Six Clicks: More Linux single-board computers

There are many great Linux-powered single-board computers, starting with the new Raspberry Pi B+.
Topic: Big Data
1 of 6 Steven Vaughan-Nichols/ZDNet

Raspberry Pi B+

DIY and gadget fans alike love the Raspberry Pi. Now, they'll have more to love with the new Raspberry Pi B+.

For the same $35 price as the original Model B, the Model B+ still uses the ARM-based Broadcom BCM2835 chipset, 512MB of RAM, and 700MHz low-power ARM1176JZ-F applications processor.

The improvements come with a more space efficient microSD card slot in place of the SDCard slot. It also has four USB 2.0 ports instead of the Model B's two. The new single-board computer (SBC) also has 40 general purpose Input/Output (GPIO) pins. The Model B has only 26. The B+ also has a single four-pin port for both both audio and video output. The audio circuit should also deliver cleaner sound with a new low-noise power supply.

Last, but not least, the B+ uses a half to full watt less power. If you're been working on Raspberry Pi projects that require long battery life, this is the SBC you've been waiting for.

2 of 6 Steven Vaughan-Nichols/ZDNet

HummingBoard SBC Family

OK, let's make this clear from the top: SolidRun's three HummingBoard SBCs are more expensive than the Raspberry Pi. The least powerful HummingBoard-i1 starts at $44.99 without a $10 power supply, while the top-of the-line HummingBoard-i2eX begins at $99.99 and goes up to $116.99, but, again, without a power supply.

On the other hand, what you get for your money is an SBC with a much more powerful processor and more RAM. It ranges from a Freescale i.MX6 with one or two Cortex-A9 cores running at 1GHz, with a C880 GPU for the two lower-end SBCs and a Vivante GC2000 for the high-end i23X. The bottom-end i1 comes with 51MBs of RAM while the other two models come with a GB of RAM.

For storage, the i1 and i2 use a MicroSD slot while the i2Ex adds a mSATA II interface. The former two use fast ethernet for network connectivity and come with two USB 2.0 ports; the i2Ex supports gigabit ethernet and includes four USB 2.0 ports.

The vendor states that all three support Android 4.4.x, Yocto, Ubuntu, Debian, FreeBSD and others.

Put it all together and what you get are powerful SBCs for users who want to tackle projects too big for a Raspberry Pi.

3 of 6 Steven Vaughan-Nichols/ZDNet


The $89 CubieTruck from CubieBoard is a hacker's delight. How much so? Would you believe someone used 22 CubieBoards to build a cluster to handle a Hadoop big data project? Believe it.

The CubieTruck, the most powerful of the current generation of CubieBoard SBCs, is powered by an AllWinner A20 System on a Chip (SoC) with dual ARM Cortex-A7 CPUs and an ARM Mali400 MP2 GPU. It also has 2GBs of RAM, a SATA 2.0 interface, 100Mbps ethernet, a pair of USB 2.0 ports, and, in an unusual twist for an SBC, built-in wifi and Bluetooth. For storage it uses MicroSD.

This device will run Android 4.2.2, Fedora 20, and several versions of Lubuntu, a lightweight version of Ubuntu.

If you want a SBC with plenty of oomph, you should check this one out.

4 of 6 Steven Vaughan-Nichols/ZDNet


What's that you say? You want real speed on your SBC? Well, Hardkernel's Ordoid-U3 features a Samsung Exynos 4412 Prime SoC, with a 1.7GHz quad-core ARM Cortex-A9 and a 3D accelerator Mali-400 Quad Core 440MHz. That's about as fast as you're going to get on an ARM-powered SBC these days.

It will cost you $65, but with all that processor power, you'll also get 2GBs of RAM, a MicroSD slot, four USB 2.0 ports and an 100Mbps ethernet port.

I like this board, with its support for Android and many versions of Linux, a lot. I think you will too.

5 of 6 Steven Vaughan-Nichols/ZDNet


You knew you could build a supercomputer out of Raspberry Pi boards right? Well, the Adapteva Parallella, with its dual-core ARM A9 processor and 16-core Epiphany-III Accelerator chip, 1GB of RAM, microSD card, two USB 2.0 ports, 10/100/1000 Ethernet, and HDMI connection, was designed from the get-go for supercomputing parallel-processing.

Yes, you read that right. The chipset that makes the parallel-processing magic has 16, count 'em, 16 cores.

That chip is also the reason the board comes with one honking huge heatsink covering most of the board. It's one hot chip!

The company has just starting shipping a new family of Parallella boards. These are the Microserver, $119; Desktop, $149; and Embedded, $249. All versions come with 1GB of RAM and gigabit ethernet. The main difference between the three models is the size of the expansion I/O. In addition, the microserver doesn't come with USB or HDMI ports.

Now, while Parallella supports Ubuntu, Debian, and FreeBSD for its operating system, to really get the most from it you need to be interested in learning parallel processing. This is not the kind of programming you learned in school with C, Java, or Python. But, if you are interested in supercomputing, and you don't have a "real" supercomputer to call your own, the Parallella family is your next best bet.

6 of 6 Steven Vaughan-Nichols/ZDNet


OK, so the Arduino is not an SBC per se. Still, since it's really the first important open hardware board, I think it deserves mention. The Arduino has been around since 2005 and it's always been intended to help introduce people to embedded computing.

Today, there 20 different Arduino boards. Some are ready to go, while with others you'll need to assemble them yourself — soldering iron not included.

The best way to get going with Arduino is with the Arduino Starter Kit. To make programs that will work on the boards, you'll need to download the open source Arduino software developer environment for your Mac, Linux, or Windows PC. 

Most Arduino boards don't even have an operating system as such. The few that do, such as the Yún, use OpenWRT, an embedded Linux distribution.

If you're the kind of person, like yours truly, who grew up with a soldering gun in one hand and an oscilloscope probe in the other, you'll like the Arduino gear. Otherwise, you'll be better off with hardware that's not quite so hard.

Related Galleries

Azure Synapse Analytics data lake features: up close

Related Galleries

Azure Synapse Analytics data lake features: up close

Pitfalls to Avoid when Interpreting Machine Learning Models

Related Galleries

Pitfalls to Avoid when Interpreting Machine Learning Models

When chatbots are a very bad idea
When Chatbots are a very bad idea ZDNet

Related Galleries

When chatbots are a very bad idea

How ubiquitous AI will permeate everything we do without our knowledge.
How ubiquitous AI will permeate everything we do without our knowledge ZDNet

Related Galleries

How ubiquitous AI will permeate everything we do without our knowledge.

Streaming becomes mainstream

Related Galleries

Streaming becomes mainstream

Photos: How FC Barcelona uses football player data to win games

Related Galleries

Photos: How FC Barcelona uses football player data to win games

Heart and sleep apps that work with the Apple Watch

Related Galleries

Heart and sleep apps that work with the Apple Watch