supercomputer

28 Articles

The 13.5 Million Core Computer

November 23, 2022 by 55 Comments

Having a dual- or quad-core CPU is not very exotic these days and CPUs with 12 or even 16 cores aren’t that rare. The Andromeda from Cerebras is a supercomputer with 13.5 million cores. The company claims it is one of the largest AI supercomputers ever built (but not the largest) and can perform 120 Petaflops of “dense compute.”

We aren’t sure about the methodology, but they also claim more than one exaflop of “AI computing.” The computer has a fabric backplane that can handle 96.8 terabits per second between nodes. According to a post on Extreme Tech, the core technology is a 3-plane wafer processor, WSE-2. One plane is for communications, one holds 40 GB of static RAM, and the math plane has 850,000 independent cores and 3.4 million floating point units.

The data is sent to the cores and collected by a bank of 64-core AMD EPYC 3 processors. Andromeda is optimized to handle sparse matrix computations. The company claims that the performance scales “almost linearly.” That is, as you double the number of cores used, you roughly half the total run time.

The machine is available for remote use and cost about $35 million to build. Since it uses 500 kW at peak run times, it isn’t free to operate, either. Extreme Tech notes that the Frontier computer at Oak Ridge National Labs is both larger and more precise, but it cost $600 million, so you’d expect it to be more capable.

Most homebrew “supercomputers” we see are more for learning how to work with clusters than trying to hit this sort of performance. Of course, if you have a modern graphics card, OpenCL and CUDA will let you do some of this, too, but at a much lesser scale.

The Descendants Of Ancient Computers

May 31, 2020 by 13 Comments

Building computers from discrete components is a fairly common hobby project, but it used to be the only way to build a computer until integrated circuits came on the scene. If you’re living in the modern times, however, you can get a computer like this running easily enough, but if you want to dive deep into high performance you’ll need to understand how those components work on a fundamental level.

[Tim] and [Yann] have been working on replicating circuitry found in the CDC6600, the first Cray supercomputer built in the 1960s. Part of what made this computer remarkable was its insane (for the time) clock speed of 10 MHz. This was achieved by using bipolar junction transistors (BJTs) that were capable of switching much more quickly than typical transistors, and by making sure that the support circuitry of resistors and capacitors were tuned to get everything working as efficiently as possible.

The duo found that not only are the BJTs used in the original Cray supercomputer long out of production, but the successors to those transistors are also out of production. Luckily they were able to find one that meets their needs, but it doesn’t seem like there is much demand for a BJT with these characteristics anymore.

[Tim] also posted an interesting discussion about some other methods of speeding up circuitry like this, namely by using reach-through capacitors and Baker clamps. It’s worth a read in its own right, but if you want to see some highlights be sure to check out this 16-bit computer built from individual transistors.

Hackaday Links Column Banner

Hackaday Links: March 29, 2020

March 29, 2020 by 19 Comments

It turns out that whacking busted things to fix them works as well on Mars as it does on Earth, as NASA managed to fix its wonky “mole” with a little help from the InSight lander’s robotic arm. Calling it “percussive maintenance” is perhaps a touch overwrought; as we explained last week, NASA prepped carefully for this last-ditch effort to salvage the HP³ experiment, and it was really more of a gentle nudge that a solid smack with the spacecraft’s backhoe bucket. From the before and after pictures, it still looks like the mole is a little off-kilter, and there was talk that the shovel fix was only the first step in a more involved repair. We’ll keep an ear open for more details — this kind of stuff is fascinating, and beats the news from Earth these days by a long shot.

Of course, the COVID-19 pandemic news isn’t all bad. Yes, the death toll is rising, the number of cases is still growing exponentially, and billions of people are living in fear and isolation. But ironically, we’re getting good at community again, and the hacker community is no exception. People really want to pitch in and do something to help, and we’ve put together some resources to help. Check out our Hackaday How You Can Help spreadsheet, a comprehensive list of what efforts are currently looking for help, plus what’s out there in terms of Discord and Slack channels, lists of materials you might need if you choose to volunteer to build something, and even a list of recent COVID-19 Hackaday articles if you need inspiration. You’ll also want to check out our calendar of free events and classes, which might be a great way to use the isolation time to better your lot.

Individual hackers aren’t the only ones pitching in, of course. Maybe of the companies in the hacker and maker space are doing what they can to help, too. Ponoko is offering heavy discounts for hardware startups to help them survive the current economic pinch. They’ve also enlisted other companies, like Adafruit and PCBWay, to join with them in offering similar breaks to certain customers.

More good news from the fight against COVID-19. Folding@Home, the distributed computing network that is currently working on folding models from many of the SARS-CoV-2 virus proteins, has broken the exaFLOP barrier and is now the most powerful computer ever built. True, not every core is active at any given time, but the 4.6 million cores and 400,000-plus GPUs in the network pushed it over from the petaFLOP range of computers like IBM’s Summit, until recently the most powerful supercomputer ever built. Also good news is that Team Hackaday is forming a large chunk of the soul of this new machine, with 3,900 users and almost a million work units completed. Got an old machine around? Read Mike Sczcys’ article on getting started and join Team Hackaday.

