Breaking News
Adapteva's $100 Parallella Supercomputer Platform Now Shipping
7/23/2013

< Previous   Image 3 of 3   

Laptop connected to a 42-board, 756-CPU Parallella cluster, which consumes less than 500W!
Laptop connected to a 42-board, 756-CPU Parallella cluster, which consumes less than 500W!

< Previous   Image 3 of 3   

Return to Article

View Comments: Newest First | Oldest First | Threaded View
<<   <   Page 3 / 5   >   >>
adapteva
User Rank
Manager
Re: eLink
adapteva   7/24/2013 1:41:31 PM
mlloyd - A few years back we decided to stay away from the high speed SERDES and use a source synchronous LVDS interface instead. This way we could attach to low cost as well as high end FPGAs.  The interface does use a fair amount of pins (8 data lanes) but can provide up to 16Gb/s total bandwidth with a 500MHz clock. This turned out to be a good choice because the low cost zynq 7010 and 7020 don't currently support high speed serdes.

Max The Magnificent
User Rank
Blogger
Re: Thanks!
Max The Magnificent   7/24/2013 1:32:19 PM
NO RATINGS
@Adapteve: Once we put together a cluster of 1,000 of these Parallella boards...

Once you do, I want to see pictures!!!

mlloyd
User Rank
Rookie
eLink
mlloyd   7/24/2013 12:30:07 PM
NO RATINGS
Congratulations, Andreas, on shipping your product and nearing your large milestones.  I am curious about your eLink interface between the Zynq and Epiphany.  Since you designed the Epiphany from scratch, is it able to make use of the high-speed transceivers offered by the Zynq?  We commonly interface processors to FPGAs in my business, and it is always a challenge to find processors that have high-speed buses that we can use to communicate with our FPGAs.

adapteva
User Rank
Manager
Re: Backend
adapteva   7/24/2013 11:30:57 AM
@lorincz_#1

The Parallella actually has a ~10Gb/s memory mapped low-latency link (through the "PEC" connector) that can be used to to construct some interesting large scale topologies. See the specs here.http://www.parallella.org/board

 

lorincz_#1
User Rank
Rookie
Backend
lorincz_#1   7/24/2013 11:24:53 AM
NO RATINGS
Does anyone know why they didn't include some type of high speed message passing bus?

The supercomputing capability of such a scalable setup seems limited to the latency/throughput of the ethernet bus. 

adapteva
User Rank
Manager
Re: Thanks!
adapteva   7/24/2013 9:22:25 AM
NO RATINGS
Rick- Thanks! Once we put together a cluster of 1,000 of these Parallella boards we will definitely be getting up into real "supercomputer" territory, although the definition is a moving target.

rick merritt
User Rank
Author
Re: Thanks!
rick merritt   7/24/2013 9:06:12 AM
NO RATINGS
Congrats, Andreas. It will be interesting to see what people do with these boards--perhpas will will get a mini-Top 500.

adapteva
User Rank
Manager
Re: Thanks!
adapteva   7/24/2013 8:18:49 AM
NO RATINGS
Dylan- Thanks! Shipping 6,300 boards is the one that really matters to us. The expectations of 5,000 KS backers has been a big weight to carry for 9 months now. Fortunately for us they have been incredibly patient and understanding. We'll definitely keep you posted. If you don't hear from us something i wrong:-)

adapteva
User Rank
Manager
Re: Awesome
adapteva   7/24/2013 7:30:56 AM
NO RATINGS
Thanks! We did reach profitability briefly but were not able to sustain the momentum. As you know, profitability is not a stable state:-)

eewiz
User Rank
CEO
Re: Awesome
eewiz   7/24/2013 2:15:36 AM
NO RATINGS
oh ok.. i see that Adapteva cores are already shipping in other products

http://www.adapteva.com/products/system-products/

<<   <   Page 3 / 5   >   >>
August Cartoon Caption Winner!
August Cartoon Caption Winner!
"All the King's horses and all the KIng's men gave up on Humpty, so they handed the problem off to Engineering."
5 comments
Top Comments of the Week
Like Us on Facebook

Datasheets.com Parts Search

185 million searchable parts
(please enter a part number or hit search to begin)
EE Times on Twitter
EE Times Twitter Feed
Flash Poll
Radio
LATEST ARCHIVED BROADCAST
David Patterson, known for his pioneering research that led to RAID, clusters and more, is part of a team at UC Berkeley that recently made its RISC-V processor architecture an open source hardware offering. We talk with Patterson and one of his colleagues behind the effort about the opportunities they see, what new kinds of designs they hope to enable and what it means for today’s commercial processor giants such as Intel, ARM and Imagination Technologies.