Winner Announced: What would you do with 48 cores?
The response to the “What would you do with 48 cores?” contest has been positively overwhelming! We closed the contest on March 24th with a grand total of 840 entries – and it’s clear from your participation and enthusiasm AND the # of page views (~35,000 views) that servers are indeed fun, and that we on the AMD server marketing team should do these things more often.
The challenge for all the judges?? It was extremely difficult for us to pick just one winner! Your responses had us pondering the various ways to make the world a better place; from using technology to help improve the climate, to helping our children excel in the math and sciences, to working to cure the various diseases that take the ones we love. On the lighter side we even learned of the relationship between bacon and server processors 
! Now the only thing to do is to roll-out the winners David Letterman style. Here is the list of the top 3 finalist as well as our grand prize winner. For those of you in the top 3, expect a personal note from me very soon!
3. Daniel Morgan – Daniel created a YouTube video titled “Technology for Congestion Relief”. Daniel takes on the damage to the environment, productivity, and quality of life that automobile traffic causes. Check out Daniel’s creative video showing how technology can be used for traffic congestion management.
2. Shay Kilby – created an essay titled “Volume Rendering with AMD”. In this essay Shay writes how software paired with AMD Opteron™ 6100 Series processors can be used to change the world one voxel at a time by rendering volume data. Shay provides stunning views of a mummy, and a beetle using Fovia’s High Definition Volume Rendering® (HDVR®).
1. OUR WINNER, Roy Keyes – posted a blog titled “48 Cores to Beat Cancer” which shows how AMD Opteron™ 6100 Series processors can help to enhance radiation cancer treatments.
In the end we could only choose one winner (that’s what the lawyers tell me) -- But we really wanted to acknowledge the talent and originality shown by our two runners-up -- so for Daniel and Shay, we will be offering each of you an unofficial runner-up prize consisting of one of our previous generation 1P processors.
Lastly, I would like to provide honorable mention to two of our other participants. You can imagine after reading thru 800+ of these, when the judges found a few that made us laugh, we had to share.
- Ryan Allen’s essay on the “Best Bacon Plan Ever”
- Chris Well’s video “Magny-Cours, it’s a lot of cores”
Thanks for your submissions, we had a lot of fun with it, and hope you did too. See you again soon!
For official contest rules please click here: http://blogs.amd.com/work/48-cores-official-contest-rules/
Anita Tulsiani is in Product Marketing for Server/Workstation products at AMD. Her postings are her own opinions and may not represent AMD’s positions, strategies or opinions. Links to third party sites are provided for convenience and unless explicitly stated, AMD is not responsible for the contents of such linked sites and no endorsement is implied.
POSTED IN: AMD Opteron
SHARE:
COMMENTS: 16
I’m just going to take the 8600 popsicle sticks I used to make a giant AMD logo and turn them into a giant logo based gaming chair……
http://www.youtube.com/watch?v=bmlGP94gLNs
Congratulations to all of the winners! Best of luck to Roy Keyes on what looks to be excellent work in an area desperate for innovation.
Congratulations to the winners!
(do you have any tips for improving our applications?)
(perhaps “entries” is clearer; I looked at this as a grant application and am interested in improving)
I’d like to see the other contest entries or at least some of the better ones? Is there any way to see them?
I don’t think the applications are particularly visionary. Useful yes, interesting yes. Visionary and forward-looking, not so much.
Comments:
One thing that seems to have been missed is a distinction between 48-cores versus using a cloud or a large server.
There was an emphasis on using the compute power of 48-cores. But any of the winning ideas could have been implemented with existing server technologies or cloud computing is aggregate CPU speed was the main resource that was needed.
Was there an application that could have benefitted from the tighter architectural coupling of a 48-core machine but would not have run satisfactorily on a symmetric multiprocessor system or a cloud computing platform?
We received over 800 entries. We will make some of them available.
Sanjay>But any of the winning ideas could have been implemented with existing server technologies
Not at all; if it is the case then the solutions to render interactively at high quality 2K cube of volumetric data would be readily available and it is definitely not the case. It has become practical with the latest multi-core MIMD architecture progress in the conjunction with development of rendering algorithms utilizing this architecture efficiently.
Such applications are not readily available because until multi-core machines came long, they had to be run on expensive multi-processor servers such as a 64-processor machine.
From a computer power perspective, by and large, it seems that whats happened is that a server that would be about the size of a fridge has now been squeezed into a tower case.
Hence my contention that running a volumetric rendering application on a 48-core machine may be nice, but does not constitute ground breaking innovation, only a lower barrier to entry for those who could not afford big iron before.
Corporate users will use the platform to do what they’ve already been doing. I’m curious about what (serious) non-corporate entries had to say.
There were a lot of people who wanted to run folding@home or render their video in different formats.
My point exactly. Mere variations on a theme.
I thought the goal of the contest was to look for some innovative transformative application of a many-core system that would be used on the desktop or workgroup server.
Hence these are rather banal uses of 48-cores, which have already been done for years.
Aren’t there any more comments from people.
It seemed that once the winner(s) were announced, interest in this dropped like a stone.
Thats a pity. But I’m still hoping to see some of the other interesting entries. No doubt there will be some good ideas in there.
I’ve debated whether or not to reply, and have decided to. I entered this contest through an essay and, clearly, did not win. The custom application I am using, without question, is better run on a shared memory machine instead of a cluster, unfortunately, the application doesn’t exactly save-the-world in a feel good way, therefore, it fails the PR test, but not the technical test. I am a cancer survivor myself, which included radiation therapy, I understand a thing or two about the social value of improving its targeting. In that spirit, I salute the winner.
Ultimately, AMD gets to choose who it wants for the contest and I support them in their decision. I’m disappointed to not have been selected as the winning entry, and I also share your criticisms of how well their applications are differentiated on a shared memory system versus a networked cluster. Like you, I’d be interested in seeing other proposals that may have good technical reasoning, if not great save-the-world PR elements.
OK cool to see- useful comments are always welcome! Peace.
Pingback: Wanted: Opteron Guest Blogger | AMD at Work