Super Computing With Mini ITX Cluster

March 16, 2017 by Will Sweatman 16 Comments

[Colin Alston] was able to snag a handful of Mini ITX motherboards for cheap and built a mini super computer he calls TinyJaguar. Named partly after the AMD Sempron 2650 APU, the TinyJaguar boasts four, yes that’s four MSI AM1I Mini-ITX motherboards, each with 4GB of DDR memory.

A Raspberry Pi with custom software manages the cluster, and along with some TTL and relays, controls the power to the four nodes. The mini super computer resides in a custom acrylic case held together by an array of 3D printed parts and fasteners.There’s even a rack-like faceplate near the bottom to host the RPi, an Ethernet switch, an array of status LEDs, and the two buttons.

With 16 total cores of computing power (including GPU), the TinyJaguar is quite capable of doing some pretty cool stuff such as running Jupyter notebook with IPyParallel. [Colin] ran into some issues getting the GPU to behave with PyOpenCL. It took a bit of pain and time, but in the end he was able to get the GPUs up, and wrote a small message passing program to show two of the cores were up and working together.

Be sure to check out [Colin’s] super computer project page, specifically the ten project logs that walk through everything that went into this build. He also posted his code if you want to take a look under the hood.

80-PIC32 Cluster Does Fractals

December 19, 2016 by Al Williams 26 Comments

One way to get around limitations in computing resources is to throw more computers at the problem. That’s why even cheap consumer-grade computers and phones have multiple cores in them. In supercomputing, it is common to have lots of processors with sophisticated sharing mechanisms.

[Henk Verbeek] decided to take 80 inexpensive PIC32 chips and build his own cluster programmed in — of all things — BASIC. The devices talk to each other via I2C. His example application plots fractals on another PIC32-based computer that has a VGA output. You can see a video of the device in action, below.

Continue reading “80-PIC32 Cluster Does Fractals” →

Hackaday Prize Entry: A Cluster Of Exoskeletons

October 25, 2016 by Brian Benchoff 5 Comments

The current trend of 3D printed prosthetic hands have one rather large drawback: you can’t use them if you already have two hands. This might seem like a glib objection, but one of last week’s Hackaday Prize posts pointed this out rather well – sometimes a meat machine needs mechanical assistance.

BEOWULF, [Chad Paik]’s entry for the Hackaday Prize, is the answer to this problem. It’s a mechanical exoskeleton for grip enhancement, stroke rehabilitation, and anyone else that doesn’t have the strength they need to get through the day.

This project solves the problem of weak arm strength through – you guessed it – 3D-printed parts, a linear actuator on the forearm, and a few force sensors on the fingertips. Control is obtained through a Thalmic Labs Myo, but the team behind the BEOWULF is currently working on a custom muscle activity sensor that is more compact and isn’t beholden to VC investors. You can check out a video of this exoskeleton below.

Continue reading “Hackaday Prize Entry: A Cluster Of Exoskeletons” →

Clustering A Lot Of Raspberry Pi Zeros

September 18, 2016 by Jenny List 55 Comments

It became something of a cliché a few years ago in online discussions, whenever a new single board computer was mentioned someone would pop up and say something like “Imagine a Beowulf cluster…“. Back then it was said largely in jest, but with the current generation of boards it’s a distinct possibility. Who hasn’t looked at a Raspberry Pi and idly thought about a cluster of them, or even created one!

[Electronoob] did just that, creating a variety of Raspberry Pi cluster configurations, the most impressive of which is a stack of 32 Pi Zeros mounted together with stand-offs. The plan was to network it via USB, for which he initially considered building a backplane, but was put off by the cost of vertical USB connectors and instead went for a wired approach. If there is a lesson to be learned from his experiences it is that buying very cheap USB cables is a minefield: his pile of eBay specials turned out to have significant numbers of faults. He’s now faced with a stark choice, solder 32 sets of USB pads on the base of each Zero or buy better cables.

The stack of Zeros is pretty impressive, but so what, you think. It’s still not working properly. But the Zero cluster isn’t his only work. He’s also created a set of very nicely executed Ethernet clusters using the larger Pi boards, and the way he’s mounted them on top of compact Ethernet switches sets them apart from some of the more spaghetti-like Pi clusters.

It’s true a Pi cluster won’t cut it in the world of supercomputers, you could almost certainly buy more bang for your buck without too much effort. But it does represent a very accessible way to learn about cluster computing, and you have to admit it a stack of Zeros does look rather impressive.

We’ve seen quite a few Pi clusters here since 2012, the biggest of which is probably this 120 node behemoth, complete with screens.

BitCluster Brings A New Way To Snoop Through BitCoin Transactions

July 23, 2016 by Mike Szczys 8 Comments

Mining the wealth of information in the BitCoin blockchain is nothing new, but BitCluster goes a long way to make sense of the information you’ll find there. The tool was released by Mathieu Lavoie and David Decary-Hetu, PH.D. on Friday following their talk at HOPE XI.

