Raspberry Pis Chained Together Provide Massive Computing Muscle

by Ostatic Staff - May. 21, 2013

As we've covered before, when it comes to the top open source stories of the last 12 months, it's clear that one of the biggest is the proliferation of tiny, inexpensive Linux-based computers at some of the smallest form factors ever seen. The Linux-based Raspberry Pi, priced at $25 and $35, leads the pack among these devices.

But in a new twist on what Raspberry Pi devices are capable of, they're being chained together to form supercomputers and powerful clusters. If it sounds like a joke, you may be surprised at the enormous computing power these lash-ups are capable of. They may even have the power to democratize supercomputing-level data crunching at very low price points.

Late last year, as we noted here, news came from the University of Southampton that Professor Simon Cox and his team of researchers had lashed together an actual supercomputer made of 64 credit card-sized Raspberry Pis using Lego pieces as the glue for the cluster.  The imaginative result is shown in the photo above.

Professor Cox said: “As soon as we were able to source sufficient Raspberry Pi computers we wanted to see if it was possible to link them together into a supercomputer. We installed and built all of the necessary software on the Pi starting from a standard Debian Wheezy system image and we have published a guide so you can build your own supercomputer.”

You can get the guide to making your own Raspberry Pi supercomputer here, and find more information on Cox's version here.

Meanwhile, Boise University PhD candidate Joshua Kiepert has mashed up a 32-node cluster from Raspberry Pi devices. He has a great account of his effort and the architecture he used online (PDF). As The Register notes, Kiepert's total bill of materials came to $1967.21, and:

"As he contemplated the design for such a cluster, Kipert settled on Arch Linux for its tiny size. He eschewed the Pi's micro-USB port as a power source, as he felt it would complicate cabling, instead using a 5V pin on the machines' I/O headers...The Pis were also overclocked, using Turbo Mode, to give Kiepert the grunt needed to run his simulations."

According to Kiepert, "by building my own cluster I could outfit it with anything I might need directly."

These lash-ups of many Raspberry Pis working in tandem may sound like novelties, but if you look at Cox's and Kiepert's accounts of their projects it becomes clear that they are harnessing massive compute power at very low price points.

There is a need for supercomputers and clusters that can perform ongoing, dedicated computing tasks without racking up huge bills for researchers. The Raspberry Pi may end up being part of the equation that solves for that need.

 Photo Credit: University of Southampton and Professor Simon Cox