Advertisement
In this video from SC12, David Cassell from NAG describes the company’s programming environment for the new Intel Xeon Phi.
Performance sensitive routines are updated with each library release to work efficiently on parallel computing systems using standards such as OpenMP and MPI,” the group said. “NAG has tuned and tested parallel routines for running on the Intel Xeon Phi coprocessor, which offers an order of magnitude more threads than traditional multicore processors.”
Read the Full Story.
Mathew Colgrove from The Portland Group writes that an invitation to present the PGI Accelerator programming model at the Society of Exploration Geophysics turned into a Geocode porting project that delivered a 5X speedup over their original MPI/OpenMP version in just five hours.
This article focuses on the process by which the PGI Accelerator directives were applied to SEISMIC_CPML. It should give you an understanding of the basic steps involved, some optimization tips, identify several potential pitfalls, and most of all, demonstrate how you can apply the PGI Accelerator directives to your own code. Note that you can download the full source code used for each step and work along with this article if you like.
Read the Full Story.
In this video from SC12, Scott Lasica from Rogue Wave Software describes the company’s Totalview debugging tools for the new Intel Xeon Phi.
Rogue Wave’s rich set of development tools and libraries, which will now also include support for the Intel Xeon Phi coprocessor, bring valuable capabilities to software developers,” said Bill Savage, vice president and general manager of the Intel Developer Products Division. “Rogue Wave is a great demonstration of how Intel Xeon Phi coprocessors can help tool vendors port more quickly and add more features.”
Read the Full Story.
In this video, Mark O’Connor from Allinea demonstrates the company’s new MAP performance profiler tool at SC12 in Salt Lake City.
Allinea MAP eschews the classic instrumentation-based MPI timeline in favor of a dynamic sampling engine that claims to scale tens of thousands of processes whilst adding just 5% to the total runtime.”
Read the Full Story.
This week the The Portland Group announced plans to extend its PGI Accelerator compiler technology with OpenACC to Intel Xeon Phi coprocessors. Using their existing code bases and with only minor modifications to their build scripts, software developers using PGI Accelerator compilers will soon be able to target Intel Xeon Phi coprocessors as well.
Until now, industry CIOs and government and university lab directors have encountered difficult decisions when evaluating accelerator technologies,” said Douglas Miles, director, The Portland Group. “Do we wait to see which technologies win out? Which programming model or models do we choose? How do we retrain developers to use these programming models? Will our applications be future-proof and portable across the different types of host CPUs and coprocessors? PGI Accelerator compilers will make Xeon Phi coprocessors programmable using standard OpenACC directives that are fully compatible with the accelerator application development efforts already under way at most significant HPC centers and sites.”
First announced in mid-2009, PGI Accelerator compilers provide a high-level coprocessor programming model intended for scientists, engineers and other domain experts who aren’t full-time programmers. PGI Accelerator Fortran 2003, C and C++ compilers enable programmers to offload compute-intensive portions of an application to a coprocessor by adding portable compiler directives, treated as comments by other compilers, to existing standard-compliant programs and recompiling with appropriate compiler options. In 2012, the PGI Accelerator compilers were enhanced to support the new de facto standard OpenACC directives.
Read the Full Story.
Today Rogue Wave Software announced that their upcoming TotalView debugger release will include expanded support for the newest platforms that are advancing both the HPC and commercial markets. With the release of TotalView 8.11, debugging support will be offered for the IBM Blue Gene/Q platform, NVIDIA CUDA 4.2, and the OpenACC capability of Cray CCE 8.0. TotalView 8.11 will also offer early access support for the Intel Xeon Phi coprocessor.
TotalView is our tool of choice for debugging on the Blue Gene/Q platform, especially following our extremely positive experiences with it on the predecessor JuGene machine, which was the world’s largest Blue Gene/P, having approximately 300,000 cores. We need to support applications that use hybrid MPI and OpenMP, and TotalView gives us the ability to debug our very-scalable and most-complex, distributed, multi-threaded applications,” stated Bernd Mohr, Deputy Department Head of Application Support at Jülich Supercomputing Centre.
Read the Full Story or check out Rogue Wave at SC12 booth #3418.
Scaling applications at scale takes planning, work, and the right tools. At iVEC and CSIRO in Australia, users are busy preparing for their next-generation Cray Cascade supercomputer by deploying Allinea DDT in their existing environment.
The purchase of this advanced supercomputer and suite of development tools represents a great investment in the future of Australian research,” said Applications Specialist Rebecca Hartman-Baker. “Having access to such power will allow scientists to explore areas of research that were previously unavailable.”
Read the Full Story.
This week Allinea Software announced the release of its high-performance scalable parallel debugger, Allinea DDT, for IBM BlueGene/Q systems.
Debugging at scale is a daunting problem for systems like Mira, the IBM BlueGene/Q systems at Argonne National Laboratory. Capable of carrying out 10 quadrillion calculations per second on over 750,000 cores, Mira presented a real coding challenge for Allinea, who worked closely with ANL over the last two years to adapt DDT’s scalable architecture to the BlueGene/Q architecture.
Allinea DDT is the only debugger capable of debugging at this scale,” said David Lecomber, CTO of Allinea Software. “We are extremely proud to be part of this project and to be assisting such important scientific work.”
Read the Full Story.
