Heck what are those 8 cards?

Any specs?

Recycled 295s......LOL.

---

"Freedom is just Chaos, with better lighting." Alan Dean Foster

Heck what are those 8 cards?

Any specs?

From what I gather 50 core x86 cards running at 1.8Ghz based on 40nm fabrication.

The promised release ones are meant to be 80 core based upon 22nm fabrication.

Thinking about what a tesla card cost`s,
One thing is certain, they will put a big hole in anyone`s credit card.
and that setup must be on a PCIe expansion board, which cost a fair bit,
There is a heck of a lot of dollars worth in that photo.

Oh yeah, that's sounds about right, the pic has unobtainium in It, I guess, for Me at least. But I have more important problems right now...

Looks like each one of those boars has an 8pin and a 6pin going to it. For at least 225w to each card, or 1800w of 12v goodness.

The next question is who makes that power configuration/chassis. Could be a custom Intel job for show, but it is a good setup for the massive GPU guys as well.

Edit: Here is the whole article where that image came from.
http://www.thinq.co.uk/2011/6/20/intel-pushes-hpc-space-knights-corner

Those are 8-45nm Knights Ferry cards, Knights Corner is supposed to be I guess 22nm in size, among other changes that is.

---

The T1 Trust, PRR T1 Class 4-4-4-4 #5550, 1 of America's First HST's

Before we know it Who? will be back, lol

Claggy

Before we know it Who? will be back, lol

Claggy

By then Who? Might be a Doctor... ;)

---

The T1 Trust, PRR T1 Class 4-4-4-4 #5550, 1 of America's First HST's

Before we know it Who? will be back, lol

Claggy

By then Who? Might be a Doctor... ;)

Last I heard, he was working on a v12 Jaguar........
I think he was using the handle v12extreme.

---

"Freedom is just Chaos, with better lighting." Alan Dean Foster

Before we know it Who? will be back, lol

Claggy

By then Who? Might be a Doctor... ;)

Last I heard, he was working on a v12 Jaguar........
I think he was using the handle v12extreme.

What kind of RAC does a V12 Jaguar have?

---

Before we know it Who? will be back, lol

Claggy

By then Who? Might be a Doctor... ;)

Last I heard, he was working on a v12 Jaguar........
I think he was using the handle v12extreme.

What kind of RAC does a V12 Jaguar have?

It uses British made Unobtainium, that's enough for Me.

---

The T1 Trust, PRR T1 Class 4-4-4-4 #5550, 1 of America's First HST's

Before we know it Who? will be back, lol

Claggy

By then Who? Might be a Doctor... ;)

Last I heard, he was working on a v12 Jaguar........
I think he was using the handle v12extreme.

What kind of RAC does a V12 Jaguar have?

It uses British made Unobtainium, that's enough for Me.

I had bookmarked the site, but I've been though a couple of hard drive crashes since then and lost a lot.

I'll come up with something...he was a PR man for Intel...should be googleable somewhere.

I think his name was Francois.


EDIT...

Francois Piednoel

---

"Freedom is just Chaos, with better lighting." Alan Dean Foster

Over a year ago, I thought the article Compilers and More: Knights Ferry Versus Fermi was a well-balanced comparison. Some details have changed since, but it still looks like a good overview to me.

                                                                   Joe

I honestly think this new processor would be a complete waste to use for SETI@Home or any other distributed computing application. It's not even useful as a server CPU. I don't see any uses for it, really.

I honestly think this new processor would be a complete waste to use for SETI@Home or any other distributed computing application. It's not even useful as a server CPU. I don't see any uses for it, really.

Can you explain further? Why do you say that?


Happy crunchin',
Martin

---

See new freedom: Mageia Linux
Take a look for yourself: Linux Format
The Future is what We all make IT (GPLv3)

I honestly think this new processor would be a complete waste to use for SETI@Home or any other distributed computing application. It's not even useful as a server CPU. I don't see any uses for it, really.

That would be a bit premature, Until someone makes an app, We'll never know, so I'll reserve judgement, as It's too early for saying that.

---

The T1 Trust, PRR T1 Class 4-4-4-4 #5550, 1 of America's First HST's

It's not even useful as a server CPU.

Niether are GPUs, yet they can do considerably more processing than a server CPU can.

---

Grant
Darwin NT

This looks like the Tyan system that I paid for that they took the money and ran. It was going to have 8 590s with 3 power supplies. About 25k worth. I would have loved to see what the RAC would have been on that.

---

So from what I gather its basically 50 Pentium cpu's on a card. In order to make use of it you would have to either:
1. Run 50 wu at a time
or
2. The app would need to multi-thread (and run however many instances / number of threads).

The 2nd case would require the cpu apps to be become multi-threaded to take advantage of all those cores.

There is no mention of it being OpenCL capable.

One of the articles mentioned a clock speed of 1.8Ghz so they aren't that fast either, but then if you can use all 50 threads its not so important.

How difficult would it be get Astropulse or Multibeam to multi-thread I wonder.

So from what I gather its basically 50 Pentium cpu's on a card. In order to make use of it you would have to either:
1. Run 50 wu at a time
or
2. The app would need to multi-thread (and run however many instances / number of threads).

The 2nd case would require the cpu apps to be become multi-threaded to take advantage of all those cores.

There is no mention of it being OpenCL capable.

One of the articles mentioned a clock speed of 1.8Ghz so they aren't that fast either, but then if you can use all 50 threads its not so important.

How difficult would it be get Astropulse or Multibeam to multi-thread I wonder.

Since the standard CPUs are able to do OpenCL I would hope that these coprocessors would also be able to run it, but we may not know for a little while.

Maybe the apps can be done in a way so the tasks will be sent to the card and it splits it up and it thinks best. Like we see with the GPUs now.

Running 50 at a time might see interesting results in how saturated the PCIe bus gets.

---

SETI@home classic workunits: 93,865 CPU time: 863,447 hours
Join the [url=http://tinyurl.com/8y46zvu]BP6/VP6 User Group[

Maybe the apps can be done in a way so the tasks will be sent to the card and it splits it up and it thinks best. Like we see with the GPUs now.

+1 for heterogeneous processing...

---

"Living by the wisdom of computer science doesn't sound so bad after all. And unlike most advice, it's backed up by proofs." -- Algorithms to live by: The computer science of human decisions.

1.8GHz is higher than I've seen. 1.2 more like it. These sort of speeds may be chosen to keep the heat down. How does the above system cope with heat dissipation if there's no space between the cards? Again I may have answered my own question here...

How does the above system cope with heat dissipation if there's no space between the cards?

Checkout the image earlier in the thread from Wembley.
Like most rack mount systems air is drawn in from the front & expelled out the rear.

---

Grant
Darwin NT