Go Back   Hardware Canucks > NEWS & REVIEWS > Press Releases & Tech News

    
Reply
 
LinkBack Thread Tools Display Modes
  #1 (permalink)  
Old March 22, 2009, 08:33 AM
omgwtf's Avatar
Hall Of Fame
 
Join Date: Mar 2007
Location: Montreal
Posts: 1,904
Default GPU computing/supercomputers

NVIDIA® Tesla™ GPU computing solutions enable the necessary transition to energy efficient parallel computing power. With 240 cores per processor and based on the revolutionary NVIDIA® CUDA™ parallel computing architecture, Tesla scales to solve the world’s most important computing challenges—more quickly and accurately.


Tesla Architecture
  • Massively-parallel many-core architecture
  • 240 scalar processor cores per GPU
  • Integer, single-precision and double-precision floating point operations
  • Hardware Thread Execution Manager enables thousands of concurrent threads per GPU
  • Parallel shared memory enables processor cores to collaborate on shared information at local cache performance
  • Ultra-fast GPU memory access with 102 GB/s peak bandwidth per GPU
  • IEEE 754 single-precision and double-precision floating point
  • Each Tesla C1060 GPU delivers 933 GFlops Single Precision and 78 GFlops Double Precision performance
Software Development Tools
  • C language compiler, debugger, profiler, and emulation mode for debugging
  • Standard numerical libraries for FFT (Fast Fourier Transform), BLAS (Basic Linear Algebra Subroutines), and CuDPP (CUDA Data Parallel Primitives)
Product Details
  • 3 or 4 Tesla C1060 Computing Processors with 4GB of dedicated memory per GPU
  • 2.33 GHz+ Quad-core AMD Phenom or Opteron, -- OR -- Quad-core Intel Core 2 or Xeon
  • Minimum system memory: 12 GB for 3 Tesla C1060s and 16 GB for 4 Tesla C1060s (at least 4GB per Tesla C1060)
  • 12GB+ system memory (at least 4GB per Tesla C1060)
  • 1200-1350 Watt Power supply
  • Acoustics < 45dbA
Supported Platforms
  • Microsoft® Windows® XP 64-bit and 32-bit (64-bit recommended)
  • Linux® 64-bit and 32-bit (64-bit recommended)
    • Red Hat Enterprise Linux 4 and 5
    • SUSE 10.1, 10.2 and 10.3
Has anyone touched one of these babies? Can't even find a price for them..to satisfy my curiosity.
__________________
Memento mori

Last edited by omgwtf; March 22, 2009 at 08:43 AM.
Reply With Quote
  #2 (permalink)  
Old March 22, 2009, 09:12 AM
BrainEater's Avatar
Hall Of Fame
 
Join Date: Mar 2007
Location: Calgary
Posts: 2,513
Default

the Tesla c870 gpu's (1060's little bro) are about 1500$

A 4 GPU c870 rig will cost you just under ten grand.



edit : here yah go :

Tycrid Platform Technologies Inc. NVIDIA Tesla C1060
__________________
Intel 3930k /rIVe/32 Gb vengeance LP/Nvidia TITAN/760/760/Intel 520's/WD raptors/etc...

Last edited by BrainEater; March 22, 2009 at 09:19 AM.
Reply With Quote
  #3 (permalink)  
Old March 22, 2009, 11:15 AM
Mr. Gruntsworthy's Avatar
Allstar
F@H
 
Join Date: Sep 2008
Location: Courtice, ONT
Posts: 738
Default

I can just imagine folding with that thing... well, it's the exact sort of thing it was developed for, basically... o.o
__________________
MAIN RIG:
Case: Antec 900
CPU : Intel E8400 45nm Core 2 Duo @ 3GHz 1333MHz FSB
RAM : 4GB DDR2 800MHz (OCZ Platinum)
HDD : Main: 120GB SSD SATA2 (OCZ) Storage: 3TB SATA2 (WD Green) Backup: 500GB External USB (Hitatchi)
GFX : HD4850 (512MB)
PSU : 610W PC Power Silencer
MoBo: Asus P5Q SE
View : Acer H213H 21.5" @ 1920 x 1080

Oshawa Deathmatch Community -- An open group for Durham/Toronto Airsoft, Paintball, and Nerf players. Come join!

I am a Level/Game Designer! http://mikeseguin.webs.com/
Reply With Quote
  #4 (permalink)  
Old March 22, 2009, 11:27 AM
ipaine's Avatar
Hall Of Fame
F@H
 
Join Date: Apr 2008
Location: Edmonton, AB
Posts: 2,055

My System Specs

Default

Does anyone know if programs such as ArcGIS, stuff from Remsoft, or any other GIS software would benefit from one of these or does everything have to be coded right from the start to take advantage of their power?
__________________
"Nothing sucks more than that moment during an argument when you realize you're wrong."


Reply With Quote
  #5 (permalink)  
Old March 22, 2009, 02:06 PM
omgwtf's Avatar
Hall Of Fame
 
Join Date: Mar 2007
Location: Montreal
Posts: 1,904
Default

ipaine: The application developer has to modify their application to take the compute-intensive kernels and map them to the GPU. The rest of the application remains on the CPU. Mapping a function to the GPU involves rewriting the function to expose the parallelism in the function and adding “C” keywords to move data to and from the GPU.

