Taipei, Wednesday, April 16, 2014 21:16 (GMT+8)
partly cloudy
Taipei
24°C
Nvidia introduces new standard for HPC with Kepler-based Tesla GPUs
Press release, May 16; Joseph Tsai, DIGITIMES [Wednesday 16 May 2012]

Nvidia has unveiled a new family of Tesla GPUs based on the Nvidia Kepler GPU computing architecture, which, according to the company, makes GPU-accelerated computing easier and more accessible for a broader range of high performance computing (HPC) scientific and technical applications.

The new Nvidia Tesla K10 and K20 GPUs are computing accelerators built to handle complex HPC problems. Designed with a focus on high performance and power efficiency, Kepler is three times as efficient as its predecessor, the Nvidia Fermi architecture, which itself established a new standard for parallel computing when introduced two years ago, Nvidia claimed.

Nvidia said it has developed a set of architectural technologies that make the Kepler GPUs high performing and highly energy efficient, as well as more applicable to a wider set of developers and applications including SMX Streaming Multiprocessor, Dynamic Parallelism and Hyper-Q

Optimized for customers in oil and gas exploration and the defense industry, a single Tesla K10 accelerator board features two GK104 Kepler GPUs that deliver an aggregate performance of 4.58-teraflops of peak single-precision floating point and 320GB per second memory bandwidth.

The Nvidia Tesla K20 GPU is the new flagship of the Tesla GPU product family, designed for the most computationally intensive HPC environments. The Tesla K20 is planned to be available in the fourth quarter of 2012.

The Tesla K20 is based on the GK110 Kepler GPU. This GPU delivers three times more double precision compared to Fermi architecture-based Tesla products and it supports the Hyper-Q and dynamic parallelism capabilities.

In addition to the Kepler architecture, Nvidia has also released a preview of the Cuda 5 parallel programming platform. Available to more than 20,000 members of Nvidia's GPU Computing Registered Developer program, the platform will enable developers to begin exploring ways to take advantage of the new Kepler GPUs, including dynamic parallelism.

The Cuda 5 parallel programming model is planned to be widely available in the third quarter of 2012.

Nvidia Tesla K20 GPU accelerator

Nvidia Tesla K20 GPU accelerator
Photo: Company

Realtime news
  • China market: Local smartphone vendors accelerating LTE launches

    Mobile + telecom | 3min ago

  • Taiwan market: Epson launches CISS printers

    IT + CE | 6min ago

  • O-Film expected to produce for 35% of handset touch panels in 2014

    Displays | 56min ago

  • Taiwan solar cell makers see surging orders from China

    Green energy | 59min ago

  • Taiwan market: Epson launches CISS printers

    Before Going to Press | 17min ago

  • FIC ruggedized tablet subsidiary receives procurement orders from Brazil

    Before Going to Press | 41min ago

  • Unitech to ramp up capacity for any-layer HDI and flex-rigid boards

    Before Going to Press | 42min ago

  • China white-box tablet vendors gaining foothold

    Before Going to Press | 1h 18min ago

Pause
 | 
View more
MSI product press release
China smartphone tracker
2014 global smartphone market forecast
  • China flat panel display industry outlook, 2013-2016

    This Digitimes Research Special Report outlines and analyzes the key trends and players that are influencing the continued development of China large-size and small-to medium size TFT LCD panel industry, while providing related growth forecasts for the industry through 2016.

  • TD-LTE market developments and forecast, 2014-2016

    TD-LTE has become the mainstream global choice for the construction of 4G asymmetric frequency spectrum networks. This Digitimes Research Special Report examines the current global status of TD-LTE, and forecasts developments in the TD-LTE supply chain in China and worldwide through 2016.