Parallella: A Supercomputer For Everyone (Kickstarter)
I just want to give a shout about a Kickstarter project that I think deserve some attention.
Basically a well established hardware chip design firm, want's to make a $99 parallel "super" computer and needs help financing mass production to lower the cost of chip production. In return for our support they promise full disclosure for all hardware and dedicated open source development tools.
So for a $99 Kickstarter pledge you get:
- Dual-core ARM A9 CPU
- Epiphany Multicore Accelerator (16 or 64 cores)
- 1GB RAM
- MicroSD Card
- USB 2.0 (two)
- Two general purpose expansion connectors
- Ethernet 10/100/1000
- HDMI connection
- Ships with Ubuntu OS
- Ships with free open source Epiphany development tools that include C compiler, multicore debugger, Eclipse IDE, OpenCL SDK/compiler, and run time libraries.
- Dimensions are 3.4'' x 2.1''
The reason I am so excited about this, is that this thing is more or less made for next generation autopilots.
- Full blown extended Kalman filter with all the bells and whistles? Would not even break a sweat doing that.
- Real-time visual navigation and object detection? You bet!
- Generate aerial maps on the fly. Be my guest!
You get the idea, and all this would be possible using a 86x54mm board consuming 5W on average. Needless to say I am a bit excited. :)
The $99 pledge is for a 16 core version, but there is talk about a $199 64-core version if they get enough to ramp up the production.
For more information visit the kickstarter here:
http://www.kickstarter.com/projects/adapteva/parallella-a-supercomp...
-
Comment by Wouter van Verre on September 29, 2012 at 2:31pm -
That looks really interesting! I think I might back this one.
-
Comment by Peter Meister on September 29, 2012 at 7:19pm -
I most definitely am backing this. I have been in the HPC business for over 15 years. It would be great to see what we could accomplish in the DIY field with a system like this. Thanks for highlighting this John !!!
-
Comment by Juan Carlos Hdez on September 30, 2012 at 3:05am -
very interesting. backed.
-
Comment by Christoph Name on September 30, 2012 at 3:43am -
Well I have no clue how you actually programm such a device efficiently. From OpenMP I know it's not that easy to take good advantage from a multicore CPU after it really depens on your application.
The advertisement sounds really amazing, but I am uncertain after I just dived into the single core 32Bit uC-World.
Anyway ... I still have a fewe weeks to make up my mind.
-
Developer Comment by John Arne Birkeland on September 30, 2012 at 7:05am -
Parallel computing is becoming easier every day. For example you now have parallel_for() and parallel_while() and other such helper functions in C++. A parallel_for() loop will for example try and run all the iterations in parallel, depending on how many cores you have.
You may also just pretend that you have 16 separates processors. This approach might actually be very well suited for Autopilot's where you want to make sure certain tasks are always executed on time. Each core could then be assigned essential tasks like sensor filtering, stabilization, GPS navigation etc.
And as you may have noticed, the Parallella system will support development using OpenCL. A dedicated programming language for paralell multi-core processing.
-
Comment by Batbileg Batjargal on September 30, 2012 at 7:53pm -
Very cool. Backed.
-
Comment by Austin H on October 1, 2012 at 1:02pm
-
backed it !
-
Comment by Andy on October 1, 2012 at 9:08pm -
Backed as well. Thanks for bringing this up.
The low power requirements and open source nature of their proposal is intriguing. Hope it goes through.
-
Comment by Brian Becker on October 2, 2012 at 6:50am
-
Curious to know what a guy (me) would do with this. I have no programming skills but still know enough to know this is worth backing. What will I do with it when I get it? Thoughts?
Social Networking
Top Content
Contests
Season Two of the Trust Time Trial (T3) Contest has now begun. The fourth round is an accuracy round for multicopters, which requires contestants to fly a cube. The deadline is April 14th.A list of all T3 contests is here.
Groups
-
ArduCopter Italian Group
86 members
-
Documentation Editors
51 members
-
ArduPlane User Group
671 members
-
ArduCopter User Group
1280 members
-
Taiwan ROC ArduPilot Gro…
8 members
Advertisement
© 2013 Created by Chris Anderson.
Powered by
You need to be a member of DIY Drones to add comments!
Join DIY Drones