[James J. Guthrie] just published a rather formal announcement that his 4-node Raspberry Pi cluster greatly outperforms a 64-node version. Of course the differentiating factor is the version of the hardware. [James] is using the Raspberry Pi 2 while the larger version used the Model B.
We covered that original build almost three years ago. It’s a cluster called the Iridris Pi supercomputer. The difference is a 700 MHz single core versus the 900 Mhz quad-core with double-the ram. This let [James] benchmark his four-node-wonder at 3.048 gigaflops. You’re a bit fuzzy about what a gigaflops is exactly? So were we… it’s a billion floating point operations per second… which doesn’t matter to your human brain. It’s a ruler with which you can take one type of measurement. This is triple the performance at 1/16th the number of nodes. The cost difference is staggering with the Iridris ringing in at around £2500 and the light-weight 4-node built at just £120. That’s more than an order of magnitude.
Look, there’s nothing fancy to see in [James’] project announcement. Yet. But it seems somewhat monumental to stand back and think that a $35 computer aimed at education is being used to build clusters for crunching Ph.D. level research projects.
Noting that funding for science has run dry for many researchers, [Gaurav] has built a supercomputer from 200 Playstation 3 consoles housed and chilled inside an old refrigerated shipping trailer. His mission at UMass Dartmouth from the National Science Foundation is simulating black hole collisions with an eye on learning something about gravitational waves.
Dr. [Gaurav Khanna] is no stranger to using PS3 supercomputers to do meaningful science. Seven years ago he proposed a 16-PS3 supercomputer running Linux and managed to convince Sony to donate four consoles. The university kicked in funding for another 8 and [Gaurav] ponied up for the last four out of his own pocket. He dubbed it the “PS3 Gravity Grid” and received international attention for the cluster. For equivalent performance, it cost him only 10% the price of a real supercomputer. This led to published papers on both hacked supercomputers and gravity waves. But that rig is looking a little old today. Enter the Air Force.
Dr. [Khanna] was not the only one using PS3s to crunch data – back in 2010 the US Air Force built the “Condor Cluster” of 1,760 PS3s to perform radar imaging of entire cities and do neuromorphic AI research. With their hardware now expired, the Air Force donated 200 of the PS3s to [Gaurav] for his new build. Now that he has wired them up, the Air Force is donating another 220 for a not-snicker-proofed total of 420.
For those sceptical that the now 8-year-old hardware is still cost-effective, even with free consoles it is marginal. RAM is an issue and modern graphics cards are each equivalent to 20 PS3s. Ever the popular target these days, Sony has the PS4 OS locked down from the get go – thanks Sony. The next cluster planned will be with PCs and graphics cards. For now, [Gaurav] has plenty of calculations that need crunching and a queue of colleagues have formed behind him.
Multi-node RasPi clusters seem to be a rite of passage these days for hackers working with distributed computing. [Dave’s] 40-node cluster is the latest of the super-Pi creations, and while it’s not the biggest we’ve featured here, it may be the sleekest.
The goal of this project—aside from the obvious desire to test distributed software—was to keep the entire package below the size of a full tower desktop. [Dave’s] design packs the Pi’s in groups of 4 across ten individual cards that easily slide out for access. Each is wired (through beautiful cable management, we must say) to one of the 2 24-port switches at the bottom of the case. The build uses an ATX power supply up top that feeds into individual power for the Pi’s and everything else, including his HD array—5 1TB HD’s, expandable to 12—a wireless router, and a hefty fan assembly.
Perhaps the greatest achievement is the custom acrylic case, which [Dave] lasered out at the Dallas Makerspace (we featured it here last month). Each panel slides off with the press of a button, and the front/back panels provide convenient access to the internal network via some jacks. If you’ve ever been remotely curious about a build like this one, you should cruise over to [Dave’s] page immediately: it’s one of the most meticulously well-documented projects we’ve seen in a long time. Videos after the break.
Continue reading “40-Node Raspi Cluster”
In retrospect, it was only a matter of time before someone turned a bunch of Raspberry Pis into a supercomputer.