GPU computing is enabled by the massively parallel architecture of NVIDIA’s GPUs called the CUDA architecture. The CUDA architecture consists of 100s of processor cores that operate together to crunch through the data set in the application.

My surprising question is, how in the hell do programmers need this kind of power? ..quote: Tesla scales to solve the world’s most important computing challenges: Game physics, graphics, video & audio, programming tools.. Nevertheless it's a beautiful machine and takes it to another level..
__________________
Memento mori

Last edited by omgwtf; March 22, 2009 at 02:38 PM.
Reply With Quote
  #6 (permalink)  
Old March 22, 2009, 02:42 PM
Chilly's Avatar
Hall Of Fame
F@H
 
Join Date: Sep 2007
Location: Ontario
Posts: 2,593
Default

I hope you guys realise these are slightly tweaked, rebadged GeForce's right? Not worth the cost in most cases, like the Quadro's they have driver and hardware level(extra DRAM, etc) optimizations but IMO its *usually* not worth the mark up. Thats not to say that there aren't cases where these(with their optimizations) would be useful over a regular GeForce with CUDA enabled but...
__________________
"Let's face it, we're not changing the world. We're building a product that helps people buy more crap - and watch porn."
Bill Watkins, CEO Seagate Technology

Dual Xeon X5650 @ 4.1GHz | EVGA Classified SR-2
2x4GB G.Skill at 1496MHz (9-9-9-24) | XFX 4890 XXX 1GB at 1000/1000MHz
Intel SSD X25-M 80GB Gen2 | Western Digital 1TB Black | SilverStone Zeus 1200W
Sound Blaster X-Fi XtremeMusic | Sony MDR-V700 | Windows Seven x64


Stuff for Sale! You must buy it, you can't look away!
Reply With Quote
  #7 (permalink)  
Old March 22, 2009, 06:46 PM
Top Prospect
 
Join Date: Dec 2007
Posts: 76
Default

Quote:
Originally Posted by Chilly View Post
I hope you guys realise these are slightly tweaked, rebadged GeForce's right? Not worth the cost in most cases, like the Quadro's they have driver and hardware level(extra DRAM, etc) optimizations but IMO its *usually* not worth the mark up. Thats not to say that there aren't cases where these(with their optimizations) would be useful over a regular GeForce with CUDA enabled but...
The Quadros are slightly more different in that they can output higher-bit colors. Up to 12-bit on the new C-class.

These, however, are basically Quadros without the extra graphics features, and with several Cuda optimizations.
Reply With Quote
  #8 (permalink)  
Old March 22, 2009, 07:22 PM
Chilly's Avatar
Hall Of Fame
F@H
 
Join Date: Sep 2007
Location: Ontario
Posts: 2,593
Default

Quote:
Originally Posted by Salavat23 View Post
The Quadros are slightly more different in that they can output higher-bit colors. Up to 12-bit on the new C-class.

These, however, are basically Quadros without the extra graphics features, and with several Cuda optimizations.
You see, thats the thing though, the Quadros often have a few features over the regular GeForce's(Ram, optimized hardware, driver tweaks, etc) that have proven them a must have, regardless of their mark up, for studios.

I'm not quite convinced with the Tesla, that those few optimizations are a "must have" as opposed to a regular GeForce with CUDA enabled. I guess what I'm trying to say is that while the Tesla may be worth it to a few development houses out there, its yet to be proven that its "worth" the extra cost, as with any new product I guess.
__________________
"Let's face it, we're not changing the world. We're building a product that helps people buy more crap - and watch porn."
Bill Watkins, CEO Seagate Technology

Dual Xeon X5650 @ 4.1GHz | EVGA Classified SR-2
2x4GB G.Skill at 1496MHz (9-9-9-24) | XFX 4890 XXX 1GB at 1000/1000MHz
Intel SSD X25-M 80GB Gen2 | Western Digital 1TB Black | SilverStone Zeus 1200W
Sound Blaster X-Fi XtremeMusic | Sony MDR-V700 | Windows Seven x64


Stuff for Sale! You must buy it, you can't look away!
Reply With Quote
  #9 (permalink)  
Old March 22, 2009, 07:31 PM
Top Prospect
 
Join Date: Dec 2007
Posts: 76
Default

I remember the Telsas were all over the news over a year ago. There would be three versions out supposedly.

The main benefit of them is the ability to scale them to multitudes of units I believe.
Reply With Quote
  #10 (permalink)  
Old March 25, 2009, 08:54 PM
MVP
 
Join Date: Apr 2007
Location: Burnaby, BC
Posts: 252

My System Specs

Default

Unfortunately, only a certain subset of applications will run on this. Many applications still require the general purpose nature of a CPU.
Reply With Quote
Reply


Thread Tools
Display Modes

Similar Threads
Thread Thread Starter Forum Replies Last Post
Hi guys,there is a great conference about mobile computing-ljh leijinghang Press Releases & Tech News 4 December 11, 2008 02:43 PM
AMD Announces Stream Computing Will Be Mainstream yamawho O/S's, Drivers & General Software 8 November 14, 2008 11:16 AM
Distributed Computing Histories..... BrainEater HardwareCanucks F@H Team 9 September 10, 2008 02:16 PM
mobile computing forum... Mr. Friendly Mobile Computing 8 January 1, 2008 12:07 PM
Distributed computing. SHaCK Suggestions & Feedback 5 February 28, 2007 11:59 AM