And finally, just because we all need a little joy in our lives right now, and because many of you are going through sports withdrawal, we present what could prove to be the new spectator sports sensation: marble racing. Longtime readers will no doubt recognize the mad genius of Martin and his Marble Machine X, the magnificent marble-dropping music machine that’s intended as a follow-up to the original Marble Machine. It’s also a great racetrack, and Martin does an amazing job doing both the color and turn-by-turn commentary in the mock race. It’s hugely entertaining, and a great tour of the 15,000-piece contraption. And when you’re done with the race, it’s nice to go back to listen to the original Marble Machine tune — it’s a happy little song for these trying times.

The New Xbox: Just How Fast Is 12 TeraFLOPS?

March 23, 2020 by 73 Comments

Microsoft’s new Xbox Series X, formerly known as Project Scarlet, is slated for release in the holiday period of 2020. Like any new console release, it promises better graphics, more immersive gameplay, and all manner of other superlatives in the press releases. In a sharp change from previous generations, however, suddenly everybody is talking about FLOPS. Let’s dive in and explore what this means, and what bearing it has on performance.

Continue reading “The New Xbox: Just How Fast Is 12 TeraFLOPS?” →

Add A Bit Of Soviet-Era Super-Computing To Your FPGA

May 3, 2019 by 11 Comments

The MESM-6 project is focused on bringing the 1960s Soviet BESM-6 computer to the modern age of FPGAs and HDLs. At the moment the team behind this preservation effort consists out of [Evgeniy Khaluev], [Serge Vakulenko] and [Leo Broukhis], who are covering the efforts on the Russian-language project page.

The BESM-6 (in Russian: БЭСМ-6, ‘Bolshaya Elektronno-Schetnaya Mashina’ or ‘large electronic computing machine’) was a highly performing Soviet super computer that was first launched in 1968 and in production for the next 19 years. Its system clock ran at 9 MHz using an astounding number of discrete components, like 60,000 transistors and 170,000 diodes, capable of addressing 192 kB of memory in total. Of the 355 built, a few survive to this day, with one on display at the London Science Museum (pictured above). Many more images and information can be found on its Russian Wikipedia page.

For those not gifted with knowledge of the Russian language, the machine-translated summary reveals that the project goal is to make a softcore in SystemVerilog that is compatible with user mode BESM-6, using the same Pascal compiler as originally used with that system. Further goals include at least 24 kB of data memory, 96 kB of command memory and the addition of modern peripherals such as SPI and I2C.

The system is meant to be integrated with the Arduino IDE, using the Pascal compiler to make it highly accessible to anyone with an interest in programming a system like this. Considering the MIT license for the project, one could conceivably use a bit of Soviet-era computing might in one’s future FPGA efforts.

If after watching the BESM-6 video — included below — you feel inspired to start your own Soviet-computing project, we’d like to wish you luck the Russian way: Ни пуха ни пера!

Continue reading “Add A Bit Of Soviet-Era Super-Computing To Your FPGA” →

Seymour Cray, Father Of The Supercomputer

January 15, 2019 by 59 Comments

Somewhere in the recesses of my memory there lives a small photograph, from one of the many magazines that fed my young interests in science and electronics – it was probably Popular Science. In my mind I see a man standing before a large machine. The man looks awkward; he clearly didn’t want to pose for the magazine photographer. The machine behind him was an amazing computer, its insides a riot of wires all of the same color; the accompanying text told me each piece was cut to a precise length so that signals could be synchronized to arrive at their destinations at exactly the right time.

My young mind was agog that a machine could be so precisely timed that a few centimeters could make a difference to a signal propagating at the speed of light. As a result, I never forgot the name of the man in the photo – Seymour Cray, the creator of the supercomputer. The machine was his iconic Cray-1, the fastest scientific computer in the world for years, which would go on to design nuclear weapons, model crashes to make cars safer, and help predict the weather.

Very few people get to have their name attached so firmly to a product, let alone have it become a registered trademark. The name Cray became synonymous with performance computing, but Seymour Cray contributed so much more to the computing industry than just the company that bears his name that it’s worth taking a look at his life, and how his machines created the future.

Continue reading “Seymour Cray, Father Of The Supercomputer” →

ILLIAC Was HAL 9000’s Granddaddy

October 18, 2018 by 42 Comments

Science fiction is usually couched in fact, and it’s fun to look at an iconic computer like HAL 9000 and trace the origins of this artificial intelligence gone wrong. You might be surprised to find that you can trace HAL’s origins to a computer built for the US Army in 1952.

If you are a fan of the novel and movie 2001: A Space Oddessy, you may recall that the HAL 9000 computer was “born” in Urbana, Illinois. Why pick such an odd location? Urbana is hardly a household name unless you know the Chicago area well. But Urbana has a place in real-life computer history. As the home of the University of Illinois at Urbana–Champaign, Urbana was known for producing a line of computers known as ILLIAC, several of which had historical significance. In particular, the ILLIAC IV was a dream of a supercomputer that — while not entirely successful — pointed the way for later supercomputers. Sometimes you learn more from failure than you do successes and at least one of the ILLIAC series is the poster child for that.

The Urbana story starts in the early 1950s. This was a time when the 1945 book “First Draft of a Report on the EDVAC” was sweeping through the country from its Princeton origins. This book outlined the design and construction of the Army computer that succeeded ENIAC. In it, Von Neumann proposed changes to EDVAC that would make it a stored program computer — that is, a computer that treats data and instructions the same.

Continue reading “ILLIAC Was HAL 9000’s Granddaddy” →