I wonder, if this was elaborate reverse engeneering of hn crowd to reach Parallella financing target :)
Parallella: A Supercomputer For Everyone is Dying
61–67 of 67 posts
Re: Parallella: A Supercomputer For Everyone is Dying
#62Well, they made it and more:
Funding: $788,138 of $750,000
Re: Parallella: A Supercomputer For Everyone is Dying
#63At some point in the next 24 hours, the backers who signed up for one unit will need to ask themselves if they would rather have two units or zero. I can't speak for the other 1800 people in my bin, but I just decided on two.
Isn't that (multiples as pledge awards) a direct violation of the letter and spirit of kickstarters new rules?
Re: Parallella: A Supercomputer For Everyone is Dying
#64Re: Parallella: A Supercomputer For Everyone is Dying
#65I did my part. I'm very much the archetypal broke college student at the moment, but I won't always be. I have big plans in the Artificial Intelligence and Machine Learning sectors, and I can't imagine a better, cheaper solution to get started on working with multi-agent systems. I desperately want to see this sort of pricing for cluster computing available in the future, when I have the scratch and knowledge necessa…
Here is one cheaper solution: simulate a multi agent system on your laptop. You don't need parallel hardware to ave multiple agents.
Re: Parallella: A Supercomputer For Everyone is Dying
#66Earlier quoted context omitted.
And there's the problem: my personal bias from years and years of GPU programming is that I'd rather target 4 cores with 16-way SIMD than 64 cores each with scalar, or to quote Seymour Cray - "If you were plowing a field, which would you rather use: Two strong oxen or 1024 chickens?" Besides, this is 28 nm technology and 15x15 mm, no? That's 225 mm^2. AMD's 28 nm Tahiti is 365 mm^2 with 4.3B transistors, making this…
> I'd rather target 4 cores with 16-way SIMD than 64 cores each with scalar You're assuming problems that are suitable for SIMD. If you have problems suitable for SIMD, use a GPU. Lots of problems are NOT suitable for SIMD. If those 64 data streams all happen to require branches regularly, for example, your 4x 16-way SIMD is going to be fucked. > Besides, this is 28 nm technology and 15x15 mm, no? Where did you get t…
Mea culpa on the die size...
That said, this is where the R&D done by AMD and NVIDIA have expanded what is amenable to running on a GPU. Specifically, instructions like vote and fast atomic ops can alleviate a lot of branching in algorithms that would otherwise be divergent. It's not a panacea, but it works surprisingly well, and it's causing the universe of algorithms that run well on GPUs to grow IMO.
What I worry about with Parallela is that by having only scalar cores, and lots of them, it has solved issues with branch divergence in exchange for potential collisions reading from and writing data to memory. The ideal balance of SIMD width versus cores count is a question AMD, Intel, and Nvidia are all investigating right now. But again, ~26M transistors - no room for SIMD...
Re: Parallella: A Supercomputer For Everyone is Dying
#67Earlier quoted context omitted.
And there's the problem: my personal bias from years and years of GPU programming is that I'd rather target 4 cores with 16-way SIMD than 64 cores each with scalar, or to quote Seymour Cray - "If you were plowing a field, which would you rather use: Two strong oxen or 1024 chickens?" Besides, this is 28 nm technology and 15x15 mm, no? That's 225 mm^2. AMD's 28 nm Tahiti is 365 mm^2 with 4.3B transistors, making this…
Why plow a field with 1024 chickens, when you can plow it with 1M worms ? The GA144's F18 core has ~20 thousand transistors, and is asynchronous , and if you make the die size the size of an Opteron, and if you wait until you can pack 20B transistors on a die, you get---one million---cores.
That chip would so cool if only its native internal representation were 32-bit... Sigh...
But it's way better than this monstrosity: http://web.media.mit.edu/~bates/Summary_files/BatesTalk.pdf