Today The Portland Group announced that customers can now purchase PGI Fortran C, C++ compilers and development tools directly from SGI. Under the new agreement, SGI will resell PGI Fortran C and C++ compilers and development tools with SGI’s line of servers based on Intel processors and NVIDIA GPU accelerators.
The process of procuring software packages separate from an HPC hardware system can often be difficult and time-consuming,” said Douglas Miles, director of The Portland Group. “Now, customers who purchase x64 or x64+GPU processor-based systems from SGI can easily include a widely-used and integrated set of high-performance compilers and development tools that enable porting, debugging and tuning of MPI and OpenMP parallel science and engineering applications.”
The PGI compiler suite includes support for a wide array of programming languages and their PGI Accelerator Fortran 2003 and C99 compilers also include support for the OpenACC directive-based x64+GPU programming model, CUDA Fortran extensions for NVIDIA GPUs, and CUDA C/C++ extensions for 64-bit and 32-bit x86 CPUs. Read the Full Story.
This week Allinea announced that the U.K.’s Atomic Weapons Establishment (AWE) is using the company’s DDT debugging tool to develop scientific code for their powerful Blackthorn and Willow Bull supercomputers.
As the size of jobs we need to do have grown, so have our debugging needs. It was really important to move to a debugger that is extremely fast and stable at any scale”, said Ken Atkinson, AWE’s High Performance Computing Strategy and Procurement Manager. “Allinea DDT was the obvious choice and we have been extremely happy with the product since it was adopted – its user interface makes debugging effective and straightforward for our application developers, even when we are working with large-scale hybrid CPU/GPU applications.”
Read the Full Story.
This week Rogue Wave Software announced its product portfolio supporting Intel’s pending Xeon Phi coprocessor. Designed for highly-parallel workloads , the Intel Xeon Phi will scale to over 50 Intel Architecture cores, making it particularly crucial for developers to have a full spectrum of development tools and components to help parallelize applications.
Rogue Wave’s rich set of development tools and libraries, which will now also include support for the Intel Xeon Phi coprocessor, bring valuable capabilities to software developers,” said Bill Savage, vice president and general manager of the Intel Developer Products Division. “Rogue Wave is a great demonstration of how Intel Xeon Phi coprocessors can help tool vendors port more quickly and add more features.”
Rogue Wave’s strategic product lines will offer Intel Xeon Phi coprocessor support including SourcePro C++, IMSL Numerical Libraries, TotalView debugger, and the ThreadSpotter cache memory optimizer. Read the Full Story.
Allinea Software reports that the company is opening up new possibilities for application developers and scientists at the Swedish National Infrastructure for Computing (SNIC). Allinea software development tools will be deployed across SNIC’s six major facilities for computational science in a wide variety of fields.
With so many users and systems, it was important to have a single tool that would support our environments well – and would enable our users to work at scale. Allinea DDT was clearly the right tool.” said Jacko Koster, Director of SNIC. “This is a step towards creating a core production environment that will be common on all SNIC resources.”
Read the Full Story.
Rogue Wave Software has announced that their latest ThreadSpotter 2012.1 release with support for Cray XE supercomputers.
Scientists and engineers expressed to Rogue Wave their need to employ ThreadSpotter to optimize the performance of parallel applications on the Cray XE platform. As Cray Supercomputers are strategic platforms for our customers, we are happy to quickly respond to these requests so our products stay at the cutting-edge of the HPC market,” stated Chris Gottbrath, Product Manager at Rogue Wave. “Rogue Wave’s ThreadSpotter 2012.1 will help Cray XE users be more productive and produce more efficient programs.”
According to the company, ThreadSpotter helps eliminate performance issues by identifying problematic sites in the code where a change could make the program far more efficient. Read the Full Story.
In this video, Rafael Asenjo from the University of Malaga presents: Main Benefits of Task Parallel Frameworks.
In this talk we will address several task parallel frameworks, like Intel Threading Building Blocks (TBB), Open MP 3.0, and Cilk, to name a few. We will also show some experimental results in the context of pipeline and wave-front parallel algorithms for which we have obtained significant performance improvements by rewriting the pthread version of the codes so as to rely on TBB. As an extra benefit, TBB also provides high level templates in order to improve the code development productivity.
Recorded at the HPC Advisory Council Workshop 2012 in Malaga. Download the slides (PDF).
This past Spring, I was fortunate enough to tour the new CSCS Swiss national supercomputing center, a state-of-the-art facility cooled with lake water. Now CSCS is refreshing their development tools by moving to Allinea DDT.
CSCS hosts Switzerland’s premier HPC systems including the Monte Rosa and Todi Cray XE6 and XK6 machines. With over 50,000 CPU cores and around 200 powerful NVIDIA Tesla X2090 GPU cards on site, developers at CSCS are aiming to achieve significant scientific breakthroughs with their applications.
Allinea DDT meets exactly the debugging needs of our users and allows us to take an important step forward in parallel debugging,” says Thomas Schoenemeyer, Associate Director Technology Integration of CSCS. “Our developers appreciate the unique debugging features such as the excellent parallel stack viewer and other integrated features that allow them to create distinct groups, which are very useful when debugging thousands of processes. The novel sparklines are exceptionally useful for examining data in parallel.”
Read the Full Story.
Advertisement
insideHPC.com is a production of insideHPC, LLC. © 2006-2013 Sitemap
