Sign up for our newsletter and get the latest HPC news and analysis.

Is AMD Back in HPC? New FirePro GPU Does 2.53 Tflops

amd-firepro-s9150-server-graphics

Today AMD announced their new FirePro S9150 GPU with 2.52 Teraflops of double-precision performance and a maximum power consumption of 235 watts. “AMD FirePro S9150 ushers in a new era of supercomputing. Its memory configuration, compute capabilities and performance per watt are unmatched in its class, and can help take supercomputers to the next level of performance and energy efficiency.”

Video: The Evolution of HPC

roadmap

Marc Hamilton presented this talk at the recent GTC Conference in Japan. “The next stop for GP-GPU computing is the exciting new field of machine learning for big data analytics. Using Convolution Neural Networks running on GPUs, it is now for the first time practical to process the huge amounts of unstructured text, voice, image, and video data available on the web to perform classification, clustering, regression, and recommendation.”

ArrayFire GPU Software Library Moving to Open Source

John Melonakos

Over at the Not Only Luck blog, John Melonakos describes why his company has chosen to open source ArrayFire GPU and accelerator software tools.

Video: An Introduction to OpenACC

people-bronson-bronson_seattle

In this video, Bronson Messer from ORNL presents: An Introduction to OpenACC. “OpenACC is gaining momentum and adoption,” said Duncan Poole, President of the OpenACC Standards Group. “Developers benefit because using OpenACC directives makes parallel programming more productive and collaboration easier. Large, legacy codes are easier to maintain and accelerated code is more portable across HPC systems.”

Nvidia Salutes Women Who CUDA

cuda2

This week Nvidia salutes Women who use CUDA for incredible science and engineering. They’ve compiled 30 profiles so far, and the advice they share from their experiences is quite inspiring. “It’s a good way to remind people that women write code, participate in open-source projects, and invent things,” said Lorena Barba from George Washington University. “It’s important to make the technology world more attractive to female students and show them examples of women who are innovators.”

Porting to Accelerators with Hybrid Fortran

Hybrid Fortran unified code: Separate loop structures for CPU and GPU, automatic code conversion to higher dimensions when needed.

Over at Typhoon Computing, Michel Müller writes programmers looking to port their code to accelerators now have a new tool called Hybrid Fortran. “This python-based preprocessor parses annotations together with your Fortran code structure, declarations, accessors and procedure calls, and then writes separate versions of your code – once for CPU with OpenMP parallelization and once for GPU with CUDA Fortran.”

HPCAC Spain Conference Publishes Agenda for Sept. 24

hpcac_2014spain_thumb_150px (1)

The HPC Advisory Council has published the preliminary agenda for the 3rd Annual HPCAC Spain Conference. The event takes place Sept. 24 at the University of Cantabria in Santander, Spain.

Can Nvidia Outsmart Intel in HPC?

harvey

Over at the Motley Fool, Srdjan Bejakovic writes that Nvidia and Intel are battling it out in the HPC space. But before you go off and invest, insideHPC wants to make sure you know the rest of the story.

Increasing the Efficiency of Your GPU-enabled Cluster with rCUDA

rcud

“The use of GPUs to accelerate applications is mainstream nowadays, but their adoption in current high performance computing clusters is primarily impaired by the trend of including accelerators in all the nodes of the cluster, as this presents several drawbacks related with increased costs. In this talk we introduce the remote GPU virtualization mechanism, intended to address the drawbacks of GPU computing. The rCUDA remote GPU virtualization framework will also be presented.”

Farber to Teach All-Day Tutorial At SC14

Farber-SC14

Over at TechEnablement, Rob Farber writes that he will be teaching an all-day tutorial “From ‘Hello World’ to Exascale Using x86, GPUs and Intel Xeon Phi Coprocessors” (tut106s1) at SC14 in New Orleans. The tutorial takes place on Sunday November 16, 2014.