Thursday, March 30, 2017
Search
  
Submit your own News for
inclusion in our Site.
Click here...
Breaking News
Optimizations Promise Up to 30 Percent Performance Uplift in Ashes of the Singularity DirectX 12 Benchmark With AMD Ryzen
SpaceX To Launch First Reused Rocket
A Look At Samsung S8's Biometric Technologies
Semiconductor Market Showed Recovery Signs In 2016
Toshiba Investors Approve Memory Chip Sale
Micron's 1xnm DRAM Process Has Low Yields, Says Report
Windows 10 Creators Update Coming April 11
Samsung Galaxy S8 and S8+ Are Official, Along With Bixby And New Gear 360
Active Discussions
Which of these DVD media are the best, most durable?
How to back up a PS2 DL game
Copy a protected DVD?
roxio issues with xp pro
Help make DVDInfoPro better with dvdinfomantis!!!
menu making
Optiarc AD-7260S review
cdrw trouble
 Home > News > General Computing > Improvi...
Last 7 Days News : SU MO TU WE TH FR SA All News

Friday, June 21, 2013
Improving Power and Programming Are Key Factors For Advancing To Exascale Computing


Key challenges the industry needs to overcome in order to reach exascale computing by the end of this decade are power and programming, according to Nvidia.

Theoretically, an exascale system - 100 times more computing capability than today's fastest systems - could be built with only x86 processors, but it would require as much as 2 gigawatts of power - the entire output of the Hoover Dam.

On the other hand, the GPUs in an exascale system built with NVIDIA Kepler K20 processors would consume about 150 megawatts. So, a hybrid system that efficiently utilizes CPUs with higher-performance GPU accelerators is the best bet to tackle the power problem.

However, the ndustry needs to look for power efficiencies in other areas.

Speaking at this week's 2013 International Supercomputing Conference (ISC) in Leipzig, Germany, NVIDIA's Chief Scientist, Bill Dally, said that reaching exascale, would require a 25x improvement in energy efficiency ? 50 gigaflops per watt vs. the 2 gigaflops per watt from today's most efficient systems.

And manufacturing process advances alone will not achieve this goal. At best, this will only deliver about a 2.2x improvement in performance per watt, leaving an energy efficiency gap of 12x that will need to be reached by other means.

Dally believes that a combination of more efficient circuit design and better processor architectures can help close the gap ? delivering 3x and 4x improvements in performance per watt, respectively.

Dally?s engineering team at NVIDIA is exploring a number of new approaches, including utilizing hierarchical register files, two-level scheduling, optimizing temporal SIMT, and other advanced techniques - all designed to maximize energy efficiency in every way possible.

Dally says that second big challenge to overcome is making it easier for developers to program these large-scale systems.

Although parallel computing is not hard - at least in Dally's view - programmers, programming tools and the architecture each need to 'play their positions.' For example, programmers should focus on designing better algorithms - and not worry about optimization or mapping. Leave that to the programming tools, which are much more effective at these types of tasks than humans.

And the architecture - well, it just needs to provide the underlying compute power, and otherwise "stay out of the way."

On top of this, Dally notes that tools and programming models need to continue improving, making it even easier for programmers to maximize both performance and energy efficiency.

Potential improvements Dally is investigating in this area include using collection-oriented programming methods, which will continue to make the process of programing large-scale machines quicker and easier.

By focusing on these areas, Dally believes that exascale computing is within our reach by the end of the decade.


Previous
Next
Facebook Bug Exposes Emails, Phone Numbers Of 6 million Facebook users        All News        Nano Research Technique Allows Petabyte Storage On A DVD
Facebook Bug Exposes Emails, Phone Numbers Of 6 million Facebook users     General Computing News      Apple Prevails In Legal Battle Against Samsung in Tokyo

Get RSS feed Easy Print E-Mail this Message

Related News
Prepare For Battle GeForce GTX Bundle Includes For Honor or Tom Clancy's Ghost Recon Wildlands Games
Nvidia Chips To Power Paccar Driverless Trucks
NVIDIA Partners with Bosch For Self-driving System Based on The DRIVE PX Xavier Platform
New Nvidia Game Ready Driver Boosts GeForce DirectX 12 Performance
NVIDIA Jetson TX2 Credit card-sized Platform Released
GeForce GTX 1080 Ti, Nvidia's fastest Card Yet, Arrives At $699
Nvidia, and AMD Increase GPU Attach Rates While Total GPU Shipments Remain Flat
Nvidia Reports Record Revenue Driven by Gaming GPUs Datacenter And Tegra
NVIDIA Powers Supercomputing Workstations with New Quadro GP100, Announce Quadro Pascal Refresh
Nvidia Announces New Ubisoft Game Bundles With GTX 1070, GTX 1080 Cards
NVIDIA Announces the G-SYNC HDR Technology
CES: Nvidia Announces New SHIELD TV, GeForce NOW service, AI Car Projects

Most Popular News
 
Home | News | All News | Reviews | Articles | Guides | Download | Expert Area | Forum | Site Info
Site best viewed at 1024x768+ - CDRINFO.COM 1998-2017 - All rights reserved -
Privacy policy - Contact Us .