could inspect them and evaluate how they’d performed while under water.

But wait—why were they there in the first place?

As bizarre as it seems to sink hundreds of servers into the ocean, there are actually several very good reasons to do so. According to the UN, about 40 percent of the world’s population lives within 60 miles of an ocean. As internet connectivity expands to cover most of the globe in the next few years, millions more people will come online, and a lot more servers will be needed to manage the increased demand and data they’ll generate.

In densely-populated cities real estate is expensive and can be hard to find. But know where there’s lots of cheap, empty space? At the bottom of the ocean. This locale also carries the added benefit of being really cold (depending where we’re talking, that is; if you’re looking off the coast of, say, Mumbai or Abu Dhabi, the waters are warmer).

Servers generate a lot of heat, and datacenters use most of their electricity for cooling. Keeping not just the temperature but also the humidity level constant is important for optimal functioning of the servers; neither of these vary much 100 feet under water.

Finally, installing data centers on the ocean floor is, surprisingly, much faster than building them on land. Microsoft claims its server-holding cylinders will take less than 90 days to go from factory ship to operation, as compared to the average two years it takes to get a terrestrial data center up and running.

Microsoft’s Special Projects team operated the underwater data center for two years, and it took a full day to dredge it up and bring it to the surface. One of the first things researchers did was to insert test tubes into the container to take samples of the air inside; they’ll use it to try to determine how gases released from the equipment may have impacted the servers’ operating environment.

The container was filled with dry nitrogen upon deployment, which seems to have made for a much better environment than the oxygen that land-bound servers are normally surrounded by; the failure rate of the servers in the water was just one-eighth that of Microsoft’s typical rate for its servers on land. The team thinks the nitrogen atmosphere was helpful because it’s less corrosive than oxygen. The fact that no humans entered the container for the entirety of its operations helped, too (no moving around of components or having to turn on lights or adjust the temperature).

Ben Cutler, a project manager in Microsoft’s Special Projects research group who leads Project Natick, believes the results of this phase of the project are sufficient to show that underwater data centers are worth pursuing. “We are now at the point of trying to harness what we have done as opposed to feeling the need to go and prove out some more,” he said.

Cutler envisions putting underwater datacenters near offshore wind farms to power them sustainably. The data centers of the future will require less human involvement, instead being managed and run primarily by technologies like robotics and AI. In this kind of “lights-out” datacenter, the servers would be swapped out about once every five years, with any that fail before then being taken offline.

The final step in this phase of Project Natick is to recycle all the components used for the underwater data center, including the steel pressure vessel, heat exchangers, and the servers themselves—and restoring the sea bed where the cylinder rested back to its original condition.

If Cutler’s optimism is a portent of things to come, it may not be long before the ocean floor is dotted with sustainable datacenters to feed our ever-increasing reliance on our phones and the internet.

Image Credit: Microsoft Continue reading

Posted in Human Robots