The Raspi supercomputer is the result of a project headed up by University of Southampton professor [Simon Cox]. Included in the team are a gaggle of grad students and [Simon]’s 6-year-old son who graciously provided the material, design, and logistics for the custom LEGO case.
The Iridris-Pi supercomputer, as the team calls their creation, consists of 64 Raspberry Pis, all configured for parallel processing using a lightweight version of MPI. [Simon] was kind enough to put up an excellent guide for turning two (or more) Raspberry Pis into a supercomputer.
The machine has a full 1 TB of disk space provided by a 16 GB SD card in each node. Although the press release doesn’t go over the computational capabilities of the Iridris-Pi, the entire system can be powered from a single 13 A supply.
If you’re wondering what it would take to get a Raspberry Pi supercomputer into the TOP500 list of supercomputers, a bit of back-of-the-envelope computation given the Raspi’s performance and the fact the 500th fastest computer can crank out about 60 TeraFLOPS/s, we’ll estimate about 1.4 Million Raspis would be needed. At least it’s a start.
In August, 2010, [Alexander Yee] and [Shigeru Kondo] won a respectable amount of praise for calculating pi to more digits than anyone else. They’re back again, this time doubling the number of digits to 10 Trillion.
The previous calculation of 5 Trillion digits of Pi took 90 days to calculate on a beast of a workstation. The calculations were performed on 2x Xeon processors running at 3.33 GHz, 96 Gigabytes of RAM, and 32 Terabytes worth of hard drives. The 10 Trillion digit attempt used the same hardware, but needed 48 Terabytes of disk to store everything.
Unfortunately, the time needed to calculate 10 Trillion digits didn’t scale linearly. [Alex] and [Shigeru] waited three hundred and seventy-one days for the computer to finish the calculations. The guys used y-cruncher, a multithreaded pi benchmarking tool written by [Alex]. y-cruncher calculates hexadecimal digits of pi; conveniently, it’s fairly easy to find the nth hex digit of pi for verification.
If you’re wondering if it would be faster to calculate pi on a top 500 supercomputer, you’d be right. Those boxes are a little busy predicting climate change, nuclear weapons yields, and curing cancer, though. Doing something nobody else has ever done is still an admirable goal, especially if it means building an awesome computer.
In our hubris, we pat ourselves on the back when we’re able to pull data off our old SCSI drives. [Chris Fenton]’s attempt to get an OS for a homebrew Cray-1 puts us rightfully to shame.
Last year we saw [Chris]’ fully functional 1/10th scale Cray-1 supercomputer built around FPGA. While the reproduction was nearly cycle-accurate, [Chris] hasn’t had an opportunity to test out his system because of the lack of available Cray software. A former Cray employee heard of his plight and loaned an 80 Megabyte CDC 9877 disk pack to in the hope of getting some system software.
[Chris] acquired a monstrous 100 pound disk drive to read the disk pack, but after 30 years in storage a lot of electrical problems cropped up. Since reading the drive digitally proved to be an exercise in futility, [Chris] hit upon the idea of taking analog data straight from the read head. This left him with a magnetic image of the disk pack that was ready for some data analysis.
After the disk image was put up on the Internet, the very talented [Yngve AAdlandsvik] figured out the data, header, and error correction formats and sent [Chris] a Python script to tease bits from the analog image. While no one is quite sure what is on the disk pack provided by the Cray employee, [Chris] is remarkably close to bringing the Cray-1 OS back from the dead. There’s also a great research report [Chris] wrote as penance for access to the CDC disk drive. Any Hack A Day readers feel like looking over the data and possibly giving [Chris] a hand?
[Chris Fenton] spent a year and a half constructing a 1/10th scale Cray-1 reproduction. The famous supercomputer was meticulously modelled in a field programmable gate array for a “nearly cycle-accurate” reproduction. [Chris’] hardware of choice for the project is a Xilinx Spartan-3E 1600 development board, using 75-80% of the available resources. The finished product runs at 33 MHz and is missing a few functions but it sounds like they don’t affect code execution. We like that he didn’t stop with the processor implementation, but also took the time to produce a case for the development board that looks just like the original.
Unlike the Atari 2600 FPGA project, we’re not quite sure what we’d use this for. But that doesn’t diminish the excellence of his work.