Feeds

GPU-CPU hybrids promise the moon*

*Code headaches not included

Build a business case: developing custom apps

HPC Blog I've been busy all day talking, listening, and maybe even learning a thing or two at the 2010 GPU Tech Conference. The speaker from the NOAA session (topic of another blog) put the move toward GPUs into perspective toward the end of his talk today with two key points.

His first point was that big HPC advances tend to come at around 10-year intervals. Citing the moves from vector processing to MPP to COTS (Common Off The Shelf systems) to GPU-hybrid systems as evidence, he pointed out that each jump required significant changes in mindset, tools, management, and code. However, the benefits from each jump were profound, with cost savings and performance increases on the order of 20x to 50x.

To illustrate this, he compared the current state-of-the-art 2.3 PFlop DOE Jaguar system to what a 1 PFlop GPU-based system would look like. The difference is amazing, to wit:

GPU Alternative Performance 2.3 PFlop/s 1.0 PFlop/s
Cores 250,000 CPU cores 1,000 Fermi GPUs
Size 284 cabinets 10 cabinets
Power draw 7-10 MW 0.5 MW
Cost $50-$100 million $5-$10 million
Reliability measured in hours measured in weeks

By every measure, GPU + CPU hybrids are a better deal in terms of price/performance, facilities, availability, and even the initial price tag for the hardware. What this table doesn't take into account is the time, labor, and heartache involved in making code "GPU-riffic." (Yeah, that's my own term for it; hoping it catches on, but doubt it will.) This task is getting easier over time as the ecosystem around CUDA becomes more robust and varied. ®

Boost IT visibility and business value

More from The Register

next story
The Return of BSOD: Does ANYONE trust Microsoft patches?
Sysadmins, you're either fighting fires or seen as incompetents now
Microsoft: Azure isn't ready for biz-critical apps … yet
Microsoft will move its own IT to the cloud to avoid $200m server bill
Shoot-em-up: Sony Online Entertainment hit by 'large scale DDoS attack'
Games disrupted as firm struggles to control network
Cutting cancer rates: Data, models and a happy ending?
How surgery might be making cancer prognoses worse
Silicon Valley jolted by magnitude 6.1 quake – its biggest in 25 years
Did the earth move for you at VMworld – oh, OK. It just did. A lot
VMware's high-wire balancing act: EVO might drag us ALL down
Get it right, EMC, or there'll be STORAGE CIVIL WAR. Mark my words
prev story

Whitepapers

Implementing global e-invoicing with guaranteed legal certainty
Explaining the role local tax compliance plays in successful supply chain management and e-business and how leading global brands are addressing this.
Endpoint data privacy in the cloud is easier than you think
Innovations in encryption and storage resolve issues of data privacy and key requirements for companies to look for in a solution.
Scale data protection with your virtual environment
To scale at the rate of virtualization growth, data protection solutions need to adopt new capabilities and simplify current features.
Boost IT visibility and business value
How building a great service catalog relieves pressure points and demonstrates the value of IT service management.
High Performance for All
While HPC is not new, it has traditionally been seen as a specialist area – is it now geared up to meet more mainstream requirements?