Parallella: The $99 Linux supercomputer

ZDNet ^ | 15 April 2013 | Steven J. Vaughan-Nichols

[ShadowAce]

Chip-company Adapteva announced on April 15th at the Linux Collaboration Summit in San Francisco, California, that they've built their first Parallella parallel-processing board for Linux supercomputing, and that they'll be sending them to their 6,300 Kickstarter supporters and other customers by this summer.

Say hi to Parallella, the $99 Linux-powered supercomputer. (Image: The Linux Foundation)

Linux has long been the number one supercomputer operating system. But while you could build your own Linux supercomputer using commercial off-the-shelf (COTS) products, it wouldn't be terribly fast. You needed hardware that could support massively parallel computing — the cornerstone of modern supercomputing. 

What Adapteva has done is create a credit-card sized parallel-processing board. This comes with a dual-core ARM A9 processor and a 64-core Epiphany Multicore Accelerator chip, along with 1GB of RAM, a microSD card, two USB 2.0 ports, 10/100/1000 Ethernet, and an HDMI connection. If all goes well, by itself, this board should deliver about 90 GFLOPS of performance, or — in terms PC users understand — about the same horse-power as a 45GHz CPU.

This board will use Ubuntu Linux 12.04 for its operating system. To put all this to work, the platform reference design and drivers are now available.

Why would you want a $99 supercomputer?

Well, besides the fact that it would be really cool, Adapteva CEO Andreas Olofsson explained:

Historically, serial processing [conventional computing] improved so quickly that in most applications, there was no need for massively parallel processing. Unfortunately, serial processing performance has now hit a brick wall, and the only practical path to scaling performance in the future is through parallel processing. To make parallel software applications ubiquitous, we will need to make parallel hardware accessible to all programmers, create much more productive parallel programming methods, and convert all serial programmers to parallel programmers.

And of course, Olofsson added, to "make parallel computing accessible to everyone so we can speed up the adoption of parallel processing in the industry", the Parallella had to be created. Olofsson admitted that his company couldn't have done it by itself. The project required, and got, the support of other hardware OEMs, including Xilinx, Analog Devices, Intersil, Micron, Microchip, and Samtec. The companies have enabled Adapteva to bring its first per-production boards to San Francisco, and soon, to its eager programmer customers.

[BornToBeAmerican]

MS one step closer to the grave or open source

Damn FAST!!!!

[PapaBear3625]

Parallel computing isn’t of much use to most ordinary folks.

Scene rendering uses parallel computing. You think the average person might like being able to create nice-looking animation? You think video game designers might like that capability?

[ThomasMore]

Holographic Computing.

[Bullish]

Does this basically mean a virtually unlimited amount of RAM to run programs fast as lightning?

[OneWingedShark]

> Anyone using C# “await()” functionality yet? > > This is really cool...! Na, I use Ada's TASK construct, which has been included in the language since it's inception in 1983... which means that it foresaw this situation (parallel computation) more than thirty years ago. Indeed, they've been adding features to the tasking-model to ease some multi-core issues (core preference, and such).

[familyop]

Very interesting for 3D graphics rendering and other engineering considerations.

[PrairieDawg]

You wrote: Can I predict global warming with this rig?

No, but you can probably debunk it with this rig.

FRegards,
PrairieDawg

[Ernest_at_the_Beach]

Fun.

[Errant]

Among the many benefits: a couple of orders of magnitude in increased processing ability for the same amount of energy, at greatly reduced costs in hardware.

[Mr. K]

ADA? how quaint~!

(just kidding)

[OneWingedShark]

ADA?

No, Ada. It's a proper name, nor an acronym (I'd hate to program in the Americans with Disabilities Act) -- Ironically the one time that casing is important (well, besides string internals).

how quaint~!

quaint     [kweynt]
adjective

1. having an old-fashioned attractiveness or charm; oddly picturesque: a quaint old house.
2. strange, peculiar, or unusual in an interesting, pleasing, or amusing way: a quaint sense of humor.
3. skillfully or cleverly made.
4. Obsolete. wise; skilled.

Hm, which definition do you mean? In the case of:
1. Yes, strong-typing with an aim towards static analysis is very old-fashioned compared to the newfangled web/scripting languages.
2. A language designed with tangible goals like maintainability, large-systems design, and early detection of errors is quite unusual.
3. I agree.
4. Again, I agree.

(just kidding)

Awww.. :(
;)

But seriously, the new 2012 standard is nice: we got some fine tuning of TASK-constructs, contracts (pre- & post-conditions, along w/ type invariants), quantified expressions (for all and for some), conditionals (if X in positive then 128.0 / X else 0.0) [same w/ case, definite plus for having CASE-coverage (since Ada83)], some new containers.

I can get you some links if you're interested.

[Mr. K]

Nah, just call me when they want to reprogram it all in c#

:P

[OneWingedShark]

Nah, just call me when they want to reprogram it all in c#

I don't see that happening for things like Avionics, certainly not things like the Apache helicopter and the F-22.

[Mr. K]

I am actually getting a lot of calls for c# conversions.

Is Ada still used that extensively? Is there a high demand for it (and high$$$)? Maybe I should look into it.

I have been joking with you a bit about this (my language is better than yours Ntah! nyah!) but it is all in good fun.

I love doing avionics and hardware control- most of the embedded systems work I get calls for are in c++

[OneWingedShark]

I love doing avionics and hardware control- most of the embedded systems work I get calls for are in c++

Yeah, it sounds like a great field -- I'd love to get in, but I'm stuck with the "entry-level catch-22" -- you know, the entry-level job "requires X years of experience in the industry"... which you can't get w/o an entry-level job.

I have been joking with you a bit about this (my language is better than yours Ntah! nyah!) but it is all in good fun.

I could tell that.

Is Ada still used that extensively? Is there a high demand for it (and high$$$)? Maybe I should look into it.

It seems to be; the life-cycles in aeronautics are LONG compared to elsewhere (the 747 started production in 1968) ~ given that the F-22 is [mostly] done in Ada (I heard there were things like cockpit-control were C++) and the 777 is entirely Ada there's going to be some demand for it. (Again, the only reason I can see I'm not getting the jobs is because companies [in general] want people "pre-packaged" for the position [i.e. no training required].)

I am actually getting a lot of calls for c# conversions.

I've been involved in some C# conversions myself, in user applications, the notable one being a conversion from Access to C#. I cam in on the tail and basically had to fix porting issues (they changed development environs from VS 2008 to 2010 and .NET 2.0 to 3.5 as well as moving the system to other machines) -- I initially got the job of "fix it so it compiles" [reference fix-up]... and then put on maintenance -- the thing is kind of a wreck in that it's more a transliteration than a conversion/translation, and you can tell it was mostly developed by intern sorts who had no experience maintaining a system. (You quickly learn to hate "cut and paste programming".)