HPCwire: AccelerEyes Upgrades Jacket Software for GPU Computing

Since 1986 - Covering the Fastest Computers
in the World and the People Who Run Them

Translation Disclaimer

Subscribe | Sign In

Visit additional Tabor Communication Publications

Leading HPC
Solution Providers

April 06, 2010

AccelerEyes Upgrades Jacket Software for GPU Computing

Enhanced double precision functionality and preparing for Fermi

ATLANTA, April 5 -- AccelerEyes today announced a new version of its Jacket GPU computing software that integrates double precision linear algebra, is easier to setup, performs faster, and is less expensive. This new Jacket version 1.3 has been designed and developed with tremendous participation of AccelerEyes' customers and prospctive customers who shaped the product feature set to meet their needs and requirements.

AccelerEyes develops Jacket, a software platform that delivers GPU computing power to desktop users of MATLAB and other very high level languages. It enables faster prototyping and problem solving across a range of government, manufacturing, energy, media, biomedical, financial, and scientific research applications. The new Jacket 1.3 integrates the ability for Jacket users to perform double precision linear algebra functionality just in tme for new NVIDIA hardware based on the FERMI architecture. To obtain this LAPACK functionality for GPUs, Jacket 1.3 users will need to license the Jacket DLA add-on for a couple of hundred US dollars.

AccelerEyes has partnered with EM Photonics to extend the Jacket product family to support double precision linear algebra by itegrating CULA tools to the platform. Jacket 1.3, together with Jacket DLA, offers double precision linear algebra functionality for inv, svd, eig, lu, chol, mpower, norm, and mldivide.

"Our ability to integrate CULA tools into the Jacket product family not only allows AccelerEyes to broaden the functionality we an deliver to our customers but demonstrates the extensibility of Jacket to incorporate a broad set of functions and libraries or many types of applications," said John Melonakos, CEO of AccelerEyes.

Jacket 1.3 also provides new and expanded functional capabilities for conv2, sort, sortrows, eps, nnz, sind, cosd, tand, nextpow2, nthroot,and interp2. Support for Jacket's "gfor" function has also been expanded or added in 1.3 for kron, cross, permute, repmat, circshift, cat, and filter.

In addition, Jacket 1.3 makes it easier for users to install, setup, and configure Jacket on both Windows and Linux platforms. The CUDA toolkit is now prepackaged with Jacket. Jacket automatically ensures prerequisite software is installed and provides automated license activation for Windows systems This release supports MATLAB 2010A as well as previous versions of MATLAB back to R2006B. Together, these improvements enable researchers to get right to work on science, engineering, and analytics versus spending time in system configuration steps.

Finally, AccelerEyes is also making Jacket much more accessible for MATLAB users interested in tapping into GPUs. With the 1.3 release, AccelerEyes has reduced the price of the Jacket base license by nearly 50 percent. Commercial and industrial users can now license Jacket for under $1,000, government and research organization for $800, and academic institutions can license the base Jacket software for $350.

Jacket automatically connects M-code to GPUs through its compile-on-the-fly system, users are able to accelerate their applications across CUDA-capable GPUs in realtime. Jacket eliminates the need to re-program MATLAB in C, C++, and CUDA to run on GPU-based workstations and clusters -- a productivity step that can take months to years to complete for computationally intensive problems.

For example, at the New York University Stern Center for Research Computing (SCRC), a center devoted to providing world-class computational facilities and services to researchers at the Stern School of Business at NYU is taking advantage of Jacket 1.3 performance and functionality improvements to accelerate applications of interest to the resarch community in finance, according to Norman White, Faculty Director at SCRC. "Initial results are very promising in Jacket 1.3, with an 8-10 times performance increase for code that does very large multiple regressions," he said. "Jacket allows our researchers to tap into our centers' computing resources using the MATLAB language, which they are already familiar with, minimizing the amount of time it takes to determine the value that GPUs can deliver for any given problem."

Pricing and availability

Jacket 1.3 is available immediately at $999, $800, and $350 for commercial, government/research, and academic customers, respectively.

About AccelerEyes

AccelerEyes launched in 2007 to commercialize Jacket, the first software platform to deliver productivity in GPU computing. With advanced language processing and runtime technology to transform CPU applications to high performance GPU codes, Jacket extends from desktop workstation performance to also fully leverage GPU clusters. Based in Atlanta, Ga., the privately-held company markets Jacket for a range of defense, intelligence, biomedical, financial, research, and academic applications. Additional information is available at www.accelereyes.com.

-----

Source: AccelerEyes, LLC

Share Options

Subscribe

Subscribe to HPCwire

Discussion

There are 0 discussion items posted.

Join the Discussion

Become a Registered User Today!

Registered Users Log in join the Discussion

RSS Feeds

Feeds by Topic

Feeds by Industry

Feeds by Content Type

Subscribe to All Content

All

May 23, 2012

May 22, 2012

May 21, 2012

May 18, 2012

May 17, 2012

May 16, 2012

May 15, 2012

May 14, 2012

Feature Articles

NVIDIA Works On CPU Co-Dependency Issues with Kepler GPU

NVIDIA is telling everyone that the GK110, its new Kepler GPU aimed at supercomputing, is all about improving performance per watt. But the other driving theme behind the new architecture is reducing the GPU's reliance on its CPU host. How well it accomplishes both these goals areas could determine the success of the new chip in high performance computing.
Read more...

OpenACC Starts to Gather Developer Mindshare

PGI, Cray, and CAPS enterprise are moving quickly to get their new OpenACC-supported compilers into the hands of GPGPU developers. At NVIDIA's GPU Technology Conference this week, there was plenty of discussion around the new HPC accelerator framework, and all three OpenACC compiler makers, as well as NVIDIA, were talking up the technology.
Read more...

NVIDIA Launches Kepler Into HPC

NVIDIA has introduced its first Kepler-generation GPU product for high performance computing, and revealed some of the inner working of the new architecture. The announcement took place at the kickoff of the company's GPU Technology Conference taking place this week in San Jose, California.
Read more...