I greatly enjoyed sitting in on the talk which began with some BitCoin basics. The cryptocurrency uses user generated “wallets” which are essentially addresses that identify transactions. Each is established using key pairs and there are roughly 146 million of these wallets in existence now

If you’re a thrifty person you might think you can get one wallet and use it for years. That might be true of the sweaty alligator-skin nightmare you’ve had in your back pocket for a decade now. It’s not true when it comes to digital bits — they’re cheap (some would say free). People who don’t generate a new wallet for every transaction weaken their BitCoin anonymity and this weakness is the core of BitCluster’s approach.

Every time you transfer BitCoin (BTC) you send the network the address of the transaction when you acquired the BTCs and sign it with your key to validate the data. If you reuse the same wallet address on subsequent transactions — maybe because you didn’t spend all of the wallet’s coins in one transaction or you overpaid and have the change routed back to your wallet. The uniqueness of that signed address can be tracked across those multiple transactions. This alone won’t dox you, but does allow a clever piece of software to build a database of nodes by associating transactions together.

Mathieu’s description of first attempts at mapping the blockchain were amusing. The demonstration showed a Python script called from the command line which started off analyzing a little more than a block a second but by the fourth or fifth blocks hit the process had slowed to a standstill that would never progress. This reminds me of some of the puzzles from Project Euler.

After a rabbit hole of optimizations the problem has been solved. All you need to recreate the work is a pair of machines (one for Python one for mondoDB) with the fastest processors you can afford, a 500 GB SSD, 32 GB of RAM (but would be 64 better), Python 64-bit, and at least a week of time. The good news is that you don’t have to recreate this. The 200GB database is available for download through a torrent and the code to navigate it is up on GitHub. Like I said, this type of blockchain sleuthing isn’t new but a powerful open source tool like this is.

Both Ransomware and illicit markets can be observed using this technique. Successful, yet not-so-cautious ransomers sometimes use the same BitCoin address for all payments. For example, research into a 2014 data sample turned up a ransomware instance that pulled in $611k (averaging $10k per day but actually pulling in most of the money during one three-week period). If you’re paying attention you know using the same wallet address is a bad move and this ransomware was eventually shut down.

Illicit markets like Silk Road are another application for BitCluster. Prior research methods relied on mining comments left by customers to estimate revenue. Imagine if you had to guess at how well Amazon was doing reading customer reviews and hoping they mentioned the price? The ability to observe BTC payment nodes is a much more powerful method.

A good illicit market won’t use just one wallet address. But to protect customers they use escrow address and these do get reused making cluster analysis possible. Silk Road was doing about $800k per month in revenue at its height. The bulk of purchases were for less than $500 with only a tiny percentage above $1000. But those large purchases were likely to be drug purchases of a kilo or more. That small sliver of total transactions actually added up to about a third of the total revenue.

It’s fascinating to peer into transactions in this manner. And the good news is that there’s plenty of interesting stuff just waiting to be discovered. After all, the blockchain is a historical record so the data isn’t going anywhere. BitCluster is intriguing and worth playing with. Currently you can search for a BTC address and see total BTC in and out, then sift through income and expense sorted by date, amount, etc. But the tool can be truly great with more development. On the top of the wishlist are automated database updates, labeling of nodes (so you can search “Silk Road” instead of a numerical address), visual graphs of flows, and a hosted version of the query tool (but computing power becomes prohibitive.)

Raspberry Pi Cluster Build Shows How And What

May 26, 2016 by Richard Baguley 36 Comments

Raspberry Pi clusters are a dime a dozen these days. Well, maybe more like £250 for a five-Pi cluster. Anyway, this project is a bit different. It’s exquisitely documented.

[Nick Smith] built a 5-node Pi 3 cluster from scratch, laser-cutting his own acrylic case and tearing down a small network switch to include in the design. It is, he happily admits, a solution looking for a problem. [Smith] did an excellent job of documenting how he designed the case in CAD, prototyped it in wood, and how he put the final cluster together with eye-catching clear acrylic.

Of interest is that he even built his own clips to hold the sides of the case together and offers all of the files for anyone who wants to build their own. Head over to his page for the complete bill of materials (we didn’t know Pis were something you could order in 5-packs). And please, next time you work on a project follow [Nick’s] example of how to document it well, and how to show what did (and didn’t) work.

If 5 nodes just doesn’t do it for you, we suggest this 120-node screen-equipped monster, and another clear-acrylic masterpiece housing 40 Pis. This stuff really isn’t only for fun and games. Although it wasn’t Pi-based, here’s a talk at Hackaday Belgrade about an ARM-based SBC cluster built to crunch numbers for university researchers.

Designing A High Performance Parallel Personal Cluster

May 9, 2016 by Brian Benchoff 86 Comments

Kristina Kapanova is a PhD student at the Bulgarian Academy of Sciences. Her research is taking her to simulations of quantum effects in semiconductor devices, but this field of study requires a supercomputer for billions of calculations. The college had a proper supercomputer, and was getting a new one, but for a while, Kristina and her fellow ramen-eating colleagues were without a big box of computing. To solve this problem, Kristina built her own supercomputer from off-the-shelf ARM boards.