• Welcome to TechPowerUp Forums, Guest! Please check out our forum guidelines for info related to our community.

Cray Builds Supercomputer Blades with Tesla 20 Series GPU Compute Processors

btarunr

Editor & Senior Moderator
Staff member
Joined
Oct 9, 2007
Messages
47,884 (7.38/day)
Location
Dublin, Ireland
System Name RBMK-1000
Processor AMD Ryzen 7 5700G
Motherboard Gigabyte B550 AORUS Elite V2
Cooling DeepCool Gammax L240 V2
Memory 2x 16GB DDR4-3200
Video Card(s) Galax RTX 4070 Ti EX
Storage Samsung 990 1TB
Display(s) BenQ 1440p 60 Hz 27-inch
Case Corsair Carbide 100R
Audio Device(s) ASUS SupremeFX S1220A
Power Supply Cooler Master MWE Gold 650W
Mouse ASUS ROG Strix Impact
Keyboard Gamdias Hermes E2
Software Windows 11 Pro
NVIDIA's Tesla 20 series GPU compute processors have made their way into CRAY's latest supercomputer, the XE6. In these, Tesla units are installed into blades, which are networked using Cray's fast Gemini system-interconnect increasing the throughput, or efficiency of these GPUs in HPC applications. "The combination of new Gemini system interconnect - paired with NVIDIA's Tesla - will provide XE6 a powerful combination of scalability and production-quality, GPU-based high performance computing (HPC) in a single system," said Cray VP Barry Bolding. According to him, the supercomputing giant's move to adopt Tesla into its blade systems (which populate the high-end segment) is after seeing the technology mature on the company's mid-range and deskside systems. Bolding noted that the company will collaborate further with NVIDIA in advancing GPU compute processors for HPC applications.



View at TechPowerUp Main Site
 
Cray seems to be a bit behind the curve given yesterday's news. Shouldnt they be using Fermi, and not Tesla units, for the blades? And shouldn't they also say with the announcement that Kepler blades will be available from Q2 2011? ie. buy Cray backbone today, test it with a couple of Fermi blades during 2010, and then upgrade and expand with Kepler next year?

139a.jpg
 
Oh, they are using Fermi. It's just that someone at Nvidia likes to create confusion with the names. Tesla means two different things

- Tesla (as seen in the chart) is the compute architecture inside GT200 (and maybe in G80).

- Tesla as mentioned in the article is the family of GPUs dedicated to HPC, just like geforce is for gaming cards and Quadro is the professional graphics card.
 
dayum that is going to be one hella frikken powerfull supercomputer. whoever buys these should be pretty pleased with next gen Tesla GPU's and so on.

folders dream machine :P
 
I'm waiting for someone to make that one statement. We all know what that is ;)
 
^Post it for an infraction.
 
Can it run Minesweeper?
 
I just can't take this thread seriously with these post's :roll:
 
Back
Top