Parallella: A Supercomputer For Everyone is Dying (opens in new tab)

With 31 hours to go. If they don't raise $138,000 in 31 hours they get squat.

The title suggests they get nothing if they don't raise the $750,000 target, which is true.

Nearly $618k as of 16 minutes after your post...

EDIT: $624,602 44 minutes after your post.... I'm getting more optimistic.

According to kicktraq, they're indeed trending towards not making it (trending to 651k, 86% of goal):

http://www.kicktraq.com/projects/adapteva/parallella-a-super...

They only have 31 hours left to raise the rest of the money. Considering they fact that they took several weeks to reach this point it seems unlikely that they will hit the goal, and therefore they will not secure any of the pledged funds.

I've done my master thesis on GPGPU, so maybe I can help out a bit. I'm not yet too familiar with Epiphany's design however. From what I could grasp what sets them apart the most is a different memory architecture compared to multicore CPUs, where the individual cores seem to be optimized for accessing adjacent memory locations as well as the locations of the direct neighbors. This is one point where the architecture seems to be similar to GPUs, although GPUs have a very different memory architecture again - for the programmer it might look similar however, especially when using OpenCL.

The main point where Epiphany is diverging from GPUs is that the individual cores are complete RISC environments. This could mainly be a big plus when it comes to branching and subprocedure calls (although NVIDIA is catching up on the later point with Kepler 2). On GPUs the kernel subprocedures currently all need to be inlined and branches mean that the cores that aren't executing the current branch are just sleeping - Epiphany cores seem to be more independent in that regard. I still expect an efficient programming model to be along the same lines as CUDA/OpenCL for epiphany however - which is a good thing btw., this model has been very successful in the high performance community and it's actually quite easy to understand - much easier than cache optimizing for CPU for example.

If we compare epiphany to CPU what's mainly missing is the CPU's cache architecture, hyperthreading, long pipelines per core, SSE on each core, possibly out-of-order and intricate branch prediction (not sure on those last ones). The missing caches might be a bit of a problem. The memory bandwidth they specify seems pretty good to me, but from personal experience I'd add another 20-30% to the achievable bandwidth if you have a good cache (which GPU has since Fermi for example). The other simplifications I actually like a lot - to me it makes much more sense to have a massive parallel system where you can just specify everything as scalar instead of doing all the SSE and hyperthreading hoops like on CPUs - optimizing for CPU is quite a pain compared to those new models.

It's closer to regular multi-core CPU computing than GPU computing. It's general purpose cores.

What sets it apart is that the cores are tiny, with little per-core memory (though all cores can transparently access each-others memory as well as main memory), and so the architecture is well suited for scaling up the number of cores with quite low power consumption.

So for problems that can be parallelized reasonably well, but with more complex data dependencies than what a GPU is good for, this might be a good fit.

I'd put it somewhere in the middle between GPU's (for embarrassingly parallel tasks) and general purpose CPU's with high throughput per core.

Also, this looks like it'd be possible to fit in the power envelope of really small embedded systems, like e.g. cellphones and tablets....

Before more developers have these systems, it'll be hard to say how useful they'll be, but the architecture looks exciting.

That's why I supported it - I really want to see how this type of architecture can be exploited, and whether or not it'll prove to be cost effective and/or simpler to work with than GPU's for the right type of problems.

IMO this combines some of the worse features of Cell (e.g. local memory and DMA) and GPUs, and while the power efficiency is good the absolute performance is very low. For a parallel noob who's using OpenMP/OpenCL I don't think it's any better than a desktop PC because programming it is going to feel the same and performance is going to be equal or lower. And if you don't use the libraries then you're in low-level ninjas-only land — the extremely simple and flexible hardware is good in theory because you can use it many different ways, but it also doesn't help you or give any hints about how to properly exploit it.

Typical multicore CPUs don't have nearly as many cores as Parallella. Also, from what I can from www.apteva.com/introduction, the power consumption is much lower and the interconnect is different. In Parallella, cores are laid out in a grid and cores can only talk directly to their neighbors.

It's not so much that it's a powerful computer, but a computer architecture that can scale up to be a very powerful system. The version they're trying to fund is a cost reduced version including their 16 core chip. They also have a 64 core chip, and plan to scale it much higher.

It's differentiated from GPU's in that each core is a simple but fully independent CPU core, with direct access to main system memory AND to the memory of the other cores.

This current project is most interesting as a means for people to start playing with the architecture rather than for the raw performance.

I'm embarassed to say the same. I had forgotten about this Kickstarter, so I thought this was some sort of MRI-supercomputer backup plan, like cryostasis.

Personally, I'd prefer to blame my mental auto-correct. So used to seeing poor / simple grammar mistakes on the internet, I'm in the habit of simply reading in the missing words. In this case, the title wasn't "Parallela: A Supercomputer for Everyone Who is Dying". Which makes this less interesting, but at least this could be real. And from the looks of it, it probably will be.

Isn't that (multiples as pledge awards) a direct violation of the letter and spirit of kickstarters new rules?

And no big project like this would ever fail by < $50,000. Wouldn't the founders or their investors or whatever just max their credit cards to see it through?

Does Kickstarter explicitly prohibit such things?

You were hoping for what exactly? VHDL/Verilog for the chips? Netlists?

For most people I'd assume the main thing is that the architecture is well documented and open, as well as the board, and they have released all of the architecture documentation and a lot of other material.

As much as it'd be great to have a market in other sources for the chips, unless/until the architecture has some traction that is pretty irrelevant.

GreenArrays are intriguing, but a completely different animal from this computer. The GreenArrays compute nodes are microscopic by comparison. Think 256 bytes of storage, shared with instructions and data. If you can map your problem on to them they seem very efficient.

They want to enable small scale computers to do more powerful computations, i don't think it is directly for ipad's and chromebooks, it could be useful for something like a quadcopter where there is a algorithm that gives better stability but needs more computational power.

They're actually providing much less computing power than even a low-end PC.

I think they will succeed.

They're selling something that's really cool and a lot of people want to see succeed, but for which there aren't any software applications to take advantage of yet.

That's the type of project where a lot of people could donate at the end so that it succeeds. Opposed to say, a game, where there's a ton at the beginning and than it slowly trails off.

I think they'll make it easily, sort of like what happened to Dalton Caldwell's App.net ... they got a ton of support in the last 12-24 hours.

For what it's worth, this is the first I've heard of it, and this convinced me to contribute.

Here is one cheaper solution: simulate a multi agent system on your laptop. You don't need parallel hardware to ave multiple agents.

Huh? How does this have anything to do with EC2?

Parallella is an dual core ARM board with an FPGA and a 16 core Epiphany CPU (full general purpose CPU cores with 32k static RAM built into each core - all the cores can access the memory of all other cores as well as system RAM). 1GB RAM total. Expected size around a credit card or so.

The main purpose is the Epiphany CPU, which they also have a 64 core version of. Their problem is that their current CPU's are produced using a process that gives them very low yields and very high per-CPU cost. The main goal of the bounty is to enable them to switch to a much higher yield process and bring the per-chip cost of the 16 core version down to a few dollars per chip.

Their long term roadmap is boards with 1000+ cores.

Yeah, the cloud is much more accessible "for everyone" than this chip. If you want to learn parallel programming rent a GPU instance.

Video games is exactly where heterogenous computers like this have flourished. But they don't have a game to market it yet - historically it hasn't even mattered much if that initial game used the platform anywhere near well. How did the Ouya sell, for instance? (It really resembles a Nexus 7 with a broken screen.)